You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@karaf.apache.org by Miroslav Beranič <mi...@mibesis.si> on 2018/09/07 12:22:22 UTC
Apache Karaf 4.2.2 ( 4.2.x ) Quartz support, with JDBC Persistence
and C3P0 JDBC pool
Hi guys,
I am working and talking about Apache Karaf version 4.2.2-SNAPSHOT (
current master on Github ).
I was rally happy to see Scheduler service with Quartz as default
implementation. Without much thinking, I've hooked up Quartz with JDBC ( I
would really like Karaf Scheduler ( Quartz ) to use existing Datasource and
JPA but no luck ).
When I moved forward, I did not care about JPA that much any more, as even
JDBC was not working. To begin, this is what I've changed in my Karaf
configuration:
file: etc/org.apache.karaf.scheduler.quartz.cfg
org.quartz.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.oracle.OracleDelegate
org.quartz.jobStore.useProperties=false
org.quartz.jobStore.dataSource=MyQuartzDS
org.quartz.jobStore.isClustered=true
org.quartz.jobStore.tablePrefix=q_ # Oracle has 30 char limit on
identifier, I started to cry, when I saw error message in SQL terminal. I
use Oracle 12c1
org.quartz.jobStore.clusterCheckinInterval=30000
org.quartz.dataSource.MyQuartzDS.driver=oracle.jdbc.driver.OracleDriver
org.quartz.dataSource.MyQuartzDS.URL=jdbc:oracle:thin:@
//oracle.local:1521/test
org.quartz.dataSource.MyQuartzDS.user=APP
org.quartz.dataSource.MyQuartzDS.password=app
org.quartz.dataSource.MyQuartzDS.maxConnections=5
org.quartz.dataSource.MyQuartzDS.validationQuery=select * from dual;
org.quartz.scheduler.jmx.export=true
org.quartz.scheduler.rmi.export=false
org.quartz.scheduler.rmi.proxy=false
This seems to be legit. As I saw in
org.apache.karaf.scheduler.core.Activator#doStart method, all / only
properties with key prefix "org.quartz" are forwarded to Quartz. Looks OK,
but I guess "org.quartz." ( with a dot at the end ) would be more correct.
But, minor issue.
Lets start it up. Well, no - it will not. To not write all here, all error
messages, all missing error messages, but strange behavior, let me go to,
what I think, is a solution.
So steps to make your Karaf Scheduler ( 4.2.x ) work with Quartz ( 2.2.x
) , I did this:
1.) Karaf 4.2.2 is using Quartz 2.2.1, that is OK, it could be upgraded to
2.3.0, but I do not need it ( as I am porting application from Quartz
1.8.5, this is way of my list of worries )
2.) scheduler/pom.xml has package imports defined as:
<Import-Package>
!com.mchange.*,
!oracle.*,
!org.quartz.*,
!weblogic.*,
!javax.transaction,
javax.servlet*;resolution:=optional,
org.jboss.*;resolution:=optional,
*
</Import-Package>
( I really am not OSGi expert, and all, but this looked strange to me ),
turned out be "true" ( I guess, as I am open to option, I am wrong, but end
of the day - this works for me, so for me -t his is correct )
I changed this to:
<Import-Package>
com.mchange.*;resolution:=optional,
oracle.*;resolution:=optional,
org.quartz.*,
!weblogic.*,
javax.transaction,
javax.servlet*;resolution:=optional,
org.jboss.*;resolution:=optional,
*
</Import-Package>
( removed all the ! in front of the packages and added
resolution:=optional, as not all users want persistent Quartz jobs , I
guess )
This is not all, now, when I deploy to Karaf, all sort of issue is
reported. Turned out I need to install org.apache.servicemix.bundles.quartz
bundle, from this url:
https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
( I did not install it from the URL, Karaf shell is doing this ... )
but look, here is
<!--
https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
-->
<dependency>
<groupId>org.apache.servicemix.bundles</groupId>
<artifactId>org.apache.servicemix.bundles.quartz</artifactId>
<version>2.2.1_1</version>
</dependency>
Nope, it will not work. As, same problem with Apache Karaf Scheduler's POM
file. Wrong imports. I guess someone found out this is wrong, and created
this version ( not direct next version, but few versions later on ):
<!--
https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
-->
<dependency>
<groupId>org.apache.servicemix.bundles</groupId>
<artifactId>org.apache.servicemix.bundles.quartz</artifactId>
<version>2.3.0_2</version>
</dependency>
But wait ... sure, this has different Quartz version. How nice , why not
... If we can upgrade, we should do it. But at this point, I was really
crying ... for muliple reasons, ... I was to much focused into my path and
versions I've decided to clone repository and change Quartz version and
name it 2.2.5.
So:
<dependency>
<groupId>org.apache.servicemix.bundles</groupId>
<artifactId>org.apache.servicemix.bundles.quartz</artifactId>
<version>2.3.0_2</version>
</dependency>
with changed Quartz version to 2.2.1 is "my"
<dependency>
<groupId>org.apache.servicemix.bundles</groupId>
<artifactId>org.apache.servicemix.bundles.quartz</artifactId>
<version>2.2.5</version>
</dependency>
( was this good or not, again: it works for me now, and I ran out of hair
to pull them out ... so, keep calm and move forward )
But this is not end of problems. I forgot what exactly was wrong at this
point.
I've installed feature:
scr transaction jndi jpa hibernate pac-jdbc-oracle ( my own oracle jdbc
bundle ) aries-blueprint scheduler karaf-scheduler-example
and bundles:
bundle:install
mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.c3p0/0.9.5.2_3-SNAPSHOT
Looook, I am starting to see something. Yeey baby:
2018-09-07T09:43:58,678 | INFO | activator-1-thread-1 |
JobStoreTX | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | Using db table-based data access locking (synchronization).
2018-09-07T09:43:58,680 | INFO | activator-1-thread-1 |
JobStoreTX | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | JobStoreTX initialized.
2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
QuartzScheduler | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | Scheduler meta-data: Quartz Scheduler (v2.2.1)
'MyScheduler' with instanceId 'my'
Scheduler class: 'org.quartz.core.QuartzScheduler' - running locally.
NOT STARTED.
Currently in standby mode.
Number of jobs executed: 0
Using thread pool 'org.quartz.simpl.SimpleThreadPool' - with 20 threads.
Using job-store 'org.quartz.impl.jdbcjobstore.JobStoreTX' - which
supports persistence. and is clustered.
2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
StdSchedulerFactory | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | Quartz scheduler 'MyScheduler' initialized from an
externally provided properties instance.
2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
StdSchedulerFactory | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | Quartz scheduler version: 2.2.1
2018-09-07T09:43:58,713 | INFO | activator-1-thread-1 |
AbstractPoolBackedDataSource | 107 - org.apache.servicemix.bundles.c3p0
- 0.9.5.2_3-SNAPSHOT | Initializing c3p0 pool...
com.mchange.v2.c3p0.ComboPooledDataSource [ acquireIncrement -> 3,
acquireRetryAttempts -> 30, acquireRetryDelay -> 1000, autoCommitOnClose ->
false, automaticTestTable -> null, breakAfterAcquireFailure -> false,
checkoutTimeout -> 0, connectionCustomizerClassName -> null,
connectionTesterClassName ->
com.mchange.v2.c3p0.impl.DefaultConnectionTester, contextClassLoaderSource
-> caller, dataSourceName -> z8kfsx9x1dnli4w1yck8tp|3c42672a,
debugUnreturnedConnectionStackTraces -> false, description -> null,
driverClass -> oracle.jdbc.driver.OracleDriver, extensions -> {},
factoryClassLocation -> null, forceIgnoreUnresolvedTransactions -> false,
forceSynchronousCheckins -> false, forceUseNamedDriverClass -> false,
identityToken -> z8kfsx9x1dnli4w1yck8tp|3c42672a, idleConnectionTestPeriod
-> 50, initialPoolSize -> 3, jdbcUrl ->
jdbc:oracle:thin:@//oracle.local:1521/test,
maxAdministrativeTaskTime -> 0, maxConnectionAge -> 0, maxIdleTime -> 0,
maxIdleTimeExcessConnections -> 0, maxPoolSize -> 5, maxStatements -> 0,
maxStatementsPerConnection -> 120, minPoolSize -> 1, numHelperThreads -> 3,
preferredTestQuery -> select * from dual;, privilegeSpawnedThreads ->
false, properties -> {user=******, password=******}, propertyCycle -> 0,
statementCacheNumDeferredCloseThreads -> 0, testConnectionOnCheckin ->
true, testConnectionOnCheckout -> false, unreturnedConnectionTimeout -> 0,
userOverrides -> {}, usesTraditionalReflectiveProxies -> false ]
2018-09-07T09:43:58,820 | INFO | activator-1-thread-1 |
JobStoreTX | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | ClusterManager: detected 1 failed or restarted instances.
2018-09-07T09:43:58,821 | INFO | activator-1-thread-1 |
JobStoreTX | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | ClusterManager: Scanning for instance "my"'s failed
in-progress jobs.
2018-09-07T09:43:58,826 | INFO | activator-1-thread-1 |
QuartzScheduler | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | Scheduler MyScheduler$_my started.
2018-09-07T09:43:58,837 | INFO | activator-1-thread-1 |
CommandExtension | 35 - org.apache.karaf.shell.core -
4.2.2.SNAPSHOT | Registering commands for bundle
org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT
This is it! I did it. Time for big coffee. And than I came back at the
terminal and saw this:
2018-09-07T09:43:59,021 | WARN | activator-1-thread-1 |
WhiteboardHandler | 105 - org.apache.karaf.scheduler.core -
4.2.2.SNAPSHOT | Error scheduling job
org.apache.karaf.scheduler.SchedulerError:
org.quartz.JobPersistenceException: Couldn't store job: Unable to serialize
JobDataMap for insertion into database because the value of property
'QuartzJobScheduler.Logger' is not serializable:
org.ops4j.pax.logging.slf4j.Slf4jLogger [See nested exception:
java.io.NotSerializableException: Unable to serialize JobDataMap for
insertion into database because the value of property
'QuartzJobScheduler.Logger' is not serializable:
org.ops4j.pax.logging.slf4j.Slf4jLogger]
at
org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:239)
~[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
at
org.apache.karaf.scheduler.core.WhiteboardHandler.register(WhiteboardHandler.java:140)
[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
at
org.apache.karaf.scheduler.core.WhiteboardHandler.access$100(WhiteboardHandler.java:37)
[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
at
org.apache.karaf.scheduler.core.WhiteboardHandler$1.addingService(WhiteboardHandler.java:66)
[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
at
org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:941)
[?:?]
at
org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:870)
[?:?]
at
org.osgi.util.tracker.AbstractTracked.trackAdding(AbstractTracked.java:256)
[?:?]
at
org.osgi.util.tracker.AbstractTracked.trackInitial(AbstractTracked.java:183)
[?:?]
at org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:318)
[?:?]
at org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:261)
[?:?]
at
org.apache.karaf.scheduler.core.WhiteboardHandler.<init>(WhiteboardHandler.java:71)
[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
at org.apache.karaf.scheduler.core.Activator.doStart(Activator.java:60)
[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
at
org.apache.karaf.util.tracker.BaseActivator.run(BaseActivator.java:275)
[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
at
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
[?:?]
at java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:?]
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
[?:?]
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
[?:?]
at java.lang.Thread.run(Thread.java:748) [?:?]
Caused by: org.quartz.JobPersistenceException: Couldn't store job: Unable
to serialize JobDataMap for insertion into database because the value of
property 'QuartzJobScheduler.Logger' is not serializable:
org.ops4j.pax.logging.slf4j.Slf4jLogger
at
org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1115)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
~[?:?]
at
org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886) ~[?:?]
at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
~[?:?]
at
org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
~[?:?]
... 17 more
Caused by: java.io.NotSerializableException: Unable to serialize JobDataMap
for insertion into database because the value of property
'QuartzJobScheduler.Logger' is not serializable:
org.ops4j.pax.logging.slf4j.Slf4jLogger
at
org.quartz.impl.jdbcjobstore.StdJDBCDelegate.serializeJobData(StdJDBCDelegate.java:3083)
~[?:?]
at
org.quartz.impl.jdbcjobstore.oracle.OracleDelegate.insertJobDetail(OracleDelegate.java:154)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1112)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
~[?:?]
at
org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
~[?:?]
at
org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886) ~[?:?]
at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
~[?:?]
at
org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
~[?:?]
... 17 more
WAAAT!! No no no no. Well, this was just the beginning and I did not know
this at that time. So I started to look for 'QuartzJobScheduler.Logger' in
Karaf soruce code. Waaat, who thought it is smart to store non-serializable
data in Job data, whaaat.
At that point I started to realize, why it was so much pain to get
persistence working ... ( I guess ) no one tried or even wanted to make
Quartz persistent in Karaf - since this code is in from 2014.
Scheduling is like one of the five most important features that I need from
the platform. I started to cry again. I did not test features, before I
pushed for the migration. There is no way back, so what can be done now
here.
Let's see.
I basically "re-wrote" Scheduler service in Karaf. What I did:
1.) Make sure Job data has only Serializable data -
- ScheduleOptions was made Serializable, with addition to move Trigger
creation from InternalScheduleOptions's constructor to method compile().
- getJobs() method from Scheduler interface returns String names and not
Objects
- Updated QuartzJobExecutor to remove non-serializable items from
JobDataMap storage
2.)
- Added custom StdScheduler and StdSchedulerFactory classes, to implemented
changes
- Added SchedulerStorage as support for "backward compatibility" and "short
lived" jobs. This is not good place to use it.
3.)
- Updated Scheduler Shell commands to use changed Scheduler API ( job names
are as keys )
Patch file is in this email attachements, I've also pushed changes to my
Github repository, with only this changes:
https://github.com/mibesis/apache-karaf/tree/karaf-4.2.2-quartz-jdbc
Anyone that is willing to use it -- please do, I would only ask Karf core
development team to "make sure" Quartz Job data is always fill with
Serializable data, as this makes all the difference and there is no need
for method getJobs() from Sheculer to return
Map<Object, ScheduleOptions> as it can not make data Serializable -- to
return Map<Object, ScheduleOptions> is OK, as all API calls are made over
keys, names, etc.
I need few days, to come over this and coffee cups, it is Friday now, so I
guess I will grab some cold drink ... it was big learing experiance for me,
to make this work.
So this is my first patch file, if there is something else needed or I can
support in someway, let me know. ( I know SchedulerStorage is not perfect
solution, but again: it works for me ; but I agree, this can be made better
)
I know I do not say it much ( ever ) -- Apache Karaf is really great
platform, it is just that I send email, when I have problems :D ...
Kind Regards,
Miroslav
--
Miroslav Beranič
MIBESIS
+386(0)40/814-843
miroslav.beranic@mibesis.si
http://www.mibesis.si
Re: Apache Karaf 4.2.2 ( 4.2.x ) Quartz support, with JDBC
Persistence and C3P0 JDBC pool
Posted by Miroslav Beranič <mi...@mibesis.si>.
JB,
looks like I broke the build.
