You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ignite.apache.org by Yaroslav Molochkov <mo...@gmail.com> on 2020/11/02 16:42:32 UTC
Re: 2.9.1 release proposal
Guys,
should you agree that issues with the 2.9.1-rc tag are good enough
for a maintenance release, i'd like to give it a go.
On Thu, Oct 29, 2020 at 5:33 PM Alexey Zinoviev <za...@gmail.com>
wrote:
> Let's discuss the possible planning dates for feature freeze for 2.10, for
> example? Do you have any plans or ideas?
>
> чт, 29 окт. 2020 г. в 17:24, Andrey Gura <ag...@apache.org>:
>
> > Hi,
> >
> > I agree with Alexey. We should release 2.9.1 and start preparing for
> > the 2.10 release.
> >
> > 2.x releases usually take a lot of time. So we can release 2.9.1, and
> > even 2.9.2 before 2.10.
> >
> > On Thu, Oct 29, 2020 at 12:02 PM Alexey Goncharuk
> > <al...@gmail.com> wrote:
> > >
> > > Hello folks,
> > >
> > > I think we should start both 2.9.1 and 2.10. In practice, maintenance
> > > release contains only critical and usability bugfixes (for example, I
> > would
> > > include this ticket [1] to include in 2.9.1 as it prevents users from
> > using
> > > tracing) and is released much faster than a minor release.
> > >
> > > [1] https://issues.apache.org/jira/browse/IGNITE-13640
> > >
> > > вт, 27 окт. 2020 г. в 21:41, Guru Stron <gu...@gmail.com>:
> > >
> > > > Hello,
> > > >
> > > > Agree with Pavel.
> > > >
> > > > On Tue, 27 Oct 2020 at 19:22, Pavel Tupitsyn <pt...@apache.org>
> > wrote:
> > > >
> > > > > Igniters,
> > > > >
> > > > > I think we should plan 2.10 instead of 2.9.1.
> > > > > ignite-2.9 branch was cut 4 months ago, a bunch of new features are
> > > > waiting
> > > > > to be released.
> > > > >
> > > > > On Tue, Oct 27, 2020 at 4:23 AM 18624049226 <18...@163.com>
> > wrote:
> > > > >
> > > > > > Hello,
> > > > > >
> > > > > > I suggest that the remaining document issue in version 2.9.0 can
> be
> > > > > > solved in version 2.9.1, and it is not a good practice to
> postpone
> > to
> > > > > > version 2.10.
> > > > > >
> > > > > > 在 2020/10/27 上午2:00, Maxim Muzafarov 写道:
> > > > > > > Hello,
> > > > > > >
> > > > > > > +1
> > > > > > > Should we start the discussion about the release leader and
> > release
> > > > > > dates?
> > > > > > >
> > > > > > > On Tue, 20 Oct 2020 at 10:12, Anton Vinogradov <av...@apache.org>
> > > > wrote:
> > > > > > >> +1 to start the 2.9.1 release process once as soon as 2.9
> > released.
> > > > > > >>
> > > > > > >> On Mon, Oct 19, 2020 at 8:49 PM Nikolay Izhikov <
> > > > nizhikov@apache.org>
> > > > > > wrote:
> > > > > > >>
> > > > > > >>> Hello, Yaroslav.
> > > > > > >>>
> > > > > > >>> I support the idea.
> > > > > > >>> But, we should carefully work with the release scope.
> > > > > > >>>
> > > > > > >>> IGNITE-13477 - fix for a SQL tracing that will be available
> > only in
> > > > > > 2.10
> > > > > > >>> IGNITE-13427 - fix for a new system view that exists in
> master
> > > > only,
> > > > > we
> > > > > > >>> need to include IGNITE-13409
> > > > > > >>>
> > > > > > >>> Other tickets should be checked, also.
> > > > > > >>> Is this a real fix of a released bug or we fix some issue we
> > bring
> > > > > with
> > > > > > >>> the brand new contribution.
> > > > > > >>>
> > > > > > >>>
> > > > > > >>> I propose to include the following tickets, also:
> > > > > > >>>
> > > > > > >>> CMD tools improvements:
> > > > > > >>>
> > > > > > >>> IGNITE-13488 - Command to print metric value
> > > > > > >>> IGNITE-13426 - Command to print system view content
> > > > > > >>> IGNITE-13422 - Parameter to explicitly enable experimental
> > commands
> > > > > > >>>
> > > > > > >>> IGNITE-13380 - Output IgniteSystemProperties via ignite.sh
> > > > > > >>>
> > > > > > >>> New system views:
> > > > > > >>>
> > > > > > >>> IGNITE-13409 Metastorage and DistributedMetastorage viewы.