[ERROR] Failed to execute goal
org.apache.karaf.tooling:karaf-maven-plugin:4.2.2-SNAPSHOT:verify (verify)
on project standard: Verification failures: Verification failures:
[ERROR] Feature resolution failed for [scheduler/4.2.2.SNAPSHOT]
[ERROR] Message: Unable to resolve root: missing requirement [root]
osgi.identity; osgi.identity=scheduler; type=karaf.feature;
version=4.2.2.SNAPSHOT;
filter:="(&(osgi.identity=scheduler)(type=karaf.feature)(version>=4.2.2.SNAPSHOT))"
[caused by: Unable to resolve scheduler/4.2.2.SNAPSHOT: missing requirement
[scheduler/4.2.2.SNAPSHOT] osgi.identity;
osgi.identity=org.apache.karaf.scheduler.core; type=osgi.bundle;
version="[4.2.2.SNAPSHOT,4.2.2.SNAPSHOT]"; resolution:=mandatory [caused
by: Unable to resolve org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT:
missing requirement [org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT]
osgi.wiring.package;
filter:="(&(osgi.wiring.package=org.quartz.management)(version>=2.2.0)(!(version>=3.0.0)))"]]
[ERROR] Repositories: {
[ERROR]
file:/home/miroslav/Workspace/Apache/Karaf/karaf-4.2.x/assemblies/features/standard/target/feature/feature.xml
[ERROR]
mvn:org.apache.karaf.features/framework/4.2.2-SNAPSHOT/xml/features
[ERROR] mvn:org.ops4j.pax.web/pax-web-features/7.2.3/xml/features
[ERROR] }
[ERROR] Resources: {
[ERROR] mvn:org.apache.felix/org.apache.felix.configadmin/1.9.4
[ERROR] mvn:org.apache.felix/org.apache.felix.coordinator/1.0.2
[ERROR] mvn:org.apache.felix/org.apache.felix.fileinstall/3.6.4
[ERROR]
mvn:org.apache.karaf.features/org.apache.karaf.features.core/4.2.2-SNAPSHOT
[ERROR]
mvn:org.apache.karaf.features/org.apache.karaf.features.extension/4.2.2-SNAPSHOT
[ERROR]
mvn:org.apache.karaf.scheduler/org.apache.karaf.scheduler.core/4.2.2-SNAPSHOT
[ERROR]
mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.jaxb-impl/2.2.11_1
[ERROR]
mvn:org.apache.servicemix.specs/org.apache.servicemix.specs.activation-api-1.1/2.5.0
[ERROR]
mvn:org.apache.servicemix.specs/org.apache.servicemix.specs.jaxb-api-2.2/2.5.0
[ERROR] mvn:org.fusesource.jansi/jansi/1.17.1
[ERROR] mvn:org.ops4j.pax.logging/pax-logging-api/1.10.1
[ERROR] mvn:org.ops4j.pax.logging/pax-logging-log4j2/1.10.1
[ERROR] mvn:org.ops4j.pax.url/pax-url-aether/2.5.4
[ERROR] }
I will have to take a look at this more, I do not have time now, will
continue later today or tomorrow. I guess I was to quick ... my bad.
Kind Regards,
Miroslav
V V pet., 7. sep. 2018 ob 15:26 je oseba Jean-Baptiste Onofré <
jb@nanthrax.net> napisala:
> You created the PR on your base.
>
> It should be created based on https://github.com/apache/karaf.
>
> Regards
> JB
>
> On 07/09/2018 15:20, Miroslav Beranič wrote:
> > Hi JB,
> >
> > ( I guess I did ok ),
> >
> > I've created PR at:
> > https://github.com/mibesis/apache-karaf/pull/1
> >
> > Kind Regards,
> > Miroslav
> >
> >
> >
> > V V pet., 7. sep. 2018 ob 14:56 je oseba Jean-Baptiste Onofré
> > <jb@nanthrax.net <ma...@nanthrax.net>> napisala:
> >
> > Oh sorry, PR means Pull Request on Github.
> >
> > You can find some details on the contribution guide:
> >
> > http://karaf.apache.org/community.html
> >
> > Don't hesitate to ping me (maybe you can join the Karaf Slack
> channel).
> > I would love to discuss with you about that.
> >
> > Regards
> > JB
> >
> > On 07/09/2018 14:53, Miroslav Beranič wrote:
> > > Hi JB,
> > >
> > > Scheduler with use on Cellar is my semi-next task, scheduled for
> > > beginning of October.
> > >
> > > PR? Sorry, I am not into this, so PR stands for?
> > >
> > > Kind Regards,
> > > Miroslav
> > >
> > >
> > >
> > >
> > > V V pet., 7. sep. 2018 ob 14:48 je oseba Jean-Baptiste Onofré
> > > <jb@nanthrax.net <ma...@nanthrax.net> <mailto:jb@nanthrax.net
> > <ma...@nanthrax.net>>> napisala:
> > >
> > > Hi,
> > >
> > > thanks for the your detailed update on the scheduler.
> > >
> > > I did the change in Karaf 4.2.x to be able to support Quartz
> > scheduler
> > > in cluster mode (especially when used with Cellar).
> > >
> > > I worked fine when I tested, using dynamic import, so I'm a bit
> > > surprised it doesn't work out of the box for you.
> > >
> > > Let me take a look on your patch file to identify the missing
> > parts. In
> > > the mean time, do you mind to prepare a PR ?
> > >
> > > Thanks !
> > > Regards
> > > JB
> > >
> > > On 07/09/2018 14:22, Miroslav Beranič wrote:
> > > > Hi guys,
> > > >
> > > > I am working and talking about Apache Karaf version
> > 4.2.2-SNAPSHOT (
> > > > current master on Github ).
> > > >
> > > > I was rally happy to see Scheduler service with Quartz as
> > default
> > > > implementation. Without much thinking, I've hooked up Quartz
> > with
> > > JDBC (
> > > > I would really like Karaf Scheduler ( Quartz ) to use
> existing
> > > > Datasource and JPA but no luck ).
> > > >
> > > > When I moved forward, I did not care about JPA that much any
> > more, as
> > > > even JDBC was not working. To begin, this is what I've
> > changed in my
> > > > Karaf configuration:
> > > >
> > > > file: etc/org.apache.karaf.scheduler.quartz.cfg
> > > >
> > org.quartz.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
> > > >
> > >
> >
> org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.oracle.OracleDelegate
> > > > org.quartz.jobStore.useProperties=false
> > > > org.quartz.jobStore.dataSource=MyQuartzDS
> > > > org.quartz.jobStore.isClustered=true
> > > > org.quartz.jobStore.tablePrefix=q_ # Oracle has 30 char
> limit on
> > > > identifier, I started to cry, when I saw error message in SQL
> > > terminal.
> > > > I use Oracle 12c1
> > > > org.quartz.jobStore.clusterCheckinInterval=30000
> > > >
> > > >
> > >
> >
> org.quartz.dataSource.MyQuartzDS.driver=oracle.jdbc.driver.OracleDriver
> > > >
> > >
> > org.quartz.dataSource.MyQuartzDS.URL=jdbc:oracle:thin:@
> //oracle.local:1521/test
> > > > org.quartz.dataSource.MyQuartzDS.user=APP
> > > > org.quartz.dataSource.MyQuartzDS.password=app
> > > > org.quartz.dataSource.MyQuartzDS.maxConnections=5
> > > > org.quartz.dataSource.MyQuartzDS.validationQuery=select *
> > from dual;
> > > >
> > > > org.quartz.scheduler.jmx.export=true
> > > > org.quartz.scheduler.rmi.export=false
> > > > org.quartz.scheduler.rmi.proxy=false
> > > >
> > > > This seems to be legit. As I saw in
> > > > org.apache.karaf.scheduler.core.Activator#doStart method,
> > all / only
> > > > properties with key prefix "org.quartz" are forwarded to
> > Quartz. Looks
> > > > OK, but I guess "org.quartz." ( with a dot at the end )
> > would be more
> > > > correct. But, minor issue.
> > > >
> > > > Lets start it up. Well, no - it will not. To not write all
> > here, all
> > > > error messages, all missing error messages, but strange
> > behavior,
> > > let me
> > > > go to, what I think, is a solution.
> > > >
> > > > So steps to make your Karaf Scheduler ( 4.2.x ) work with
> > Quartz (
> > > > 2.2.x ) , I did this:
> > > >
> > > > 1.) Karaf 4.2.2 is using Quartz 2.2.1, that is OK, it could
> be
> > > upgraded
> > > > to 2.3.0, but I do not need it ( as I am porting application
> > from
> > > Quartz
> > > > 1.8.5, this is way of my list of worries )
> > > > 2.) scheduler/pom.xml has package imports defined as:
> > > > <Import-Package>
> > > > !com.mchange.*,
> > > > !oracle.*,
> > > > !org.quartz.*,
> > > > !weblogic.*,
> > > > !javax.transaction,
> > > > javax.servlet*;resolution:=optional,
> > > > org.jboss.*;resolution:=optional,
> > > > *
> > > > </Import-Package>
> > > >
> > > > ( I really am not OSGi expert, and all, but this looked
> > strange to
> > > me ),
> > > > turned out be "true" ( I guess, as I am open to option, I am
> > > wrong, but
> > > > end of the day - this works for me, so for me -t his is
> > correct )
> > > >
> > > > I changed this to:
> > > > <Import-Package>
> > > > com.mchange.*;resolution:=optional,
> > > > oracle.*;resolution:=optional,
> > > > org.quartz.*,
> > > > !weblogic.*,
> > > > javax.transaction,
> > > > javax.servlet*;resolution:=optional,
> > > > org.jboss.*;resolution:=optional,
> > > > *
> > > > </Import-Package>
> > > >
> > > > ( removed all the ! in front of the packages and added
> > > > resolution:=optional, as not all users want persistent
> > Quartz jobs , I
> > > > guess )
> > > >
> > > > This is not all, now, when I deploy to Karaf, all sort of
> > issue is
> > > > reported. Turned out I need to install
> > > > org.apache.servicemix.bundles.quartz bundle, from this url:
> > > >
> > > >
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > > >
> > > > ( I did not install it from the URL, Karaf shell is doing
> > this ... )
> > > >
> > > > but look, here is
> > > > <!--
> > > >
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > > > -->
> > > > <dependency>
> > > > <groupId>org.apache.servicemix.bundles</groupId>
> > > >
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > > <version>2.2.1_1</version>
> > > > </dependency>
> > > >
> > > > Nope, it will not work. As, same problem with Apache Karaf
> > Scheduler's
> > > > POM file. Wrong imports. I guess someone found out this is
> > wrong, and
> > > > created this version ( not direct next version, but few
> versions
> > > later on ):
> > > >
> > > > <!--
> > > >
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > > > -->
> > > > <dependency>
> > > > <groupId>org.apache.servicemix.bundles</groupId>
> > > >
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > > <version>2.3.0_2</version>
> > > > </dependency>
> > > >
> > > > But wait ... sure, this has different Quartz version. How
> nice ,
> > > why not
> > > > ... If we can upgrade, we should do it. But at this point, I
> was
> > > really
> > > > crying ... for muliple reasons, ... I was to much focused
> > into my path
> > > > and versions I've decided to clone repository and change
> Quartz
> > > version
> > > > and name it 2.2.5.
> > > > So:
> > > > <dependency>
> > > > <groupId>org.apache.servicemix.bundles</groupId>
> > > >
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > > <version>2.3.0_2</version>
> > > > </dependency>
> > > > with changed Quartz version to 2.2.1 is "my"
> > > > <dependency>
> > > > <groupId>org.apache.servicemix.bundles</groupId>
> > > >
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > > <version>2.2.5</version>
> > > > </dependency>
> > > >
> > > > ( was this good or not, again: it works for me now, and I
> > ran out of
> > > > hair to pull them out ... so, keep calm and move forward )
> > > >
> > > > But this is not end of problems. I forgot what exactly was
> wrong
> > > at this
> > > > point.
> > > >
> > > > I've installed feature:
> > > >
> > > > scr transaction jndi jpa hibernate pac-jdbc-oracle ( my own
> > oracle
> > > jdbc
> > > > bundle ) aries-blueprint scheduler karaf-scheduler-example
> > > >
> > > > and bundles:
> > > > bundle:install
> > > >
> > >
> >
> mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.c3p0/0.9.5.2_3-SNAPSHOT
> > > >
> > > > Looook, I am starting to see something. Yeey baby:
> > > >
> > > > 2018-09-07T09:43:58,678 | INFO | activator-1-thread-1 |
> > > > JobStoreTX | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | Using db table-based data access locking
> > > > (synchronization).
> > > > 2018-09-07T09:43:58,680 | INFO | activator-1-thread-1 |
> > > > JobStoreTX | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | JobStoreTX initialized.
> > > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > > QuartzScheduler | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | Scheduler meta-data: Quartz Scheduler
> > (v2.2.1)
> > > > 'MyScheduler' with instanceId 'my'
> > > > Scheduler class: 'org.quartz.core.QuartzScheduler' -
> running
> > > locally.
> > > > NOT STARTED.
> > > > Currently in standby mode.
> > > > Number of jobs executed: 0
> > > > Using thread pool 'org.quartz.simpl.SimpleThreadPool' -
> > with 20
> > > threads.
> > > > Using job-store 'org.quartz.impl.jdbcjobstore.JobStoreTX'
> > - which
> > > > supports persistence. and is clustered.
> > > >
> > > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > > StdSchedulerFactory | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | Quartz scheduler 'MyScheduler'
> > initialized from an
> > > > externally provided properties instance.
> > > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > > StdSchedulerFactory | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | Quartz scheduler version: 2.2.1
> > > > 2018-09-07T09:43:58,713 | INFO | activator-1-thread-1 |
> > > > AbstractPoolBackedDataSource | 107 -
> > > > org.apache.servicemix.bundles.c3p0 - 0.9.5.2_3-SNAPSHOT |
> > Initializing
> > > > c3p0 pool... com.mchange.v2.c3p0.ComboPooledDataSource [
> > > > acquireIncrement -> 3, acquireRetryAttempts -> 30,
> > > acquireRetryDelay ->
> > > > 1000, autoCommitOnClose -> false, automaticTestTable -> null,
> > > > breakAfterAcquireFailure -> false, checkoutTimeout -> 0,
> > > > connectionCustomizerClassName -> null,
> > connectionTesterClassName ->
> > > > com.mchange.v2.c3p0.impl.DefaultConnectionTester,
> > > > contextClassLoaderSource -> caller, dataSourceName ->
> > > > z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > > debugUnreturnedConnectionStackTraces ->
> > > > false, description -> null, driverClass ->
> > > > oracle.jdbc.driver.OracleDriver, extensions -> {},
> > > factoryClassLocation
> > > > -> null, forceIgnoreUnresolvedTransactions -> false,
> > > > forceSynchronousCheckins -> false, forceUseNamedDriverClass
> > -> false,
> > > > identityToken -> z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > > > idleConnectionTestPeriod -> 50, initialPoolSize -> 3,
> jdbcUrl ->
> > > > jdbc:oracle:thin:@//oracle.local:1521/test,
> > > maxAdministrativeTaskTime ->
> > > > 0, maxConnectionAge -> 0, maxIdleTime -> 0,
> > > maxIdleTimeExcessConnections
> > > > -> 0, maxPoolSize -> 5, maxStatements -> 0,
> > maxStatementsPerConnection
> > > > -> 120, minPoolSize -> 1, numHelperThreads -> 3,
> > preferredTestQuery ->
> > > > select * from dual;, privilegeSpawnedThreads -> false,
> > properties ->
> > > > {user=******, password=******}, propertyCycle -> 0,
> > > > statementCacheNumDeferredCloseThreads -> 0,
> > testConnectionOnCheckin ->
> > > > true, testConnectionOnCheckout -> false,
> > > unreturnedConnectionTimeout ->
> > > > 0, userOverrides -> {}, usesTraditionalReflectiveProxies ->
> > false ]
> > > > 2018-09-07T09:43:58,820 | INFO | activator-1-thread-1 |
> > > > JobStoreTX | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | ClusterManager: detected 1 failed or
> > restarted
> > > instances.
> > > > 2018-09-07T09:43:58,821 | INFO | activator-1-thread-1 |
> > > > JobStoreTX | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | ClusterManager: Scanning for instance
> > "my"'s failed
> > > > in-progress jobs.
> > > > 2018-09-07T09:43:58,826 | INFO | activator-1-thread-1 |
> > > > QuartzScheduler | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | Scheduler MyScheduler$_my started.
> > > > 2018-09-07T09:43:58,837 | INFO | activator-1-thread-1 |
> > > > CommandExtension | 35 -
> > org.apache.karaf.shell.core -
> > > > 4.2.2.SNAPSHOT | Registering commands for bundle
> > > > org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT
> > > >
> > > > This is it! I did it. Time for big coffee. And than I came
> > back at the
> > > > terminal and saw this:
> > > >
> > > > 2018-09-07T09:43:59,021 | WARN | activator-1-thread-1 |
> > > > WhiteboardHandler | 105 -
> > > org.apache.karaf.scheduler.core
> > > > - 4.2.2.SNAPSHOT | Error scheduling job
> > > > org.apache.karaf.scheduler.SchedulerError:
> > > > org.quartz.JobPersistenceException: Couldn't store job:
> > Unable to
> > > > serialize JobDataMap for insertion into database because the
> > value of
> > > > property 'QuartzJobScheduler.Logger' is not serializable:
> > > > org.ops4j.pax.logging.slf4j.Slf4jLogger [See nested
> exception:
> > > > java.io <http://java.io>
> > <http://java.io>.NotSerializableException: Unable to
> > > serialize JobDataMap for
> > > > insertion into database because the value of property
> > > > 'QuartzJobScheduler.Logger' is not serializable:
> > > > org.ops4j.pax.logging.slf4j.Slf4jLogger]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:239)
> > > > ~[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.register(WhiteboardHandler.java:140)
> > > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.access$100(WhiteboardHandler.java:37)
> > > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler$1.addingService(WhiteboardHandler.java:66)
> > > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > > at
> > > >
> > >
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:941)
> > > > [?:?]
> > > > at
> > > >
> > >
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:870)
> > > > [?:?]
> > > > at
> > > >
> > >
> >
> org.osgi.util.tracker.AbstractTracked.trackAdding(AbstractTracked.java:256)
> > > > [?:?]
> > > > at
> > > >
> > >
> >
> org.osgi.util.tracker.AbstractTracked.trackInitial(AbstractTracked.java:183)
> > > > [?:?]
> > > > at
> > > >
> > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:318)
> > > [?:?]
> > > > at
> > > >
> > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:261)
> > > [?:?]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.<init>(WhiteboardHandler.java:71)
> > > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > > at
> > > >
> > org.apache.karaf.scheduler.core.Activator.doStart(Activator.java:60)
> > > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.util.tracker.BaseActivator.run(BaseActivator.java:275)
> > > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > > at
> > > >
> > >
> >
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> > > > [?:?]
> > > > at
> > java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:?]
> > > > at
> > > >
> > >
> >
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> > > > [?:?]
> > > > at
> > > >
> > >
> >
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> > > > [?:?]
> > > > at java.lang.Thread.run(Thread.java:748) [?:?]
> > > > Caused by: org.quartz.JobPersistenceException: Couldn't
> > store job:
> > > > Unable to serialize JobDataMap for insertion into database
> > because the
> > > > value of property 'QuartzJobScheduler.Logger' is not
> > serializable:
> > > > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1115)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> > ~[?:?]
> > > > at
> > org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > > > ~[?:?]
> > > > ... 17 more
> > > > Caused by: java.io <http://java.io>
> > <http://java.io>.NotSerializableException:
> > > Unable to serialize
> > > > JobDataMap for insertion into database because the value of
> > property
> > > > 'QuartzJobScheduler.Logger' is not serializable:
> > > > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.StdJDBCDelegate.serializeJobData(StdJDBCDelegate.java:3083)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.oracle.OracleDelegate.insertJobDetail(OracleDelegate.java:154)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1112)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> > ~[?:?]
> > > > at
> > org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > > > ~[?:?]
> > > > at
> > > >
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > > > ~[?:?]
> > > > ... 17 more
> > > >
> > > >
> > > > WAAAT!! No no no no. Well, this was just the beginning and
> > I did not
> > > > know this at that time. So I started to look for
> > > > 'QuartzJobScheduler.Logger' in Karaf soruce code. Waaat, who
> > > thought it
> > > > is smart to store non-serializable data in Job data, whaaat.
> > > > At that point I started to realize, why it was so much pain
> > to get
> > > > persistence working ... ( I guess ) no one tried or even
> > wanted to
> > > make
> > > > Quartz persistent in Karaf - since this code is in from 2014.
> > > >
> > > > Scheduling is like one of the five most important features
> > that I need
> > > > from the platform. I started to cry again. I did not test
> > features,
> > > > before I pushed for the migration. There is no way back, so
> what
> > > can be
> > > > done now here.
> > > >
> > > > Let's see.
> > > >
> > > > I basically "re-wrote" Scheduler service in Karaf. What I
> did:
> > > >
> > > > 1.) Make sure Job data has only Serializable data -
> > > > - ScheduleOptions was made Serializable, with addition to
> > move Trigger
> > > > creation from InternalScheduleOptions's constructor to method
> > > compile().
> > > > - getJobs() method from Scheduler interface returns String
> names
> > > and not
> > > > Objects
> > > > - Updated QuartzJobExecutor to remove non-serializable items
> > from
> > > > JobDataMap storage
> > > >
> > > > 2.)
> > > > - Added custom StdScheduler and StdSchedulerFactory classes,
> to
> > > > implemented changes
> > > > - Added SchedulerStorage as support for "backward
> > compatibility" and
> > > > "short lived" jobs. This is not good place to use it.
> > > >
> > > > 3.)
> > > > - Updated Scheduler Shell commands to use changed Scheduler
> > API ( job
> > > > names are as keys )
> > > >
> > > > Patch file is in this email attachements, I've also pushed
> > changes
> > > to my
> > > > Github repository, with only this changes:
> > > >
> > > >
> > https://github.com/mibesis/apache-karaf/tree/karaf-4.2.2-quartz-jdbc
> > > >
> > > >
> > > > Anyone that is willing to use it -- please do, I would only
> > ask Karf
> > > > core development team to "make sure" Quartz Job data is
> always
> > > fill with
> > > > Serializable data, as this makes all the difference and
> there is
> > > no need
> > > > for method getJobs() from Sheculer to return
> > > > Map<Object, ScheduleOptions> as it can not make data
> > Serializable
> > > -- to
> > > > return Map<Object, ScheduleOptions> is OK, as all API calls
> > are made
> > > > over keys, names, etc.
> > > >
> > > > I need few days, to come over this and coffee cups, it is
> Friday
> > > now, so
> > > > I guess I will grab some cold drink ... it was big learing
> > experiance
> > > > for me, to make this work.
> > > >
> > > > So this is my first patch file, if there is something else
> > needed or I
> > > > can support in someway, let me know. ( I know
> > SchedulerStorage is not
> > > > perfect solution, but again: it works for me ; but I agree,
> this
> > > can be
> > > > made better )
> > > >
> > > >
> > > > I know I do not say it much ( ever ) -- Apache Karaf is
> > really great
> > > > platform, it is just that I send email, when I have problems
> > :D ...
> > > >
> > > > Kind Regards,
> > > > Miroslav
> > > >
> > > >
> > > > --
> > > > Miroslav Beranič
> > > > MIBESIS
> > > > +386(0)40/814-843
> > > > miroslav.beranic@mibesis.si
> > <ma...@mibesis.si>
> > <mailto:miroslav.beranic@mibesis.si
> > <ma...@mibesis.si>>
> > > <mailto:miroslav.beranic@mibesis.si
> > <ma...@mibesis.si>
> > > <mailto:miroslav.beranic@mibesis.si
> > <ma...@mibesis.si>>>
> > > > http://www.mibesis.si
> > >
> > > --
> > > Jean-Baptiste Onofré
> > > jbonofre@apache.org <ma...@apache.org>
> > <mailto:jbonofre@apache.org <ma...@apache.org>>
> > > http://blog.nanthrax.net
> > > Talend - http://www.talend.com
> > >
> > >
> > >
> > > --
> > > Miroslav Beranič
> > > MIBESIS
> > > +386(0)40/814-843
> > > miroslav.beranic@mibesis.si <ma...@mibesis.si>
> > <mailto:miroslav.beranic@mibesis.si
> > <ma...@mibesis.si>>
> > > http://www.mibesis.si
> >
> > --
> > Jean-Baptiste Onofré
> > jbonofre@apache.org <ma...@apache.org>
> > http://blog.nanthrax.net
> > Talend - http://www.talend.com
> >
> >
> >
> > --
> > Miroslav Beranič
> > MIBESIS
> > +386(0)40/814-843
> > miroslav.beranic@mibesis.si <ma...@mibesis.si>
> > http://www.mibesis.si
>
> --
> Jean-Baptiste Onofré
> jbonofre@apache.org
> http://blog.nanthrax.net
> Talend - http://www.talend.com
>
--
Miroslav Beranič
MIBESIS
+386(0)40/814-843
miroslav.beranic@mibesis.si
http://www.mibesis.si
Re: Apache Karaf 4.2.2 ( 4.2.x ) Quartz support, with JDBC
Persistence and C3P0 JDBC pool
Posted by Jean-Baptiste Onofré <jb...@nanthrax.net>.
You created the PR on your base.
It should be created based on https://github.com/apache/karaf.
Regards
JB
On 07/09/2018 15:20, Miroslav Beranič wrote:
> Hi JB,
>
> ( I guess I did ok ),
>
> I've created PR at:
> https://github.com/mibesis/apache-karaf/pull/1
>
> Kind Regards,
> Miroslav
>
>
>
> V V pet., 7. sep. 2018 ob 14:56 je oseba Jean-Baptiste Onofré
> <jb@nanthrax.net <ma...@nanthrax.net>> napisala:
>
> Oh sorry, PR means Pull Request on Github.
>
> You can find some details on the contribution guide:
>
> http://karaf.apache.org/community.html
>
> Don't hesitate to ping me (maybe you can join the Karaf Slack channel).
> I would love to discuss with you about that.
>
> Regards
> JB
>
> On 07/09/2018 14:53, Miroslav Beranič wrote:
> > Hi JB,
> >
> > Scheduler with use on Cellar is my semi-next task, scheduled for
> > beginning of October.
> >
> > PR? Sorry, I am not into this, so PR stands for?
> >
> > Kind Regards,
> > Miroslav
> >
> >
> >
> >
> > V V pet., 7. sep. 2018 ob 14:48 je oseba Jean-Baptiste Onofré
> > <jb@nanthrax.net <ma...@nanthrax.net> <mailto:jb@nanthrax.net
> <ma...@nanthrax.net>>> napisala:
> >
> > Hi,
> >
> > thanks for the your detailed update on the scheduler.
> >
> > I did the change in Karaf 4.2.x to be able to support Quartz
> scheduler
> > in cluster mode (especially when used with Cellar).
> >
> > I worked fine when I tested, using dynamic import, so I'm a bit
> > surprised it doesn't work out of the box for you.
> >
> > Let me take a look on your patch file to identify the missing
> parts. In
> > the mean time, do you mind to prepare a PR ?
> >
> > Thanks !
> > Regards
> > JB
> >
> > On 07/09/2018 14:22, Miroslav Beranič wrote:
> > > Hi guys,
> > >
> > > I am working and talking about Apache Karaf version
> 4.2.2-SNAPSHOT (
> > > current master on Github ).
> > >
> > > I was rally happy to see Scheduler service with Quartz as
> default
> > > implementation. Without much thinking, I've hooked up Quartz
> with
> > JDBC (
> > > I would really like Karaf Scheduler ( Quartz ) to use existing
> > > Datasource and JPA but no luck ).
> > >
> > > When I moved forward, I did not care about JPA that much any
> more, as
> > > even JDBC was not working. To begin, this is what I've
> changed in my
> > > Karaf configuration:
> > >
> > > file: etc/org.apache.karaf.scheduler.quartz.cfg
> > >
> org.quartz.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
> > >
> >
> org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.oracle.OracleDelegate
> > > org.quartz.jobStore.useProperties=false
> > > org.quartz.jobStore.dataSource=MyQuartzDS
> > > org.quartz.jobStore.isClustered=true
> > > org.quartz.jobStore.tablePrefix=q_ # Oracle has 30 char limit on
> > > identifier, I started to cry, when I saw error message in SQL
> > terminal.
> > > I use Oracle 12c1
> > > org.quartz.jobStore.clusterCheckinInterval=30000
> > >
> > >
> >
> org.quartz.dataSource.MyQuartzDS.driver=oracle.jdbc.driver.OracleDriver
> > >
> >
> org.quartz.dataSource.MyQuartzDS.URL=jdbc:oracle:thin:@//oracle.local:1521/test
> > > org.quartz.dataSource.MyQuartzDS.user=APP
> > > org.quartz.dataSource.MyQuartzDS.password=app
> > > org.quartz.dataSource.MyQuartzDS.maxConnections=5
> > > org.quartz.dataSource.MyQuartzDS.validationQuery=select *
> from dual;
> > >
> > > org.quartz.scheduler.jmx.export=true
> > > org.quartz.scheduler.rmi.export=false
> > > org.quartz.scheduler.rmi.proxy=false
> > >
> > > This seems to be legit. As I saw in
> > > org.apache.karaf.scheduler.core.Activator#doStart method,
> all / only
> > > properties with key prefix "org.quartz" are forwarded to
> Quartz. Looks
> > > OK, but I guess "org.quartz." ( with a dot at the end )
> would be more
> > > correct. But, minor issue.
> > >
> > > Lets start it up. Well, no - it will not. To not write all
> here, all
> > > error messages, all missing error messages, but strange
> behavior,
> > let me
> > > go to, what I think, is a solution.
> > >
> > > So steps to make your Karaf Scheduler ( 4.2.x ) work with
> Quartz (
> > > 2.2.x ) , I did this:
> > >
> > > 1.) Karaf 4.2.2 is using Quartz 2.2.1, that is OK, it could be
> > upgraded
> > > to 2.3.0, but I do not need it ( as I am porting application
> from
> > Quartz
> > > 1.8.5, this is way of my list of worries )
> > > 2.) scheduler/pom.xml has package imports defined as:
> > > <Import-Package>
> > > !com.mchange.*,
> > > !oracle.*,
> > > !org.quartz.*,
> > > !weblogic.*,
> > > !javax.transaction,
> > > javax.servlet*;resolution:=optional,
> > > org.jboss.*;resolution:=optional,
> > > *
> > > </Import-Package>
> > >
> > > ( I really am not OSGi expert, and all, but this looked
> strange to
> > me ),
> > > turned out be "true" ( I guess, as I am open to option, I am
> > wrong, but
> > > end of the day - this works for me, so for me -t his is
> correct )
> > >
> > > I changed this to:
> > > <Import-Package>
> > > com.mchange.*;resolution:=optional,
> > > oracle.*;resolution:=optional,
> > > org.quartz.*,
> > > !weblogic.*,
> > > javax.transaction,
> > > javax.servlet*;resolution:=optional,
> > > org.jboss.*;resolution:=optional,
> > > *
> > > </Import-Package>
> > >
> > > ( removed all the ! in front of the packages and added
> > > resolution:=optional, as not all users want persistent
> Quartz jobs , I
> > > guess )
> > >
> > > This is not all, now, when I deploy to Karaf, all sort of
> issue is
> > > reported. Turned out I need to install
> > > org.apache.servicemix.bundles.quartz bundle, from this url:
> > >
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > >
> > > ( I did not install it from the URL, Karaf shell is doing
> this ... )
> > >
> > > but look, here is
> > > <!--
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > > -->
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > >
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.2.1_1</version>
> > > </dependency>
> > >
> > > Nope, it will not work. As, same problem with Apache Karaf
> Scheduler's
> > > POM file. Wrong imports. I guess someone found out this is
> wrong, and
> > > created this version ( not direct next version, but few versions
> > later on ):
> > >
> > > <!--
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > > -->
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > >
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.3.0_2</version>
> > > </dependency>
> > >
> > > But wait ... sure, this has different Quartz version. How nice ,
> > why not
> > > ... If we can upgrade, we should do it. But at this point, I was
> > really
> > > crying ... for muliple reasons, ... I was to much focused
> into my path
> > > and versions I've decided to clone repository and change Quartz
> > version
> > > and name it 2.2.5.
> > > So:
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > >
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.3.0_2</version>
> > > </dependency>
> > > with changed Quartz version to 2.2.1 is "my"
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > >
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.2.5</version>
> > > </dependency>
> > >
> > > ( was this good or not, again: it works for me now, and I
> ran out of
> > > hair to pull them out ... so, keep calm and move forward )
> > >
> > > But this is not end of problems. I forgot what exactly was wrong
> > at this
> > > point.
> > >
> > > I've installed feature:
> > >
> > > scr transaction jndi jpa hibernate pac-jdbc-oracle ( my own
> oracle
> > jdbc
> > > bundle ) aries-blueprint scheduler karaf-scheduler-example
> > >
> > > and bundles:
> > > bundle:install
> > >
> >
> mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.c3p0/0.9.5.2_3-SNAPSHOT
> > >
> > > Looook, I am starting to see something. Yeey baby:
> > >
> > > 2018-09-07T09:43:58,678 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Using db table-based data access locking
> > > (synchronization).