> > > > > > >>> IGNITE-13408 BinaryMetadata view.
> > > > > > >>>
> > > > > > >>>
> > > > > > >>>> 19 окт. 2020 г., в 18:20, Yaroslav Molochkov <
> > > > molochkovyn@gmail.com
> > > > > >
> > > > > > >>> написал(а):
> > > > > > >>>> Hello, Igniters!
> > > > > > >>>>
> > > > > > >>>> I've compiled a list of tickets that, i think, deserve to be
> > > > > released
> > > > > > in
> > > > > > >>> a
> > > > > > >>>> minor Ignite release (meaning in 2.9.1) after 2.9. Here they
> > are:
> > > > > > >>>>
> > > > > > >>>> IGNITE-13569
> > > > > > >>>> disable archiving + walCompactionEnabled probably broke
> > reading
> > > > from
> > > > > > wal
> > > > > > >>> on
> > > > > > >>>> server restart
> > > > > > >>>>
> > > > > > >>>> IGNITE-13418
> > > > > > >>>> Deadlock on multiple cache delete
> > > > > > >>>>
> > > > > > >>>> IGNITE-13563
> > > > > > >>>> Deserializing IBinaryObject containing an IBinaryObject
> field
> > > > fails
> > > > > > >>>>
> > > > > > >>>> IGNITE-13575
> > > > > > >>>> Invalid blocking section in GridNioWorker and
> > GridNioClientWorker
> > > > > > leads
> > > > > > >>> to
> > > > > > >>>> false positive blocking thread detection
> > > > > > >>>>
> > > > > > >>>> IGNITE-13458
> > > > > > >>>> RebalancingPartitionsTotal metrics
> > > > > > >>>>
> > > > > > >>>> IGNITE-13536
> > > > > > >>>> .NET: Child processes become zombies when persistence is
> used
> > with
> > > > > > >>>> direct-io on Linux
> > > > > > >>>>
> > > > > > >>>> IGNITE-13500
> > > > > > >>>> Checkpoint read lock fail if it is taking under write lock
> > during
> > > > > the
> > > > > > >>>> stopping node
> > > > > > >>>>
> > > > > > >>>> IGNITE-13431
> > > > > > >>>> NPE during Cassandra Store initialization with PRIMITIVE
> > strategy
> > > > > > >>>>
> > > > > > >>>> IGNITE-13417
> > > > > > >>>> Cache Interceptors deserialization on client nodes
> > > > > > >>>>
> > > > > > >>>> IGNITE-13495
> > > > > > >>>> ZookeeperDiscoverySpiMBeanImpl#getCoordinator can return
> > invalid
> > > > > node
> > > > > > as
> > > > > > >>>> coordinator
> > > > > > >>>>
> > > > > > >>>> IGNITE-13479
> > > > > > >>>> Both ignite.sh and control.sh use the same JVM_OPTS.
> > Control.sh
> > > > > > doesn't
> > > > > > >>>> start if JMX port was set
> > > > > > >>>>
> > > > > > >>>> IGNITE-11312
> > > > > > >>>> JDBC: Thin driver reports incorrect property names
> > > > > > >>>>
> > > > > > >>>> IGNITE-13462
> > > > > > >>>> .NET: Thin client Dispose hangs when continuous query is
> > active on
> > > > > > .NET
> > > > > > >>>> Core 3.x
> > > > > > >>>>
> > > > > > >>>> IGNITE-13484
> > > > > > >>>> C++ odbc-example losing some values if run with 1 additional
> > node
> > > > > > >>>>
> > > > > > >>>> IGNITE-13477
> > > > > > >>>> Fix NPE in SQL tracing implementation.
> > > > > > >>>>
> > > > > > >>>> IGNITE-13435
> > > > > > >>>> Fixing some unrecorded issues command warm-up control.sh
> > > > > > >>>>
> > > > > > >>>> IGNITE-13403
> > > > > > >>>> Update JDBC metadata to match actual capabilities
> > > > > > >>>>
> > > > > > >>>> IGNITE-13427
> > > > > > >>>> The local metastorage system view fails if unmarshallable
> > values
> > > > > > present
> > > > > > >>>>
> > > > > > >>>> IGNITE-13401
> > > > > > >>>> Unsupported protocol version exception when getting cache
> > > > > > configuration
> > > > > > >>>> from Java thin client
> > > > > > >>>>
> > > > > > >>>> IGNITE-13388
> > > > > > >>>> apache-ignite deb package depends on a non-existent package
> > and
> > > > > can't
> > > > > > be
> > > > > > >>>> installed on Debian 10
> > > > > > >>>>
> > > > > > >>>> IGNITE-13397
> > > > > > >>>> NPE in logSupplierDone(UUID nodeId)
> > > > > > >>>>
> > > > > > >>>> IGNITE-13296
> > > > > > >>>> .NET: TransactionImpl finalizer can crash the process
> > > > > > >>>>
> > > > > > >>>> IGNITE-13382
> > > > > > >>>> DurableBackgroundTask can abandon incomplete task
> > > > > > >>>>
> > > > > > >>>> IGNITE-12509
> > > > > > >>>> CACHE_REBALANCE_STOPPED event raises for wrong caches in
> case
> > of
> > > > > > >>> specified
> > > > > > >>>> RebalanceDelay
> > > > > > >>>>
> > > > > > >>>> IGNITE-13072
> > > > > > >>>> Synchronization problems when different classloaders are
> used
> > for
> > > > > > >>>> deployment of same class
> > > > > > >>>>
> > > > > > >>>> IGNITE-13379
> > > > > > >>>> Exception occur on SQL caches when client reconnect
> > > > > > >>>>
> > > > > > >>>> IGNITE-13363
> > > > > > >>>> GridDhtCacheEntry::toString locks
> > > > > > >>>>
> > > > > > >>>> IGNITE-13373
> > > > > > >>>> WAL segmentns do not released on
> releaseHistoryForPreloading()
> > > > > > >>>>
> > > > > > >>>> IGNITE-13540
> > > > > > >>>> Exchange worker, waiting for new task from queue, considered
> > as
> > > > > > blocked.
> > > > > > >>>>
> > > > > > >>>> IGNITE-13491
> > > > > > >>>> Fix incorrect topology snapshot logger output about
> > coordinator
> > > > > > change.
> > > > > > >>>>
> > > > > > >>>> IGNITE-13439
> > > > > > >>>> Printing detailed classpath slowdowns node initialization
> > > > > > >>>>
> > > > > > >>>> WDYT?
> > > > > > >>>
> > > > > >
> > > > > >
> > > > >
> > > >
> >
>
Re: 2.9.1 release proposal
Posted by Alexey Zinoviev <za...@gmail.com>.
I'm ok with code freeze dates
чт, 5 нояб. 2020 г. в 12:21, Maxim Muzafarov <mm...@apache.org>:
> Folks,
>
> What may be the comfortable dates of the 2.10 release to finish all
> the related development activities?
> I suggest release Apache Ignite 2.10 prior to the end of this year and
> focus on the 3.0 activities in 2021 with only bug-fix releases for the
> 2.10.x.
>
> I suggest the following release dates for the 2.10:
>
> Code Freeze: December 16th, 2020
> Release Date: December 30th, 2020
>
> On Tue, 3 Nov 2020 at 18:15, Yaroslav Molochkov <mo...@gmail.com>
> wrote:
> >
> > Nikolay, hi!
> >
> > Done.
> >
> > On Tue, Nov 3, 2020 at 10:48 AM Nikolay Izhikov <ni...@apache.org>
> wrote:
> >
> > > Hello, Yaroslav.
> > >
> > > Looks like we have fixVersion 2.9.1 in Jira.
> > > Let’s mark all tickets with label «2.9.1-rc1» with fixVersion=2.9.1 and
> > > use fixVersion instead of label further.
> > >
> > >
> > >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20IGNITE%20AND%20fixVersion%20%3D%202.9.1
> > >
> > > > 2 нояб. 2020 г., в 19:42, Yaroslav Molochkov <mo...@gmail.com>
> > > написал(а):
> > > >
> > > > Guys,
> > > >
> > > > should you agree that issues with the 2.9.1-rc tag are good enough
> > > > for a maintenance release, i'd like to give it a go.