> > > 2018-09-07T09:43:58,680 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | JobStoreTX initialized.
> > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > QuartzScheduler | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Scheduler meta-data: Quartz Scheduler
> (v2.2.1)
> > > 'MyScheduler' with instanceId 'my'
> > > Scheduler class: 'org.quartz.core.QuartzScheduler' - running
> > locally.
> > > NOT STARTED.
> > > Currently in standby mode.
> > > Number of jobs executed: 0
> > > Using thread pool 'org.quartz.simpl.SimpleThreadPool' -
> with 20
> > threads.
> > > Using job-store 'org.quartz.impl.jdbcjobstore.JobStoreTX'
> - which
> > > supports persistence. and is clustered.
> > >
> > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > StdSchedulerFactory | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Quartz scheduler 'MyScheduler'
> initialized from an
> > > externally provided properties instance.
> > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > StdSchedulerFactory | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Quartz scheduler version: 2.2.1
> > > 2018-09-07T09:43:58,713 | INFO | activator-1-thread-1 |
> > > AbstractPoolBackedDataSource | 107 -
> > > org.apache.servicemix.bundles.c3p0 - 0.9.5.2_3-SNAPSHOT |
> Initializing
> > > c3p0 pool... com.mchange.v2.c3p0.ComboPooledDataSource [
> > > acquireIncrement -> 3, acquireRetryAttempts -> 30,
> > acquireRetryDelay ->
> > > 1000, autoCommitOnClose -> false, automaticTestTable -> null,
> > > breakAfterAcquireFailure -> false, checkoutTimeout -> 0,
> > > connectionCustomizerClassName -> null,
> connectionTesterClassName ->
> > > com.mchange.v2.c3p0.impl.DefaultConnectionTester,
> > > contextClassLoaderSource -> caller, dataSourceName ->
> > > z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > debugUnreturnedConnectionStackTraces ->
> > > false, description -> null, driverClass ->
> > > oracle.jdbc.driver.OracleDriver, extensions -> {},
> > factoryClassLocation
> > > -> null, forceIgnoreUnresolvedTransactions -> false,
> > > forceSynchronousCheckins -> false, forceUseNamedDriverClass
> -> false,
> > > identityToken -> z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > > idleConnectionTestPeriod -> 50, initialPoolSize -> 3, jdbcUrl ->
> > > jdbc:oracle:thin:@//oracle.local:1521/test,
> > maxAdministrativeTaskTime ->
> > > 0, maxConnectionAge -> 0, maxIdleTime -> 0,
> > maxIdleTimeExcessConnections
> > > -> 0, maxPoolSize -> 5, maxStatements -> 0,
> maxStatementsPerConnection
> > > -> 120, minPoolSize -> 1, numHelperThreads -> 3,
> preferredTestQuery ->
> > > select * from dual;, privilegeSpawnedThreads -> false,
> properties ->
> > > {user=******, password=******}, propertyCycle -> 0,
> > > statementCacheNumDeferredCloseThreads -> 0,
> testConnectionOnCheckin ->
> > > true, testConnectionOnCheckout -> false,
> > unreturnedConnectionTimeout ->
> > > 0, userOverrides -> {}, usesTraditionalReflectiveProxies ->
> false ]
> > > 2018-09-07T09:43:58,820 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | ClusterManager: detected 1 failed or
> restarted
> > instances.
> > > 2018-09-07T09:43:58,821 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | ClusterManager: Scanning for instance
> "my"'s failed
> > > in-progress jobs.
> > > 2018-09-07T09:43:58,826 | INFO | activator-1-thread-1 |
> > > QuartzScheduler | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Scheduler MyScheduler$_my started.
> > > 2018-09-07T09:43:58,837 | INFO | activator-1-thread-1 |
> > > CommandExtension | 35 -
> org.apache.karaf.shell.core -
> > > 4.2.2.SNAPSHOT | Registering commands for bundle
> > > org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT
> > >
> > > This is it! I did it. Time for big coffee. And than I came
> back at the
> > > terminal and saw this:
> > >
> > > 2018-09-07T09:43:59,021 | WARN | activator-1-thread-1 |
> > > WhiteboardHandler | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Error scheduling job
> > > org.apache.karaf.scheduler.SchedulerError:
> > > org.quartz.JobPersistenceException: Couldn't store job:
> Unable to
> > > serialize JobDataMap for insertion into database because the
> value of
> > > property 'QuartzJobScheduler.Logger' is not serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger [See nested exception:
> > > java.io <http://java.io>
> <http://java.io>.NotSerializableException: Unable to
> > serialize JobDataMap for
> > > insertion into database because the value of property
> > > 'QuartzJobScheduler.Logger' is not serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:239)
> > > ~[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.register(WhiteboardHandler.java:140)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.access$100(WhiteboardHandler.java:37)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler$1.addingService(WhiteboardHandler.java:66)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:941)
> > > [?:?]
> > > at
> > >
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:870)
> > > [?:?]
> > > at
> > >
> >
> org.osgi.util.tracker.AbstractTracked.trackAdding(AbstractTracked.java:256)
> > > [?:?]
> > > at
> > >
> >
> org.osgi.util.tracker.AbstractTracked.trackInitial(AbstractTracked.java:183)
> > > [?:?]
> > > at
> > >
> org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:318)
> > [?:?]
> > > at
> > >
> org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:261)
> > [?:?]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.<init>(WhiteboardHandler.java:71)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> org.apache.karaf.scheduler.core.Activator.doStart(Activator.java:60)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.util.tracker.BaseActivator.run(BaseActivator.java:275)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> > > [?:?]
> > > at
> java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:?]
> > > at
> > >
> >
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> > > [?:?]
> > > at
> > >
> >
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> > > [?:?]
> > > at java.lang.Thread.run(Thread.java:748) [?:?]
> > > Caused by: org.quartz.JobPersistenceException: Couldn't
> store job:
> > > Unable to serialize JobDataMap for insertion into database
> because the
> > > value of property 'QuartzJobScheduler.Logger' is not
> serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1115)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> ~[?:?]
> > > at
> org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > > ~[?:?]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > > ~[?:?]
> > > ... 17 more
> > > Caused by: java.io <http://java.io>
> <http://java.io>.NotSerializableException:
> > Unable to serialize
> > > JobDataMap for insertion into database because the value of
> property
> > > 'QuartzJobScheduler.Logger' is not serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.StdJDBCDelegate.serializeJobData(StdJDBCDelegate.java:3083)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.oracle.OracleDelegate.insertJobDetail(OracleDelegate.java:154)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1112)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> ~[?:?]
> > > at
> org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > > ~[?:?]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > > ~[?:?]
> > > ... 17 more
> > >
> > >
> > > WAAAT!! No no no no. Well, this was just the beginning and
> I did not
> > > know this at that time. So I started to look for
> > > 'QuartzJobScheduler.Logger' in Karaf soruce code. Waaat, who
> > thought it
> > > is smart to store non-serializable data in Job data, whaaat.
> > > At that point I started to realize, why it was so much pain
> to get
> > > persistence working ... ( I guess ) no one tried or even
> wanted to
> > make
> > > Quartz persistent in Karaf - since this code is in from 2014.
> > >
> > > Scheduling is like one of the five most important features
> that I need
> > > from the platform. I started to cry again. I did not test
> features,
> > > before I pushed for the migration. There is no way back, so what
> > can be
> > > done now here.
> > >
> > > Let's see.
> > >
> > > I basically "re-wrote" Scheduler service in Karaf. What I did:
> > >
> > > 1.) Make sure Job data has only Serializable data -
> > > - ScheduleOptions was made Serializable, with addition to
> move Trigger
> > > creation from InternalScheduleOptions's constructor to method
> > compile().
> > > - getJobs() method from Scheduler interface returns String names
> > and not
> > > Objects
> > > - Updated QuartzJobExecutor to remove non-serializable items
> from
> > > JobDataMap storage
> > >
> > > 2.)
> > > - Added custom StdScheduler and StdSchedulerFactory classes, to
> > > implemented changes
> > > - Added SchedulerStorage as support for "backward
> compatibility" and
> > > "short lived" jobs. This is not good place to use it.
> > >
> > > 3.)
> > > - Updated Scheduler Shell commands to use changed Scheduler
> API ( job
> > > names are as keys )
> > >
> > > Patch file is in this email attachements, I've also pushed
> changes
> > to my
> > > Github repository, with only this changes:
> > >
> > >
> https://github.com/mibesis/apache-karaf/tree/karaf-4.2.2-quartz-jdbc
> > >
> > >
> > > Anyone that is willing to use it -- please do, I would only
> ask Karf
> > > core development team to "make sure" Quartz Job data is always
> > fill with
> > > Serializable data, as this makes all the difference and there is
> > no need
> > > for method getJobs() from Sheculer to return
> > > Map<Object, ScheduleOptions> as it can not make data
> Serializable
> > -- to
> > > return Map<Object, ScheduleOptions> is OK, as all API calls
> are made
> > > over keys, names, etc.
> > >
> > > I need few days, to come over this and coffee cups, it is Friday
> > now, so
> > > I guess I will grab some cold drink ... it was big learing
> experiance
> > > for me, to make this work.
> > >
> > > So this is my first patch file, if there is something else
> needed or I
> > > can support in someway, let me know. ( I know
> SchedulerStorage is not
> > > perfect solution, but again: it works for me ; but I agree, this
> > can be
> > > made better )
> > >
> > >
> > > I know I do not say it much ( ever ) -- Apache Karaf is
> really great
> > > platform, it is just that I send email, when I have problems
> :D ...
> > >
> > > Kind Regards,
> > > Miroslav
> > >
> > >
> > > --
> > > Miroslav Beranič
> > > MIBESIS
> > > +386(0)40/814-843
> > > miroslav.beranic@mibesis.si
> <ma...@mibesis.si>
> <mailto:miroslav.beranic@mibesis.si
> <ma...@mibesis.si>>
> > <mailto:miroslav.beranic@mibesis.si
> <ma...@mibesis.si>
> > <mailto:miroslav.beranic@mibesis.si
> <ma...@mibesis.si>>>
> > > http://www.mibesis.si
> >
> > --
> > Jean-Baptiste Onofré
> > jbonofre@apache.org <ma...@apache.org>
> <mailto:jbonofre@apache.org <ma...@apache.org>>
> > http://blog.nanthrax.net
> > Talend - http://www.talend.com
> >
> >
> >
> > --
> > Miroslav Beranič
> > MIBESIS
> > +386(0)40/814-843
> > miroslav.beranic@mibesis.si <ma...@mibesis.si>
> <mailto:miroslav.beranic@mibesis.si
> <ma...@mibesis.si>>
> > http://www.mibesis.si
>
> --
> Jean-Baptiste Onofré
> jbonofre@apache.org <ma...@apache.org>
> http://blog.nanthrax.net
> Talend - http://www.talend.com
>
>
>
> --
> Miroslav Beranič
> MIBESIS
> +386(0)40/814-843
> miroslav.beranic@mibesis.si <ma...@mibesis.si>
> http://www.mibesis.si
--
Jean-Baptiste Onofré
jbonofre@apache.org
http://blog.nanthrax.net
Talend - http://www.talend.com
Re: Apache Karaf 4.2.2 ( 4.2.x ) Quartz support, with JDBC
Persistence and C3P0 JDBC pool
Posted by Miroslav Beranič <mi...@mibesis.si>.
Hi JB,
( I guess I did ok ),
I've created PR at:
https://github.com/mibesis/apache-karaf/pull/1
Kind Regards,
Miroslav
V V pet., 7. sep. 2018 ob 14:56 je oseba Jean-Baptiste Onofré <
jb@nanthrax.net> napisala:
> Oh sorry, PR means Pull Request on Github.
>
> You can find some details on the contribution guide:
>
> http://karaf.apache.org/community.html
>
> Don't hesitate to ping me (maybe you can join the Karaf Slack channel).
> I would love to discuss with you about that.
>
> Regards
> JB
>
> On 07/09/2018 14:53, Miroslav Beranič wrote:
> > Hi JB,
> >
> > Scheduler with use on Cellar is my semi-next task, scheduled for
> > beginning of October.
> >
> > PR? Sorry, I am not into this, so PR stands for?
> >
> > Kind Regards,
> > Miroslav
> >
> >
> >
> >
> > V V pet., 7. sep. 2018 ob 14:48 je oseba Jean-Baptiste Onofré
> > <jb@nanthrax.net <ma...@nanthrax.net>> napisala:
> >
> > Hi,
> >
> > thanks for the your detailed update on the scheduler.
> >
> > I did the change in Karaf 4.2.x to be able to support Quartz
> scheduler
> > in cluster mode (especially when used with Cellar).
> >
> > I worked fine when I tested, using dynamic import, so I'm a bit
> > surprised it doesn't work out of the box for you.
> >
> > Let me take a look on your patch file to identify the missing parts.
> In
> > the mean time, do you mind to prepare a PR ?
> >
> > Thanks !
> > Regards
> > JB
> >
> > On 07/09/2018 14:22, Miroslav Beranič wrote:
> > > Hi guys,
> > >
> > > I am working and talking about Apache Karaf version 4.2.2-SNAPSHOT
> (
> > > current master on Github ).
> > >
> > > I was rally happy to see Scheduler service with Quartz as default
> > > implementation. Without much thinking, I've hooked up Quartz with
> > JDBC (
> > > I would really like Karaf Scheduler ( Quartz ) to use existing
> > > Datasource and JPA but no luck ).
> > >
> > > When I moved forward, I did not care about JPA that much any more,
> as
> > > even JDBC was not working. To begin, this is what I've changed in
> my
> > > Karaf configuration:
> > >
> > > file: etc/org.apache.karaf.scheduler.quartz.cfg
> > > org.quartz.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
> > >
> >
> org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.oracle.OracleDelegate
> > > org.quartz.jobStore.useProperties=false
> > > org.quartz.jobStore.dataSource=MyQuartzDS
> > > org.quartz.jobStore.isClustered=true
> > > org.quartz.jobStore.tablePrefix=q_ # Oracle has 30 char limit on
> > > identifier, I started to cry, when I saw error message in SQL
> > terminal.
> > > I use Oracle 12c1
> > > org.quartz.jobStore.clusterCheckinInterval=30000
> > >
> > >
> >
> org.quartz.dataSource.MyQuartzDS.driver=oracle.jdbc.driver.OracleDriver
> > >
> > org.quartz.dataSource.MyQuartzDS.URL=jdbc:oracle:thin:@
> //oracle.local:1521/test
> > > org.quartz.dataSource.MyQuartzDS.user=APP
> > > org.quartz.dataSource.MyQuartzDS.password=app
> > > org.quartz.dataSource.MyQuartzDS.maxConnections=5
> > > org.quartz.dataSource.MyQuartzDS.validationQuery=select * from
> dual;
> > >
> > > org.quartz.scheduler.jmx.export=true
> > > org.quartz.scheduler.rmi.export=false
> > > org.quartz.scheduler.rmi.proxy=false
> > >
> > > This seems to be legit. As I saw in
> > > org.apache.karaf.scheduler.core.Activator#doStart method, all /
> only
> > > properties with key prefix "org.quartz" are forwarded to Quartz.
> Looks
> > > OK, but I guess "org.quartz." ( with a dot at the end ) would be
> more
> > > correct. But, minor issue.
> > >
> > > Lets start it up. Well, no - it will not. To not write all here,
> all
> > > error messages, all missing error messages, but strange behavior,
> > let me
> > > go to, what I think, is a solution.