> > > >
> > > > On Thu, Oct 29, 2020 at 5:33 PM Alexey Zinoviev <
> zaleslaw.sin@gmail.com>
> > > > wrote:
> > > >
> > > >> Let's discuss the possible planning dates for feature freeze for
> 2.10,
> > > for
> > > >> example? Do you have any plans or ideas?
> > > >>
> > > >> чт, 29 окт. 2020 г. в 17:24, Andrey Gura <ag...@apache.org>:
> > > >>
> > > >>> Hi,
> > > >>>
> > > >>> I agree with Alexey. We should release 2.9.1 and start preparing
> for
> > > >>> the 2.10 release.
> > > >>>
> > > >>> 2.x releases usually take a lot of time. So we can release 2.9.1,
> and
> > > >>> even 2.9.2 before 2.10.
> > > >>>
> > > >>> On Thu, Oct 29, 2020 at 12:02 PM Alexey Goncharuk
> > > >>> <al...@gmail.com> wrote:
> > > >>>>
> > > >>>> Hello folks,
> > > >>>>
> > > >>>> I think we should start both 2.9.1 and 2.10. In practice,
> maintenance
> > > >>>> release contains only critical and usability bugfixes (for
> example, I
> > > >>> would
> > > >>>> include this ticket [1] to include in 2.9.1 as it prevents users
> from
> > > >>> using
> > > >>>> tracing) and is released much faster than a minor release.
> > > >>>>
> > > >>>> [1] https://issues.apache.org/jira/browse/IGNITE-13640
> > > >>>>
> > > >>>> вт, 27 окт. 2020 г. в 21:41, Guru Stron <
> gurustronpublic@gmail.com>:
> > > >>>>
> > > >>>>> Hello,
> > > >>>>>
> > > >>>>> Agree with Pavel.
> > > >>>>>
> > > >>>>> On Tue, 27 Oct 2020 at 19:22, Pavel Tupitsyn <
> ptupitsyn@apache.org>
> > > >>> wrote:
> > > >>>>>
> > > >>>>>> Igniters,
> > > >>>>>>
> > > >>>>>> I think we should plan 2.10 instead of 2.9.1.
> > > >>>>>> ignite-2.9 branch was cut 4 months ago, a bunch of new features
> are
> > > >>>>> waiting
> > > >>>>>> to be released.
> > > >>>>>>
> > > >>>>>> On Tue, Oct 27, 2020 at 4:23 AM 18624049226 <
> 18624049226@163.com>
> > > >>> wrote:
> > > >>>>>>
> > > >>>>>>> Hello,
> > > >>>>>>>
> > > >>>>>>> I suggest that the remaining document issue in version 2.9.0
> can
> > > >> be
> > > >>>>>>> solved in version 2.9.1, and it is not a good practice to
> > > >> postpone
> > > >>> to
> > > >>>>>>> version 2.10.
> > > >>>>>>>
> > > >>>>>>> 在 2020/10/27 上午2:00, Maxim Muzafarov 写道:
> > > >>>>>>>> Hello,
> > > >>>>>>>>
> > > >>>>>>>> +1
> > > >>>>>>>> Should we start the discussion about the release leader and
> > > >>> release
> > > >>>>>>> dates?
> > > >>>>>>>>
> > > >>>>>>>> On Tue, 20 Oct 2020 at 10:12, Anton Vinogradov <av@apache.org
> >
> > > >>>>> wrote:
> > > >>>>>>>>> +1 to start the 2.9.1 release process once as soon as 2.9
> > > >>> released.
> > > >>>>>>>>>
> > > >>>>>>>>> On Mon, Oct 19, 2020 at 8:49 PM Nikolay Izhikov <
> > > >>>>> nizhikov@apache.org>
> > > >>>>>>> wrote:
> > > >>>>>>>>>
> > > >>>>>>>>>> Hello, Yaroslav.
> > > >>>>>>>>>>
> > > >>>>>>>>>> I support the idea.
> > > >>>>>>>>>> But, we should carefully work with the release scope.
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13477 - fix for a SQL tracing that will be available
> > > >>> only in
> > > >>>>>>> 2.10
> > > >>>>>>>>>> IGNITE-13427 - fix for a new system view that exists in
> > > >> master
> > > >>>>> only,
> > > >>>>>> we
> > > >>>>>>>>>> need to include IGNITE-13409
> > > >>>>>>>>>>
> > > >>>>>>>>>> Other tickets should be checked, also.