> > >
> > > So steps to make your Karaf Scheduler ( 4.2.x ) work with Quartz
> (
> > > 2.2.x ) , I did this:
> > >
> > > 1.) Karaf 4.2.2 is using Quartz 2.2.1, that is OK, it could be
> > upgraded
> > > to 2.3.0, but I do not need it ( as I am porting application from
> > Quartz
> > > 1.8.5, this is way of my list of worries )
> > > 2.) scheduler/pom.xml has package imports defined as:
> > > <Import-Package>
> > > !com.mchange.*,
> > > !oracle.*,
> > > !org.quartz.*,
> > > !weblogic.*,
> > > !javax.transaction,
> > > javax.servlet*;resolution:=optional,
> > > org.jboss.*;resolution:=optional,
> > > *
> > > </Import-Package>
> > >
> > > ( I really am not OSGi expert, and all, but this looked strange to
> > me ),
> > > turned out be "true" ( I guess, as I am open to option, I am
> > wrong, but
> > > end of the day - this works for me, so for me -t his is correct )
> > >
> > > I changed this to:
> > > <Import-Package>
> > > com.mchange.*;resolution:=optional,
> > > oracle.*;resolution:=optional,
> > > org.quartz.*,
> > > !weblogic.*,
> > > javax.transaction,
> > > javax.servlet*;resolution:=optional,
> > > org.jboss.*;resolution:=optional,
> > > *
> > > </Import-Package>
> > >
> > > ( removed all the ! in front of the packages and added
> > > resolution:=optional, as not all users want persistent Quartz jobs
> , I
> > > guess )
> > >
> > > This is not all, now, when I deploy to Karaf, all sort of issue is
> > > reported. Turned out I need to install
> > > org.apache.servicemix.bundles.quartz bundle, from this url:
> > >
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > >
> > > ( I did not install it from the URL, Karaf shell is doing this ...
> )
> > >
> > > but look, here is
> > > <!--
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > > -->
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.2.1_1</version>
> > > </dependency>
> > >
> > > Nope, it will not work. As, same problem with Apache Karaf
> Scheduler's
> > > POM file. Wrong imports. I guess someone found out this is wrong,
> and
> > > created this version ( not direct next version, but few versions
> > later on ):
> > >
> > > <!--
> > >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > > -->
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.3.0_2</version>
> > > </dependency>
> > >
> > > But wait ... sure, this has different Quartz version. How nice ,
> > why not
> > > ... If we can upgrade, we should do it. But at this point, I was
> > really
> > > crying ... for muliple reasons, ... I was to much focused into my
> path
> > > and versions I've decided to clone repository and change Quartz
> > version
> > > and name it 2.2.5.
> > > So:
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.3.0_2</version>
> > > </dependency>
> > > with changed Quartz version to 2.2.1 is "my"
> > > <dependency>
> > > <groupId>org.apache.servicemix.bundles</groupId>
> > > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > > <version>2.2.5</version>
> > > </dependency>
> > >
> > > ( was this good or not, again: it works for me now, and I ran out
> of
> > > hair to pull them out ... so, keep calm and move forward )
> > >
> > > But this is not end of problems. I forgot what exactly was wrong
> > at this
> > > point.
> > >
> > > I've installed feature:
> > >
> > > scr transaction jndi jpa hibernate pac-jdbc-oracle ( my own oracle
> > jdbc
> > > bundle ) aries-blueprint scheduler karaf-scheduler-example
> > >
> > > and bundles:
> > > bundle:install
> > >
> >
> mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.c3p0/0.9.5.2_3-SNAPSHOT
> > >
> > > Looook, I am starting to see something. Yeey baby:
> > >
> > > 2018-09-07T09:43:58,678 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Using db table-based data access locking
> > > (synchronization).
> > > 2018-09-07T09:43:58,680 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | JobStoreTX initialized.
> > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > QuartzScheduler | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Scheduler meta-data: Quartz Scheduler (v2.2.1)
> > > 'MyScheduler' with instanceId 'my'
> > > Scheduler class: 'org.quartz.core.QuartzScheduler' - running
> > locally.
> > > NOT STARTED.
> > > Currently in standby mode.
> > > Number of jobs executed: 0
> > > Using thread pool 'org.quartz.simpl.SimpleThreadPool' - with 20
> > threads.
> > > Using job-store 'org.quartz.impl.jdbcjobstore.JobStoreTX' - which
> > > supports persistence. and is clustered.
> > >
> > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > StdSchedulerFactory | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Quartz scheduler 'MyScheduler' initialized from
> an
> > > externally provided properties instance.
> > > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > > StdSchedulerFactory | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Quartz scheduler version: 2.2.1
> > > 2018-09-07T09:43:58,713 | INFO | activator-1-thread-1 |
> > > AbstractPoolBackedDataSource | 107 -
> > > org.apache.servicemix.bundles.c3p0 - 0.9.5.2_3-SNAPSHOT |
> Initializing
> > > c3p0 pool... com.mchange.v2.c3p0.ComboPooledDataSource [
> > > acquireIncrement -> 3, acquireRetryAttempts -> 30,
> > acquireRetryDelay ->
> > > 1000, autoCommitOnClose -> false, automaticTestTable -> null,
> > > breakAfterAcquireFailure -> false, checkoutTimeout -> 0,
> > > connectionCustomizerClassName -> null, connectionTesterClassName ->
> > > com.mchange.v2.c3p0.impl.DefaultConnectionTester,
> > > contextClassLoaderSource -> caller, dataSourceName ->
> > > z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > debugUnreturnedConnectionStackTraces ->
> > > false, description -> null, driverClass ->
> > > oracle.jdbc.driver.OracleDriver, extensions -> {},
> > factoryClassLocation
> > > -> null, forceIgnoreUnresolvedTransactions -> false,
> > > forceSynchronousCheckins -> false, forceUseNamedDriverClass ->
> false,
> > > identityToken -> z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > > idleConnectionTestPeriod -> 50, initialPoolSize -> 3, jdbcUrl ->
> > > jdbc:oracle:thin:@//oracle.local:1521/test,
> > maxAdministrativeTaskTime ->
> > > 0, maxConnectionAge -> 0, maxIdleTime -> 0,
> > maxIdleTimeExcessConnections
> > > -> 0, maxPoolSize -> 5, maxStatements -> 0,
> maxStatementsPerConnection
> > > -> 120, minPoolSize -> 1, numHelperThreads -> 3,
> preferredTestQuery ->
> > > select * from dual;, privilegeSpawnedThreads -> false, properties
> ->
> > > {user=******, password=******}, propertyCycle -> 0,
> > > statementCacheNumDeferredCloseThreads -> 0,
> testConnectionOnCheckin ->
> > > true, testConnectionOnCheckout -> false,
> > unreturnedConnectionTimeout ->
> > > 0, userOverrides -> {}, usesTraditionalReflectiveProxies -> false ]
> > > 2018-09-07T09:43:58,820 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | ClusterManager: detected 1 failed or restarted
> > instances.
> > > 2018-09-07T09:43:58,821 | INFO | activator-1-thread-1 |
> > > JobStoreTX | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | ClusterManager: Scanning for instance "my"'s
> failed
> > > in-progress jobs.
> > > 2018-09-07T09:43:58,826 | INFO | activator-1-thread-1 |
> > > QuartzScheduler | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Scheduler MyScheduler$_my started.
> > > 2018-09-07T09:43:58,837 | INFO | activator-1-thread-1 |
> > > CommandExtension | 35 -
> org.apache.karaf.shell.core -
> > > 4.2.2.SNAPSHOT | Registering commands for bundle
> > > org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT
> > >
> > > This is it! I did it. Time for big coffee. And than I came back at
> the
> > > terminal and saw this:
> > >
> > > 2018-09-07T09:43:59,021 | WARN | activator-1-thread-1 |
> > > WhiteboardHandler | 105 -
> > org.apache.karaf.scheduler.core
> > > - 4.2.2.SNAPSHOT | Error scheduling job
> > > org.apache.karaf.scheduler.SchedulerError:
> > > org.quartz.JobPersistenceException: Couldn't store job: Unable to
> > > serialize JobDataMap for insertion into database because the value
> of
> > > property 'QuartzJobScheduler.Logger' is not serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger [See nested exception:
> > > java.io <http://java.io>.NotSerializableException: Unable to
> > serialize JobDataMap for
> > > insertion into database because the value of property
> > > 'QuartzJobScheduler.Logger' is not serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:239)
> > > ~[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.register(WhiteboardHandler.java:140)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.access$100(WhiteboardHandler.java:37)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler$1.addingService(WhiteboardHandler.java:66)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:941)
> > > [?:?]
> > > at
> > >
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:870)
> > > [?:?]
> > > at
> > >
> >
> org.osgi.util.tracker.AbstractTracked.trackAdding(AbstractTracked.java:256)
> > > [?:?]
> > > at
> > >
> >
> org.osgi.util.tracker.AbstractTracked.trackInitial(AbstractTracked.java:183)
> > > [?:?]
> > > at
> > > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:318)
> > [?:?]
> > > at
> > > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:261)
> > [?:?]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.<init>(WhiteboardHandler.java:71)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> org.apache.karaf.scheduler.core.Activator.doStart(Activator.java:60)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> org.apache.karaf.util.tracker.BaseActivator.run(BaseActivator.java:275)
> > > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > > at
> > >
> >
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> > > [?:?]
> > > at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> [?:?]
> > > at
> > >
> >
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> > > [?:?]
> > > at
> > >
> >
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> > > [?:?]
> > > at java.lang.Thread.run(Thread.java:748) [?:?]
> > > Caused by: org.quartz.JobPersistenceException: Couldn't store job:
> > > Unable to serialize JobDataMap for insertion into database because
> the
> > > value of property 'QuartzJobScheduler.Logger' is not serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1115)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> ~[?:?]
> > > at
> org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > > ~[?:?]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > > ~[?:?]
> > > ... 17 more
> > > Caused by: java.io <http://java.io>.NotSerializableException:
> > Unable to serialize
> > > JobDataMap for insertion into database because the value of
> property
> > > 'QuartzJobScheduler.Logger' is not serializable:
> > > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.StdJDBCDelegate.serializeJobData(StdJDBCDelegate.java:3083)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.oracle.OracleDelegate.insertJobDetail(OracleDelegate.java:154)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1112)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> > ~[?:?]
> > > at
> > >
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > > ~[?:?]
> > > at
> > >
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> ~[?:?]
> > > at
> org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > > ~[?:?]
> > > at
> > >
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > > ~[?:?]
> > > ... 17 more
> > >
> > >
> > > WAAAT!! No no no no. Well, this was just the beginning and I did
> not
> > > know this at that time. So I started to look for
> > > 'QuartzJobScheduler.Logger' in Karaf soruce code. Waaat, who
> > thought it
> > > is smart to store non-serializable data in Job data, whaaat.
> > > At that point I started to realize, why it was so much pain to get
> > > persistence working ... ( I guess ) no one tried or even wanted to
> > make
> > > Quartz persistent in Karaf - since this code is in from 2014.
> > >
> > > Scheduling is like one of the five most important features that I
> need
> > > from the platform. I started to cry again. I did not test features,
> > > before I pushed for the migration. There is no way back, so what
> > can be
> > > done now here.
> > >
> > > Let's see.
> > >
> > > I basically "re-wrote" Scheduler service in Karaf. What I did:
> > >
> > > 1.) Make sure Job data has only Serializable data -
> > > - ScheduleOptions was made Serializable, with addition to move
> Trigger
> > > creation from InternalScheduleOptions's constructor to method
> > compile().
> > > - getJobs() method from Scheduler interface returns String names
> > and not
> > > Objects
> > > - Updated QuartzJobExecutor to remove non-serializable items from
> > > JobDataMap storage
> > >
> > > 2.)
> > > - Added custom StdScheduler and StdSchedulerFactory classes, to
> > > implemented changes
> > > - Added SchedulerStorage as support for "backward compatibility"
> and
> > > "short lived" jobs. This is not good place to use it.
> > >
> > > 3.)
> > > - Updated Scheduler Shell commands to use changed Scheduler API (
> job
> > > names are as keys )
> > >
> > > Patch file is in this email attachements, I've also pushed changes
> > to my
> > > Github repository, with only this changes:
> > >
> > >
> https://github.com/mibesis/apache-karaf/tree/karaf-4.2.2-quartz-jdbc
> > >
> > >
> > > Anyone that is willing to use it -- please do, I would only ask
> Karf
> > > core development team to "make sure" Quartz Job data is always
> > fill with
> > > Serializable data, as this makes all the difference and there is
> > no need
> > > for method getJobs() from Sheculer to return
> > > Map<Object, ScheduleOptions> as it can not make data Serializable
> > -- to
> > > return Map<Object, ScheduleOptions> is OK, as all API calls are
> made
> > > over keys, names, etc.
> > >
> > > I need few days, to come over this and coffee cups, it is Friday
> > now, so
> > > I guess I will grab some cold drink ... it was big learing
> experiance
> > > for me, to make this work.
> > >
> > > So this is my first patch file, if there is something else needed
> or I
> > > can support in someway, let me know. ( I know SchedulerStorage is
> not
> > > perfect solution, but again: it works for me ; but I agree, this
> > can be
> > > made better )
> > >
> > >
> > > I know I do not say it much ( ever ) -- Apache Karaf is really
> great
> > > platform, it is just that I send email, when I have problems :D ...
> > >
> > > Kind Regards,
> > > Miroslav
> > >
> > >
> > > --
> > > Miroslav Beranič
> > > MIBESIS
> > > +386(0)40/814-843
> > > miroslav.beranic@mibesis.si <ma...@mibesis.si>
> > <mailto:miroslav.beranic@mibesis.si
> > <ma...@mibesis.si>>
> > > http://www.mibesis.si
> >
> > --
> > Jean-Baptiste Onofré
> > jbonofre@apache.org <ma...@apache.org>
> > http://blog.nanthrax.net
> > Talend - http://www.talend.com
> >
> >
> >
> > --
> > Miroslav Beranič
> > MIBESIS
> > +386(0)40/814-843
> > miroslav.beranic@mibesis.si <ma...@mibesis.si>
> > http://www.mibesis.si
>
> --
> Jean-Baptiste Onofré
> jbonofre@apache.org
> http://blog.nanthrax.net
> Talend - http://www.talend.com
>
--
Miroslav Beranič
MIBESIS
+386(0)40/814-843
miroslav.beranic@mibesis.si
http://www.mibesis.si
Re: Apache Karaf 4.2.2 ( 4.2.x ) Quartz support, with JDBC
Persistence and C3P0 JDBC pool
Posted by Jean-Baptiste Onofré <jb...@nanthrax.net>.
Oh sorry, PR means Pull Request on Github.
You can find some details on the contribution guide:
http://karaf.apache.org/community.html
Don't hesitate to ping me (maybe you can join the Karaf Slack channel).
I would love to discuss with you about that.
Regards
JB
On 07/09/2018 14:53, Miroslav Beranič wrote:
> Hi JB,
>
> Scheduler with use on Cellar is my semi-next task, scheduled for
> beginning of October.
>
> PR? Sorry, I am not into this, so PR stands for?
>
> Kind Regards,
> Miroslav
>
>
>
>
> V V pet., 7. sep. 2018 ob 14:48 je oseba Jean-Baptiste Onofré
> <jb@nanthrax.net <ma...@nanthrax.net>> napisala:
>
> Hi,
>
> thanks for the your detailed update on the scheduler.
>
> I did the change in Karaf 4.2.x to be able to support Quartz scheduler
> in cluster mode (especially when used with Cellar).
>
> I worked fine when I tested, using dynamic import, so I'm a bit
> surprised it doesn't work out of the box for you.
>
> Let me take a look on your patch file to identify the missing parts. In
> the mean time, do you mind to prepare a PR ?
>
> Thanks !
> Regards
> JB
>
> On 07/09/2018 14:22, Miroslav Beranič wrote:
> > Hi guys,
> >
> > I am working and talking about Apache Karaf version 4.2.2-SNAPSHOT (
> > current master on Github ).
> >
> > I was rally happy to see Scheduler service with Quartz as default
> > implementation. Without much thinking, I've hooked up Quartz with
> JDBC (
> > I would really like Karaf Scheduler ( Quartz ) to use existing
> > Datasource and JPA but no luck ).
> >
> > When I moved forward, I did not care about JPA that much any more, as
> > even JDBC was not working. To begin, this is what I've changed in my
> > Karaf configuration:
> >
> > file: etc/org.apache.karaf.scheduler.quartz.cfg
> > org.quartz.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
> >
> org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.oracle.OracleDelegate
> > org.quartz.jobStore.useProperties=false
> > org.quartz.jobStore.dataSource=MyQuartzDS
> > org.quartz.jobStore.isClustered=true
> > org.quartz.jobStore.tablePrefix=q_ # Oracle has 30 char limit on
> > identifier, I started to cry, when I saw error message in SQL
> terminal.
> > I use Oracle 12c1
> > org.quartz.jobStore.clusterCheckinInterval=30000
> >
> >
> org.quartz.dataSource.MyQuartzDS.driver=oracle.jdbc.driver.OracleDriver
> >
> org.quartz.dataSource.MyQuartzDS.URL=jdbc:oracle:thin:@//oracle.local:1521/test
> > org.quartz.dataSource.MyQuartzDS.user=APP
> > org.quartz.dataSource.MyQuartzDS.password=app
> > org.quartz.dataSource.MyQuartzDS.maxConnections=5
> > org.quartz.dataSource.MyQuartzDS.validationQuery=select * from dual;
> >
> > org.quartz.scheduler.jmx.export=true
> > org.quartz.scheduler.rmi.export=false
> > org.quartz.scheduler.rmi.proxy=false
> >
> > This seems to be legit. As I saw in
> > org.apache.karaf.scheduler.core.Activator#doStart method, all / only
> > properties with key prefix "org.quartz" are forwarded to Quartz. Looks
> > OK, but I guess "org.quartz." ( with a dot at the end ) would be more
> > correct. But, minor issue.
> >
> > Lets start it up. Well, no - it will not. To not write all here, all
> > error messages, all missing error messages, but strange behavior,
> let me
> > go to, what I think, is a solution.
> >
> > So steps to make your Karaf Scheduler ( 4.2.x ) work with Quartz (
> > 2.2.x ) , I did this:
> >
> > 1.) Karaf 4.2.2 is using Quartz 2.2.1, that is OK, it could be
> upgraded
> > to 2.3.0, but I do not need it ( as I am porting application from
> Quartz
> > 1.8.5, this is way of my list of worries )
> > 2.) scheduler/pom.xml has package imports defined as:
> > <Import-Package>
> > !com.mchange.*,
> > !oracle.*,
> > !org.quartz.*,
> > !weblogic.*,
> > !javax.transaction,
> > javax.servlet*;resolution:=optional,
> > org.jboss.*;resolution:=optional,
> > *
> > </Import-Package>
> >
> > ( I really am not OSGi expert, and all, but this looked strange to
> me ),
> > turned out be "true" ( I guess, as I am open to option, I am
> wrong, but
> > end of the day - this works for me, so for me -t his is correct )
> >
> > I changed this to:
> > <Import-Package>
> > com.mchange.*;resolution:=optional,
> > oracle.*;resolution:=optional,
> > org.quartz.*,
> > !weblogic.*,
> > javax.transaction,
> > javax.servlet*;resolution:=optional,
> > org.jboss.*;resolution:=optional,
> > *
> > </Import-Package>
> >
> > ( removed all the ! in front of the packages and added
> > resolution:=optional, as not all users want persistent Quartz jobs , I
> > guess )
> >
> > This is not all, now, when I deploy to Karaf, all sort of issue is
> > reported. Turned out I need to install
> > org.apache.servicemix.bundles.quartz bundle, from this url:
> >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> >
> > ( I did not install it from the URL, Karaf shell is doing this ... )
> >
> > but look, here is
> > <!--
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > -->
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.2.1_1</version>
> > </dependency>
> >
> > Nope, it will not work. As, same problem with Apache Karaf Scheduler's
> > POM file. Wrong imports. I guess someone found out this is wrong, and
> > created this version ( not direct next version, but few versions
> later on ):
> >
> > <!--
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > -->
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.3.0_2</version>
> > </dependency>
> >
> > But wait ... sure, this has different Quartz version. How nice ,
> why not
> > ... If we can upgrade, we should do it. But at this point, I was
> really
> > crying ... for muliple reasons, ... I was to much focused into my path
> > and versions I've decided to clone repository and change Quartz
> version
> > and name it 2.2.5.
> > So:
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.3.0_2</version>
> > </dependency>
> > with changed Quartz version to 2.2.1 is "my"
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.2.5</version>
> > </dependency>
> >
> > ( was this good or not, again: it works for me now, and I ran out of
> > hair to pull them out ... so, keep calm and move forward )
> >
> > But this is not end of problems. I forgot what exactly was wrong
> at this
> > point.
> >
> > I've installed feature:
> >
> > scr transaction jndi jpa hibernate pac-jdbc-oracle ( my own oracle
> jdbc
> > bundle ) aries-blueprint scheduler karaf-scheduler-example
> >
> > and bundles:
> > bundle:install
> >
> mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.c3p0/0.9.5.2_3-SNAPSHOT
> >
> > Looook, I am starting to see something. Yeey baby:
> >
> > 2018-09-07T09:43:58,678 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Using db table-based data access locking
> > (synchronization).
> > 2018-09-07T09:43:58,680 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | JobStoreTX initialized.
> > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > QuartzScheduler | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Scheduler meta-data: Quartz Scheduler (v2.2.1)
> > 'MyScheduler' with instanceId 'my'
> > Scheduler class: 'org.quartz.core.QuartzScheduler' - running
> locally.
> > NOT STARTED.
> > Currently in standby mode.
> > Number of jobs executed: 0
> > Using thread pool 'org.quartz.simpl.SimpleThreadPool' - with 20
> threads.
> > Using job-store 'org.quartz.impl.jdbcjobstore.JobStoreTX' - which
> > supports persistence. and is clustered.
> >
> > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > StdSchedulerFactory | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Quartz scheduler 'MyScheduler' initialized from an
> > externally provided properties instance.
> > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > StdSchedulerFactory | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Quartz scheduler version: 2.2.1
> > 2018-09-07T09:43:58,713 | INFO | activator-1-thread-1 |
> > AbstractPoolBackedDataSource | 107 -
> > org.apache.servicemix.bundles.c3p0 - 0.9.5.2_3-SNAPSHOT | Initializing
> > c3p0 pool... com.mchange.v2.c3p0.ComboPooledDataSource [
> > acquireIncrement -> 3, acquireRetryAttempts -> 30,
> acquireRetryDelay ->
> > 1000, autoCommitOnClose -> false, automaticTestTable -> null,
> > breakAfterAcquireFailure -> false, checkoutTimeout -> 0,
> > connectionCustomizerClassName -> null, connectionTesterClassName ->
> > com.mchange.v2.c3p0.impl.DefaultConnectionTester,
> > contextClassLoaderSource -> caller, dataSourceName ->
> > z8kfsx9x1dnli4w1yck8tp|3c42672a,
> debugUnreturnedConnectionStackTraces ->
> > false, description -> null, driverClass ->
> > oracle.jdbc.driver.OracleDriver, extensions -> {},
> factoryClassLocation
> > -> null, forceIgnoreUnresolvedTransactions -> false,
> > forceSynchronousCheckins -> false, forceUseNamedDriverClass -> false,
> > identityToken -> z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > idleConnectionTestPeriod -> 50, initialPoolSize -> 3, jdbcUrl ->
> > jdbc:oracle:thin:@//oracle.local:1521/test,
> maxAdministrativeTaskTime ->
> > 0, maxConnectionAge -> 0, maxIdleTime -> 0,
> maxIdleTimeExcessConnections
> > -> 0, maxPoolSize -> 5, maxStatements -> 0, maxStatementsPerConnection
> > -> 120, minPoolSize -> 1, numHelperThreads -> 3, preferredTestQuery ->
> > select * from dual;, privilegeSpawnedThreads -> false, properties ->
> > {user=******, password=******}, propertyCycle -> 0,
> > statementCacheNumDeferredCloseThreads -> 0, testConnectionOnCheckin ->
> > true, testConnectionOnCheckout -> false,
> unreturnedConnectionTimeout ->
> > 0, userOverrides -> {}, usesTraditionalReflectiveProxies -> false ]
> > 2018-09-07T09:43:58,820 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | ClusterManager: detected 1 failed or restarted
> instances.
> > 2018-09-07T09:43:58,821 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | ClusterManager: Scanning for instance "my"'s failed
> > in-progress jobs.
> > 2018-09-07T09:43:58,826 | INFO | activator-1-thread-1 |
> > QuartzScheduler | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Scheduler MyScheduler$_my started.
> > 2018-09-07T09:43:58,837 | INFO | activator-1-thread-1 |
> > CommandExtension | 35 - org.apache.karaf.shell.core -
> > 4.2.2.SNAPSHOT | Registering commands for bundle
> > org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT
> >
> > This is it! I did it. Time for big coffee. And than I came back at the
> > terminal and saw this:
> >
> > 2018-09-07T09:43:59,021 | WARN | activator-1-thread-1 |
> > WhiteboardHandler | 105 -
> org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Error scheduling job
> > org.apache.karaf.scheduler.SchedulerError:
> > org.quartz.JobPersistenceException: Couldn't store job: Unable to
> > serialize JobDataMap for insertion into database because the value of
> > property 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger [See nested exception:
> > java.io <http://java.io>.NotSerializableException: Unable to
> serialize JobDataMap for
> > insertion into database because the value of property
> > 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger]
> > at
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:239)
> > ~[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.register(WhiteboardHandler.java:140)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.access$100(WhiteboardHandler.java:37)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler$1.addingService(WhiteboardHandler.java:66)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:941)
> > [?:?]
> > at
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:870)
> > [?:?]
> > at
> >
> org.osgi.util.tracker.AbstractTracked.trackAdding(AbstractTracked.java:256)
> > [?:?]
> > at
> >
> org.osgi.util.tracker.AbstractTracked.trackInitial(AbstractTracked.java:183)
> > [?:?]
> > at
> > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:318)
> [?:?]
> > at
> > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:261)
> [?:?]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.<init>(WhiteboardHandler.java:71)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> > org.apache.karaf.scheduler.core.Activator.doStart(Activator.java:60)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.apache.karaf.util.tracker.BaseActivator.run(BaseActivator.java:275)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> > [?:?]
> > at java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:?]
> > at
> >
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> > [?:?]
> > at
> >
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> > [?:?]
> > at java.lang.Thread.run(Thread.java:748) [?:?]
> > Caused by: org.quartz.JobPersistenceException: Couldn't store job:
> > Unable to serialize JobDataMap for insertion into database because the
> > value of property 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1115)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > ~[?:?]
> > at
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886) ~[?:?]
> > at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > ~[?:?]
> > at
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > ~[?:?]
> > ... 17 more
> > Caused by: java.io <http://java.io>.NotSerializableException:
> Unable to serialize
> > JobDataMap for insertion into database because the value of property
> > 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > at
> >
> org.quartz.impl.jdbcjobstore.StdJDBCDelegate.serializeJobData(StdJDBCDelegate.java:3083)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.oracle.OracleDelegate.insertJobDetail(OracleDelegate.java:154)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1112)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > ~[?:?]
> > at
> >
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886) ~[?:?]
> > at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > ~[?:?]
> > at
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > ~[?:?]
> > ... 17 more
> >
> >
> > WAAAT!! No no no no. Well, this was just the beginning and I did not
> > know this at that time. So I started to look for
> > 'QuartzJobScheduler.Logger' in Karaf soruce code. Waaat, who
> thought it
> > is smart to store non-serializable data in Job data, whaaat.
> > At that point I started to realize, why it was so much pain to get
> > persistence working ... ( I guess ) no one tried or even wanted to
> make
> > Quartz persistent in Karaf - since this code is in from 2014.
> >
> > Scheduling is like one of the five most important features that I need
> > from the platform. I started to cry again. I did not test features,
> > before I pushed for the migration. There is no way back, so what
> can be
> > done now here.
> >
> > Let's see.
> >
> > I basically "re-wrote" Scheduler service in Karaf. What I did:
> >
> > 1.) Make sure Job data has only Serializable data -
> > - ScheduleOptions was made Serializable, with addition to move Trigger
> > creation from InternalScheduleOptions's constructor to method
> compile().
> > - getJobs() method from Scheduler interface returns String names
> and not
> > Objects
> > - Updated QuartzJobExecutor to remove non-serializable items from
> > JobDataMap storage
> >
> > 2.)
> > - Added custom StdScheduler and StdSchedulerFactory classes, to
> > implemented changes
> > - Added SchedulerStorage as support for "backward compatibility" and
> > "short lived" jobs. This is not good place to use it.
> >
> > 3.)
> > - Updated Scheduler Shell commands to use changed Scheduler API ( job
> > names are as keys )
> >
> > Patch file is in this email attachements, I've also pushed changes
> to my
> > Github repository, with only this changes:
> >
> > https://github.com/mibesis/apache-karaf/tree/karaf-4.2.2-quartz-jdbc
> >
> >
> > Anyone that is willing to use it -- please do, I would only ask Karf
> > core development team to "make sure" Quartz Job data is always
> fill with
> > Serializable data, as this makes all the difference and there is
> no need
> > for method getJobs() from Sheculer to return
> > Map<Object, ScheduleOptions> as it can not make data Serializable
> -- to
> > return Map<Object, ScheduleOptions> is OK, as all API calls are made
> > over keys, names, etc.
> >
> > I need few days, to come over this and coffee cups, it is Friday
> now, so
> > I guess I will grab some cold drink ... it was big learing experiance
> > for me, to make this work.
> >
> > So this is my first patch file, if there is something else needed or I
> > can support in someway, let me know. ( I know SchedulerStorage is not
> > perfect solution, but again: it works for me ; but I agree, this
> can be
> > made better )
> >
> >
> > I know I do not say it much ( ever ) -- Apache Karaf is really great
> > platform, it is just that I send email, when I have problems :D ...
> >
> > Kind Regards,
> > Miroslav
> >
> >
> > --
> > Miroslav Beranič
> > MIBESIS
> > +386(0)40/814-843
> > miroslav.beranic@mibesis.si <ma...@mibesis.si>
> <mailto:miroslav.beranic@mibesis.si
> <ma...@mibesis.si>>
> > http://www.mibesis.si
>
> --
> Jean-Baptiste Onofré
> jbonofre@apache.org <ma...@apache.org>
> http://blog.nanthrax.net
> Talend - http://www.talend.com
>
>
>
> --
> Miroslav Beranič
> MIBESIS
> +386(0)40/814-843
> miroslav.beranic@mibesis.si <ma...@mibesis.si>
> http://www.mibesis.si
--
Jean-Baptiste Onofré
jbonofre@apache.org
http://blog.nanthrax.net
Talend - http://www.talend.com
Re: Apache Karaf 4.2.2 ( 4.2.x ) Quartz support, with JDBC
Persistence and C3P0 JDBC pool
Posted by Miroslav Beranič <mi...@mibesis.si>.
Hi JB,
Scheduler with use on Cellar is my semi-next task, scheduled for beginning
of October.
PR? Sorry, I am not into this, so PR stands for?
Kind Regards,
Miroslav
V V pet., 7. sep. 2018 ob 14:48 je oseba Jean-Baptiste Onofré <
jb@nanthrax.net> napisala:
> Hi,
>
> thanks for the your detailed update on the scheduler.
>
> I did the change in Karaf 4.2.x to be able to support Quartz scheduler
> in cluster mode (especially when used with Cellar).
>
> I worked fine when I tested, using dynamic import, so I'm a bit
> surprised it doesn't work out of the box for you.
>
> Let me take a look on your patch file to identify the missing parts. In
> the mean time, do you mind to prepare a PR ?
>
> Thanks !
> Regards
> JB
>
> On 07/09/2018 14:22, Miroslav Beranič wrote:
> > Hi guys,
> >
> > I am working and talking about Apache Karaf version 4.2.2-SNAPSHOT (
> > current master on Github ).
> >
> > I was rally happy to see Scheduler service with Quartz as default
> > implementation. Without much thinking, I've hooked up Quartz with JDBC (
> > I would really like Karaf Scheduler ( Quartz ) to use existing
> > Datasource and JPA but no luck ).
> >
> > When I moved forward, I did not care about JPA that much any more, as
> > even JDBC was not working. To begin, this is what I've changed in my
> > Karaf configuration:
> >
> > file: etc/org.apache.karaf.scheduler.quartz.cfg
> > org.quartz.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
> >
> org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.oracle.OracleDelegate
> > org.quartz.jobStore.useProperties=false
> > org.quartz.jobStore.dataSource=MyQuartzDS
> > org.quartz.jobStore.isClustered=true
> > org.quartz.jobStore.tablePrefix=q_ # Oracle has 30 char limit on
> > identifier, I started to cry, when I saw error message in SQL terminal.
> > I use Oracle 12c1
> > org.quartz.jobStore.clusterCheckinInterval=30000
> >
> > org.quartz.dataSource.MyQuartzDS.driver=oracle.jdbc.driver.OracleDriver
> > org.quartz.dataSource.MyQuartzDS.URL=jdbc:oracle:thin:@
> //oracle.local:1521/test
> > org.quartz.dataSource.MyQuartzDS.user=APP
> > org.quartz.dataSource.MyQuartzDS.password=app
> > org.quartz.dataSource.MyQuartzDS.maxConnections=5
> > org.quartz.dataSource.MyQuartzDS.validationQuery=select * from dual;
> >
> > org.quartz.scheduler.jmx.export=true
> > org.quartz.scheduler.rmi.export=false
> > org.quartz.scheduler.rmi.proxy=false
> >
> > This seems to be legit. As I saw in
> > org.apache.karaf.scheduler.core.Activator#doStart method, all / only
> > properties with key prefix "org.quartz" are forwarded to Quartz. Looks
> > OK, but I guess "org.quartz." ( with a dot at the end ) would be more
> > correct. But, minor issue.
> >
> > Lets start it up. Well, no - it will not. To not write all here, all
> > error messages, all missing error messages, but strange behavior, let me
> > go to, what I think, is a solution.
> >
> > So steps to make your Karaf Scheduler ( 4.2.x ) work with Quartz (
> > 2.2.x ) , I did this:
> >
> > 1.) Karaf 4.2.2 is using Quartz 2.2.1, that is OK, it could be upgraded
> > to 2.3.0, but I do not need it ( as I am porting application from Quartz
> > 1.8.5, this is way of my list of worries )
> > 2.) scheduler/pom.xml has package imports defined as:
> > <Import-Package>
> > !com.mchange.*,
> > !oracle.*,
> > !org.quartz.*,
> > !weblogic.*,
> > !javax.transaction,
> > javax.servlet*;resolution:=optional,
> > org.jboss.*;resolution:=optional,
> > *
> > </Import-Package>
> >
> > ( I really am not OSGi expert, and all, but this looked strange to me ),
> > turned out be "true" ( I guess, as I am open to option, I am wrong, but
> > end of the day - this works for me, so for me -t his is correct )
> >
> > I changed this to:
> > <Import-Package>
> > com.mchange.*;resolution:=optional,
> > oracle.*;resolution:=optional,
> > org.quartz.*,
> > !weblogic.*,
> > javax.transaction,
> > javax.servlet*;resolution:=optional,
> > org.jboss.*;resolution:=optional,
> > *
> > </Import-Package>
> >
> > ( removed all the ! in front of the packages and added
> > resolution:=optional, as not all users want persistent Quartz jobs , I
> > guess )
> >
> > This is not all, now, when I deploy to Karaf, all sort of issue is
> > reported. Turned out I need to install
> > org.apache.servicemix.bundles.quartz bundle, from this url:
> >
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> >
> > ( I did not install it from the URL, Karaf shell is doing this ... )
> >
> > but look, here is
> > <!--
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > -->
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.2.1_1</version>
> > </dependency>
> >
> > Nope, it will not work. As, same problem with Apache Karaf Scheduler's
> > POM file. Wrong imports. I guess someone found out this is wrong, and
> > created this version ( not direct next version, but few versions later
> on ):
> >
> > <!--
> >
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> > -->
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.3.0_2</version>
> > </dependency>
> >
> > But wait ... sure, this has different Quartz version. How nice , why not
> > ... If we can upgrade, we should do it. But at this point, I was really
> > crying ... for muliple reasons, ... I was to much focused into my path
> > and versions I've decided to clone repository and change Quartz version
> > and name it 2.2.5.
> > So:
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.3.0_2</version>
> > </dependency>
> > with changed Quartz version to 2.2.1 is "my"
> > <dependency>
> > <groupId>org.apache.servicemix.bundles</groupId>
> > <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> > <version>2.2.5</version>
> > </dependency>
> >
> > ( was this good or not, again: it works for me now, and I ran out of
> > hair to pull them out ... so, keep calm and move forward )
> >
> > But this is not end of problems. I forgot what exactly was wrong at this
> > point.
> >
> > I've installed feature:
> >
> > scr transaction jndi jpa hibernate pac-jdbc-oracle ( my own oracle jdbc
> > bundle ) aries-blueprint scheduler karaf-scheduler-example
> >
> > and bundles:
> > bundle:install
> >
> mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.c3p0/0.9.5.2_3-SNAPSHOT
> >
> > Looook, I am starting to see something. Yeey baby:
> >
> > 2018-09-07T09:43:58,678 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Using db table-based data access locking
> > (synchronization).
> > 2018-09-07T09:43:58,680 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | JobStoreTX initialized.
> > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > QuartzScheduler | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Scheduler meta-data: Quartz Scheduler (v2.2.1)
> > 'MyScheduler' with instanceId 'my'
> > Scheduler class: 'org.quartz.core.QuartzScheduler' - running locally.
> > NOT STARTED.
> > Currently in standby mode.
> > Number of jobs executed: 0
> > Using thread pool 'org.quartz.simpl.SimpleThreadPool' - with 20
> threads.
> > Using job-store 'org.quartz.impl.jdbcjobstore.JobStoreTX' - which
> > supports persistence. and is clustered.
> >
> > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > StdSchedulerFactory | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Quartz scheduler 'MyScheduler' initialized from an
> > externally provided properties instance.
> > 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> > StdSchedulerFactory | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Quartz scheduler version: 2.2.1
> > 2018-09-07T09:43:58,713 | INFO | activator-1-thread-1 |
> > AbstractPoolBackedDataSource | 107 -
> > org.apache.servicemix.bundles.c3p0 - 0.9.5.2_3-SNAPSHOT | Initializing
> > c3p0 pool... com.mchange.v2.c3p0.ComboPooledDataSource [
> > acquireIncrement -> 3, acquireRetryAttempts -> 30, acquireRetryDelay ->
> > 1000, autoCommitOnClose -> false, automaticTestTable -> null,
> > breakAfterAcquireFailure -> false, checkoutTimeout -> 0,
> > connectionCustomizerClassName -> null, connectionTesterClassName ->
> > com.mchange.v2.c3p0.impl.DefaultConnectionTester,
> > contextClassLoaderSource -> caller, dataSourceName ->
> > z8kfsx9x1dnli4w1yck8tp|3c42672a, debugUnreturnedConnectionStackTraces ->
> > false, description -> null, driverClass ->
> > oracle.jdbc.driver.OracleDriver, extensions -> {}, factoryClassLocation
> > -> null, forceIgnoreUnresolvedTransactions -> false,
> > forceSynchronousCheckins -> false, forceUseNamedDriverClass -> false,
> > identityToken -> z8kfsx9x1dnli4w1yck8tp|3c42672a,
> > idleConnectionTestPeriod -> 50, initialPoolSize -> 3, jdbcUrl ->
> > jdbc:oracle:thin:@//oracle.local:1521/test, maxAdministrativeTaskTime ->
> > 0, maxConnectionAge -> 0, maxIdleTime -> 0, maxIdleTimeExcessConnections
> > -> 0, maxPoolSize -> 5, maxStatements -> 0, maxStatementsPerConnection
> > -> 120, minPoolSize -> 1, numHelperThreads -> 3, preferredTestQuery ->
> > select * from dual;, privilegeSpawnedThreads -> false, properties ->
> > {user=******, password=******}, propertyCycle -> 0,
> > statementCacheNumDeferredCloseThreads -> 0, testConnectionOnCheckin ->
> > true, testConnectionOnCheckout -> false, unreturnedConnectionTimeout ->
> > 0, userOverrides -> {}, usesTraditionalReflectiveProxies -> false ]
> > 2018-09-07T09:43:58,820 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | ClusterManager: detected 1 failed or restarted
> instances.
> > 2018-09-07T09:43:58,821 | INFO | activator-1-thread-1 |
> > JobStoreTX | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | ClusterManager: Scanning for instance "my"'s failed
> > in-progress jobs.
> > 2018-09-07T09:43:58,826 | INFO | activator-1-thread-1 |
> > QuartzScheduler | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Scheduler MyScheduler$_my started.
> > 2018-09-07T09:43:58,837 | INFO | activator-1-thread-1 |
> > CommandExtension | 35 - org.apache.karaf.shell.core -
> > 4.2.2.SNAPSHOT | Registering commands for bundle
> > org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT
> >
> > This is it! I did it. Time for big coffee. And than I came back at the
> > terminal and saw this:
> >
> > 2018-09-07T09:43:59,021 | WARN | activator-1-thread-1 |
> > WhiteboardHandler | 105 - org.apache.karaf.scheduler.core
> > - 4.2.2.SNAPSHOT | Error scheduling job
> > org.apache.karaf.scheduler.SchedulerError:
> > org.quartz.JobPersistenceException: Couldn't store job: Unable to
> > serialize JobDataMap for insertion into database because the value of
> > property 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger [See nested exception:
> > java.io.NotSerializableException: Unable to serialize JobDataMap for
> > insertion into database because the value of property
> > 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger]
> > at
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:239)
> > ~[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.register(WhiteboardHandler.java:140)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.access$100(WhiteboardHandler.java:37)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler$1.addingService(WhiteboardHandler.java:66)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:941)
> > [?:?]
> > at
> >
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:870)
> > [?:?]
> > at
> >
> org.osgi.util.tracker.AbstractTracked.trackAdding(AbstractTracked.java:256)
> > [?:?]
> > at
> >
> org.osgi.util.tracker.AbstractTracked.trackInitial(AbstractTracked.java:183)
> > [?:?]
> > at
> > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:318) [?:?]
> > at
> > org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:261) [?:?]
> > at
> >
> org.apache.karaf.scheduler.core.WhiteboardHandler.<init>(WhiteboardHandler.java:71)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> > org.apache.karaf.scheduler.core.Activator.doStart(Activator.java:60)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> > org.apache.karaf.util.tracker.BaseActivator.run(BaseActivator.java:275)
> > [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> > at
> > java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> > [?:?]
> > at java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:?]
> > at
> >
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> > [?:?]
> > at
> >
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> > [?:?]
> > at java.lang.Thread.run(Thread.java:748) [?:?]
> > Caused by: org.quartz.JobPersistenceException: Couldn't store job:
> > Unable to serialize JobDataMap for insertion into database because the
> > value of property 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1115)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > ~[?:?]
> > at
> > org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> ~[?:?]
> > at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > ~[?:?]
> > at
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > ~[?:?]
> > ... 17 more
> > Caused by: java.io.NotSerializableException: Unable to serialize
> > JobDataMap for insertion into database because the value of property
> > 'QuartzJobScheduler.Logger' is not serializable:
> > org.ops4j.pax.logging.slf4j.Slf4jLogger
> > at
> >
> org.quartz.impl.jdbcjobstore.StdJDBCDelegate.serializeJobData(StdJDBCDelegate.java:3083)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.oracle.OracleDelegate.insertJobDetail(OracleDelegate.java:154)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1112)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> > ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93)
> ~[?:?]
> > at
> >
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> > ~[?:?]
> > at
> > org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886)
> ~[?:?]
> > at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> > ~[?:?]
> > at
> >
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> > ~[?:?]
> > ... 17 more
> >
> >
> > WAAAT!! No no no no. Well, this was just the beginning and I did not
> > know this at that time. So I started to look for
> > 'QuartzJobScheduler.Logger' in Karaf soruce code. Waaat, who thought it
> > is smart to store non-serializable data in Job data, whaaat.
> > At that point I started to realize, why it was so much pain to get
> > persistence working ... ( I guess ) no one tried or even wanted to make
> > Quartz persistent in Karaf - since this code is in from 2014.
> >
> > Scheduling is like one of the five most important features that I need
> > from the platform. I started to cry again. I did not test features,
> > before I pushed for the migration. There is no way back, so what can be
> > done now here.
> >
> > Let's see.
> >
> > I basically "re-wrote" Scheduler service in Karaf. What I did:
> >
> > 1.) Make sure Job data has only Serializable data -
> > - ScheduleOptions was made Serializable, with addition to move Trigger
> > creation from InternalScheduleOptions's constructor to method compile().
> > - getJobs() method from Scheduler interface returns String names and not
> > Objects
> > - Updated QuartzJobExecutor to remove non-serializable items from
> > JobDataMap storage
> >
> > 2.)
> > - Added custom StdScheduler and StdSchedulerFactory classes, to
> > implemented changes
> > - Added SchedulerStorage as support for "backward compatibility" and
> > "short lived" jobs. This is not good place to use it.
> >
> > 3.)
> > - Updated Scheduler Shell commands to use changed Scheduler API ( job
> > names are as keys )
> >
> > Patch file is in this email attachements, I've also pushed changes to my
> > Github repository, with only this changes:
> >
> > https://github.com/mibesis/apache-karaf/tree/karaf-4.2.2-quartz-jdbc
> >
> >
> > Anyone that is willing to use it -- please do, I would only ask Karf
> > core development team to "make sure" Quartz Job data is always fill with
> > Serializable data, as this makes all the difference and there is no need
> > for method getJobs() from Sheculer to return
> > Map<Object, ScheduleOptions> as it can not make data Serializable -- to
> > return Map<Object, ScheduleOptions> is OK, as all API calls are made
> > over keys, names, etc.
> >
> > I need few days, to come over this and coffee cups, it is Friday now, so
> > I guess I will grab some cold drink ... it was big learing experiance
> > for me, to make this work.
> >
> > So this is my first patch file, if there is something else needed or I
> > can support in someway, let me know. ( I know SchedulerStorage is not
> > perfect solution, but again: it works for me ; but I agree, this can be
> > made better )
> >
> >
> > I know I do not say it much ( ever ) -- Apache Karaf is really great
> > platform, it is just that I send email, when I have problems :D ...
> >
> > Kind Regards,
> > Miroslav
> >
> >
> > --
> > Miroslav Beranič
> > MIBESIS
> > +386(0)40/814-843
> > miroslav.beranic@mibesis.si <ma...@mibesis.si>
> > http://www.mibesis.si
>
> --
> Jean-Baptiste Onofré
> jbonofre@apache.org
> http://blog.nanthrax.net
> Talend - http://www.talend.com
>
--
Miroslav Beranič
MIBESIS
+386(0)40/814-843
miroslav.beranic@mibesis.si
http://www.mibesis.si
Re: Apache Karaf 4.2.2 ( 4.2.x ) Quartz support, with JDBC
Persistence and C3P0 JDBC pool
Posted by Jean-Baptiste Onofré <jb...@nanthrax.net>.
Hi,
thanks for the your detailed update on the scheduler.
I did the change in Karaf 4.2.x to be able to support Quartz scheduler
in cluster mode (especially when used with Cellar).
I worked fine when I tested, using dynamic import, so I'm a bit
surprised it doesn't work out of the box for you.
Let me take a look on your patch file to identify the missing parts. In
the mean time, do you mind to prepare a PR ?
Thanks !
Regards
JB
On 07/09/2018 14:22, Miroslav Beranič wrote:
> Hi guys,
>
> I am working and talking about Apache Karaf version 4.2.2-SNAPSHOT (
> current master on Github ).
>
> I was rally happy to see Scheduler service with Quartz as default
> implementation. Without much thinking, I've hooked up Quartz with JDBC (
> I would really like Karaf Scheduler ( Quartz ) to use existing
> Datasource and JPA but no luck ).
>
> When I moved forward, I did not care about JPA that much any more, as
> even JDBC was not working. To begin, this is what I've changed in my
> Karaf configuration:
>
> file: etc/org.apache.karaf.scheduler.quartz.cfg
> org.quartz.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
> org.quartz.jobStore.driverDelegateClass=org.quartz.impl.jdbcjobstore.oracle.OracleDelegate
> org.quartz.jobStore.useProperties=false
> org.quartz.jobStore.dataSource=MyQuartzDS
> org.quartz.jobStore.isClustered=true
> org.quartz.jobStore.tablePrefix=q_ # Oracle has 30 char limit on
> identifier, I started to cry, when I saw error message in SQL terminal.
> I use Oracle 12c1
> org.quartz.jobStore.clusterCheckinInterval=30000
>
> org.quartz.dataSource.MyQuartzDS.driver=oracle.jdbc.driver.OracleDriver
> org.quartz.dataSource.MyQuartzDS.URL=jdbc:oracle:thin:@//oracle.local:1521/test
> org.quartz.dataSource.MyQuartzDS.user=APP
> org.quartz.dataSource.MyQuartzDS.password=app
> org.quartz.dataSource.MyQuartzDS.maxConnections=5
> org.quartz.dataSource.MyQuartzDS.validationQuery=select * from dual;
>
> org.quartz.scheduler.jmx.export=true
> org.quartz.scheduler.rmi.export=false
> org.quartz.scheduler.rmi.proxy=false
>
> This seems to be legit. As I saw in
> org.apache.karaf.scheduler.core.Activator#doStart method, all / only
> properties with key prefix "org.quartz" are forwarded to Quartz. Looks
> OK, but I guess "org.quartz." ( with a dot at the end ) would be more
> correct. But, minor issue.
>
> Lets start it up. Well, no - it will not. To not write all here, all
> error messages, all missing error messages, but strange behavior, let me
> go to, what I think, is a solution.
>
> So steps to make your Karaf Scheduler ( 4.2.x ) work with Quartz (
> 2.2.x ) , I did this:
>
> 1.) Karaf 4.2.2 is using Quartz 2.2.1, that is OK, it could be upgraded
> to 2.3.0, but I do not need it ( as I am porting application from Quartz
> 1.8.5, this is way of my list of worries )
> 2.) scheduler/pom.xml has package imports defined as:
> <Import-Package>
> !com.mchange.*,
> !oracle.*,
> !org.quartz.*,
> !weblogic.*,
> !javax.transaction,
> javax.servlet*;resolution:=optional,
> org.jboss.*;resolution:=optional,
> *
> </Import-Package>
>
> ( I really am not OSGi expert, and all, but this looked strange to me ),
> turned out be "true" ( I guess, as I am open to option, I am wrong, but
> end of the day - this works for me, so for me -t his is correct )
>
> I changed this to:
> <Import-Package>
> com.mchange.*;resolution:=optional,
> oracle.*;resolution:=optional,
> org.quartz.*,
> !weblogic.*,
> javax.transaction,
> javax.servlet*;resolution:=optional,
> org.jboss.*;resolution:=optional,
> *
> </Import-Package>
>
> ( removed all the ! in front of the packages and added
> resolution:=optional, as not all users want persistent Quartz jobs , I
> guess )
>
> This is not all, now, when I deploy to Karaf, all sort of issue is
> reported. Turned out I need to install
> org.apache.servicemix.bundles.quartz bundle, from this url:
>
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
>
> ( I did not install it from the URL, Karaf shell is doing this ... )
>
> but look, here is
> <!--
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> -->
> <dependency>
> <groupId>org.apache.servicemix.bundles</groupId>
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> <version>2.2.1_1</version>
> </dependency>
>
> Nope, it will not work. As, same problem with Apache Karaf Scheduler's
> POM file. Wrong imports. I guess someone found out this is wrong, and
> created this version ( not direct next version, but few versions later on ):
>
> <!--
> https://mvnrepository.com/artifact/org.apache.servicemix.bundles/org.apache.servicemix.bundles.quartz
> -->
> <dependency>
> <groupId>org.apache.servicemix.bundles</groupId>
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> <version>2.3.0_2</version>
> </dependency>
>
> But wait ... sure, this has different Quartz version. How nice , why not
> ... If we can upgrade, we should do it. But at this point, I was really
> crying ... for muliple reasons, ... I was to much focused into my path
> and versions I've decided to clone repository and change Quartz version
> and name it 2.2.5.
> So:
> <dependency>
> <groupId>org.apache.servicemix.bundles</groupId>
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> <version>2.3.0_2</version>
> </dependency>
> with changed Quartz version to 2.2.1 is "my"
> <dependency>
> <groupId>org.apache.servicemix.bundles</groupId>
> <artifactId>org.apache.servicemix.bundles.quartz</artifactId>
> <version>2.2.5</version>
> </dependency>
>
> ( was this good or not, again: it works for me now, and I ran out of
> hair to pull them out ... so, keep calm and move forward )
>
> But this is not end of problems. I forgot what exactly was wrong at this
> point.
>
> I've installed feature:
>
> scr transaction jndi jpa hibernate pac-jdbc-oracle ( my own oracle jdbc
> bundle ) aries-blueprint scheduler karaf-scheduler-example
>
> and bundles:
> bundle:install
> mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.c3p0/0.9.5.2_3-SNAPSHOT
>
> Looook, I am starting to see something. Yeey baby:
>
> 2018-09-07T09:43:58,678 | INFO | activator-1-thread-1 |
> JobStoreTX | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | Using db table-based data access locking
> (synchronization).
> 2018-09-07T09:43:58,680 | INFO | activator-1-thread-1 |
> JobStoreTX | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | JobStoreTX initialized.
> 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> QuartzScheduler | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | Scheduler meta-data: Quartz Scheduler (v2.2.1)
> 'MyScheduler' with instanceId 'my'
> Scheduler class: 'org.quartz.core.QuartzScheduler' - running locally.
> NOT STARTED.
> Currently in standby mode.
> Number of jobs executed: 0
> Using thread pool 'org.quartz.simpl.SimpleThreadPool' - with 20 threads.
> Using job-store 'org.quartz.impl.jdbcjobstore.JobStoreTX' - which
> supports persistence. and is clustered.
>
> 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> StdSchedulerFactory | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | Quartz scheduler 'MyScheduler' initialized from an
> externally provided properties instance.
> 2018-09-07T09:43:58,689 | INFO | activator-1-thread-1 |
> StdSchedulerFactory | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | Quartz scheduler version: 2.2.1
> 2018-09-07T09:43:58,713 | INFO | activator-1-thread-1 |
> AbstractPoolBackedDataSource | 107 -
> org.apache.servicemix.bundles.c3p0 - 0.9.5.2_3-SNAPSHOT | Initializing
> c3p0 pool... com.mchange.v2.c3p0.ComboPooledDataSource [
> acquireIncrement -> 3, acquireRetryAttempts -> 30, acquireRetryDelay ->
> 1000, autoCommitOnClose -> false, automaticTestTable -> null,
> breakAfterAcquireFailure -> false, checkoutTimeout -> 0,
> connectionCustomizerClassName -> null, connectionTesterClassName ->
> com.mchange.v2.c3p0.impl.DefaultConnectionTester,
> contextClassLoaderSource -> caller, dataSourceName ->
> z8kfsx9x1dnli4w1yck8tp|3c42672a, debugUnreturnedConnectionStackTraces ->
> false, description -> null, driverClass ->
> oracle.jdbc.driver.OracleDriver, extensions -> {}, factoryClassLocation
> -> null, forceIgnoreUnresolvedTransactions -> false,
> forceSynchronousCheckins -> false, forceUseNamedDriverClass -> false,
> identityToken -> z8kfsx9x1dnli4w1yck8tp|3c42672a,
> idleConnectionTestPeriod -> 50, initialPoolSize -> 3, jdbcUrl ->
> jdbc:oracle:thin:@//oracle.local:1521/test, maxAdministrativeTaskTime ->
> 0, maxConnectionAge -> 0, maxIdleTime -> 0, maxIdleTimeExcessConnections
> -> 0, maxPoolSize -> 5, maxStatements -> 0, maxStatementsPerConnection
> -> 120, minPoolSize -> 1, numHelperThreads -> 3, preferredTestQuery ->
> select * from dual;, privilegeSpawnedThreads -> false, properties ->
> {user=******, password=******}, propertyCycle -> 0,
> statementCacheNumDeferredCloseThreads -> 0, testConnectionOnCheckin ->
> true, testConnectionOnCheckout -> false, unreturnedConnectionTimeout ->
> 0, userOverrides -> {}, usesTraditionalReflectiveProxies -> false ]
> 2018-09-07T09:43:58,820 | INFO | activator-1-thread-1 |
> JobStoreTX | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | ClusterManager: detected 1 failed or restarted instances.
> 2018-09-07T09:43:58,821 | INFO | activator-1-thread-1 |
> JobStoreTX | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | ClusterManager: Scanning for instance "my"'s failed
> in-progress jobs.
> 2018-09-07T09:43:58,826 | INFO | activator-1-thread-1 |
> QuartzScheduler | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | Scheduler MyScheduler$_my started.
> 2018-09-07T09:43:58,837 | INFO | activator-1-thread-1 |
> CommandExtension | 35 - org.apache.karaf.shell.core -
> 4.2.2.SNAPSHOT | Registering commands for bundle
> org.apache.karaf.scheduler.core/4.2.2.SNAPSHOT
>
> This is it! I did it. Time for big coffee. And than I came back at the
> terminal and saw this:
>
> 2018-09-07T09:43:59,021 | WARN | activator-1-thread-1 |
> WhiteboardHandler | 105 - org.apache.karaf.scheduler.core
> - 4.2.2.SNAPSHOT | Error scheduling job
> org.apache.karaf.scheduler.SchedulerError:
> org.quartz.JobPersistenceException: Couldn't store job: Unable to
> serialize JobDataMap for insertion into database because the value of
> property 'QuartzJobScheduler.Logger' is not serializable:
> org.ops4j.pax.logging.slf4j.Slf4jLogger [See nested exception:
> java.io.NotSerializableException: Unable to serialize JobDataMap for
> insertion into database because the value of property
> 'QuartzJobScheduler.Logger' is not serializable:
> org.ops4j.pax.logging.slf4j.Slf4jLogger]
> at
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:239)
> ~[105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> at
> org.apache.karaf.scheduler.core.WhiteboardHandler.register(WhiteboardHandler.java:140)
> [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> at
> org.apache.karaf.scheduler.core.WhiteboardHandler.access$100(WhiteboardHandler.java:37)
> [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> at
> org.apache.karaf.scheduler.core.WhiteboardHandler$1.addingService(WhiteboardHandler.java:66)
> [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> at
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:941)
> [?:?]
> at
> org.osgi.util.tracker.ServiceTracker$Tracked.customizerAdding(ServiceTracker.java:870)
> [?:?]
> at
> org.osgi.util.tracker.AbstractTracked.trackAdding(AbstractTracked.java:256)
> [?:?]
> at
> org.osgi.util.tracker.AbstractTracked.trackInitial(AbstractTracked.java:183)
> [?:?]
> at
> org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:318) [?:?]
> at
> org.osgi.util.tracker.ServiceTracker.open(ServiceTracker.java:261) [?:?]
> at
> org.apache.karaf.scheduler.core.WhiteboardHandler.<init>(WhiteboardHandler.java:71)
> [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> at
> org.apache.karaf.scheduler.core.Activator.doStart(Activator.java:60)
> [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> at
> org.apache.karaf.util.tracker.BaseActivator.run(BaseActivator.java:275)
> [105:org.apache.karaf.scheduler.core:4.2.2.SNAPSHOT]
> at
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> [?:?]
> at java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:?]
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> [?:?]
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> [?:?]
> at java.lang.Thread.run(Thread.java:748) [?:?]
> Caused by: org.quartz.JobPersistenceException: Couldn't store job:
> Unable to serialize JobDataMap for insertion into database because the
> value of property 'QuartzJobScheduler.Logger' is not serializable:
> org.ops4j.pax.logging.slf4j.Slf4jLogger
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1115)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93) ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> ~[?:?]
> at
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886) ~[?:?]
> at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> ~[?:?]
> at
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> ~[?:?]
> ... 17 more
> Caused by: java.io.NotSerializableException: Unable to serialize
> JobDataMap for insertion into database because the value of property
> 'QuartzJobScheduler.Logger' is not serializable:
> org.ops4j.pax.logging.slf4j.Slf4jLogger
> at
> org.quartz.impl.jdbcjobstore.StdJDBCDelegate.serializeJobData(StdJDBCDelegate.java:3083)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.oracle.OracleDelegate.insertJobDetail(OracleDelegate.java:154)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJob(JobStoreSupport.java:1112)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport$2.executeVoid(JobStoreSupport.java:1062)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3703)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport$VoidTransactionCallback.execute(JobStoreSupport.java:3701)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport.executeInNonManagedTXLock(JobStoreSupport.java:3787)
> ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreTX.executeInLock(JobStoreTX.java:93) ~[?:?]
> at
> org.quartz.impl.jdbcjobstore.JobStoreSupport.storeJobAndTrigger(JobStoreSupport.java:1058)
> ~[?:?]
> at
> org.quartz.core.QuartzScheduler.scheduleJob(QuartzScheduler.java:886) ~[?:?]
> at org.quartz.impl.StdScheduler.scheduleJob(StdScheduler.java:249)
> ~[?:?]
> at
> org.apache.karaf.scheduler.core.QuartzScheduler.schedule(QuartzScheduler.java:237)
> ~[?:?]
> ... 17 more
>
>
> WAAAT!! No no no no. Well, this was just the beginning and I did not
> know this at that time. So I started to look for
> 'QuartzJobScheduler.Logger' in Karaf soruce code. Waaat, who thought it
> is smart to store non-serializable data in Job data, whaaat.
> At that point I started to realize, why it was so much pain to get
> persistence working ... ( I guess ) no one tried or even wanted to make
> Quartz persistent in Karaf - since this code is in from 2014.
>
> Scheduling is like one of the five most important features that I need
> from the platform. I started to cry again. I did not test features,
> before I pushed for the migration. There is no way back, so what can be
> done now here.
>
> Let's see.
>
> I basically "re-wrote" Scheduler service in Karaf. What I did:
>
> 1.) Make sure Job data has only Serializable data -
> - ScheduleOptions was made Serializable, with addition to move Trigger
> creation from InternalScheduleOptions's constructor to method compile().
> - getJobs() method from Scheduler interface returns String names and not
> Objects
> - Updated QuartzJobExecutor to remove non-serializable items from
> JobDataMap storage
>
> 2.)
> - Added custom StdScheduler and StdSchedulerFactory classes, to
> implemented changes
> - Added SchedulerStorage as support for "backward compatibility" and
> "short lived" jobs. This is not good place to use it.
>
> 3.)
> - Updated Scheduler Shell commands to use changed Scheduler API ( job
> names are as keys )
>
> Patch file is in this email attachements, I've also pushed changes to my
> Github repository, with only this changes:
>
> https://github.com/mibesis/apache-karaf/tree/karaf-4.2.2-quartz-jdbc
>
>
> Anyone that is willing to use it -- please do, I would only ask Karf
> core development team to "make sure" Quartz Job data is always fill with
> Serializable data, as this makes all the difference and there is no need
> for method getJobs() from Sheculer to return
> Map<Object, ScheduleOptions> as it can not make data Serializable -- to
> return Map<Object, ScheduleOptions> is OK, as all API calls are made
> over keys, names, etc.
>
> I need few days, to come over this and coffee cups, it is Friday now, so
> I guess I will grab some cold drink ... it was big learing experiance
> for me, to make this work.
>
> So this is my first patch file, if there is something else needed or I
> can support in someway, let me know. ( I know SchedulerStorage is not
> perfect solution, but again: it works for me ; but I agree, this can be
> made better )
>
>
> I know I do not say it much ( ever ) -- Apache Karaf is really great
> platform, it is just that I send email, when I have problems :D ...
>
> Kind Regards,
> Miroslav
>
>
> --
> Miroslav Beranič
> MIBESIS
> +386(0)40/814-843
> miroslav.beranic@mibesis.si <ma...@mibesis.si>
> http://www.mibesis.si
--
Jean-Baptiste Onofré
jbonofre@apache.org
http://blog.nanthrax.net
Talend - http://www.talend.com