> > > >>>>>>>>>> Is this a real fix of a released bug or we fix some issue we
> > > >>> bring
> > > >>>>>> with
> > > >>>>>>>>>> the brand new contribution.
> > > >>>>>>>>>>
> > > >>>>>>>>>>
> > > >>>>>>>>>> I propose to include the following tickets, also:
> > > >>>>>>>>>>
> > > >>>>>>>>>> CMD tools improvements:
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13488 - Command to print metric value
> > > >>>>>>>>>> IGNITE-13426 - Command to print system view content
> > > >>>>>>>>>> IGNITE-13422 - Parameter to explicitly enable experimental
> > > >>> commands
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13380 - Output IgniteSystemProperties via ignite.sh
> > > >>>>>>>>>>
> > > >>>>>>>>>> New system views:
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13409 Metastorage and DistributedMetastorage viewы.
> > > >>>>>>>>>> IGNITE-13408 BinaryMetadata view.
> > > >>>>>>>>>>
> > > >>>>>>>>>>
> > > >>>>>>>>>>> 19 окт. 2020 г., в 18:20, Yaroslav Molochkov <
> > > >>>>> molochkovyn@gmail.com
> > > >>>>>>>
> > > >>>>>>>>>> написал(а):
> > > >>>>>>>>>>> Hello, Igniters!
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> I've compiled a list of tickets that, i think, deserve to
> be
> > > >>>>>> released
> > > >>>>>>> in
> > > >>>>>>>>>> a
> > > >>>>>>>>>>> minor Ignite release (meaning in 2.9.1) after 2.9. Here
> they
> > > >>> are:
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13569
> > > >>>>>>>>>>> disable archiving + walCompactionEnabled probably broke
> > > >>> reading
> > > >>>>> from
> > > >>>>>>> wal
> > > >>>>>>>>>> on
> > > >>>>>>>>>>> server restart
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13418
> > > >>>>>>>>>>> Deadlock on multiple cache delete
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13563
> > > >>>>>>>>>>> Deserializing IBinaryObject containing an IBinaryObject
> > > >> field
> > > >>>>> fails
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13575
> > > >>>>>>>>>>> Invalid blocking section in GridNioWorker and
> > > >>> GridNioClientWorker
> > > >>>>>>> leads
> > > >>>>>>>>>> to
> > > >>>>>>>>>>> false positive blocking thread detection
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13458
> > > >>>>>>>>>>> RebalancingPartitionsTotal metrics
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13536
> > > >>>>>>>>>>> .NET: Child processes become zombies when persistence is
> > > >> used
> > > >>> with
> > > >>>>>>>>>>> direct-io on Linux
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13500
> > > >>>>>>>>>>> Checkpoint read lock fail if it is taking under write lock
> > > >>> during
> > > >>>>>> the
> > > >>>>>>>>>>> stopping node
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13431
> > > >>>>>>>>>>> NPE during Cassandra Store initialization with PRIMITIVE
> > > >>> strategy
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13417
> > > >>>>>>>>>>> Cache Interceptors deserialization on client nodes
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13495
> > > >>>>>>>>>>> ZookeeperDiscoverySpiMBeanImpl#getCoordinator can return
> > > >>> invalid
> > > >>>>>> node
> > > >>>>>>> as
> > > >>>>>>>>>>> coordinator
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13479
> > > >>>>>>>>>>> Both ignite.sh and control.sh use the same JVM_OPTS.
> > > >>> Control.sh
> > > >>>>>>> doesn't
> > > >>>>>>>>>>> start if JMX port was set
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-11312
> > > >>>>>>>>>>> JDBC: Thin driver reports incorrect property names
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13462
> > > >>>>>>>>>>> .NET: Thin client Dispose hangs when continuous query is
> > > >>> active on
> > > >>>>>>> .NET
> > > >>>>>>>>>>> Core 3.x
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13484
> > > >>>>>>>>>>> C++ odbc-example losing some values if run with 1
> additional
> > > >>> node
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13477
> > > >>>>>>>>>>> Fix NPE in SQL tracing implementation.
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13435
> > > >>>>>>>>>>> Fixing some unrecorded issues command warm-up control.sh
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13403
> > > >>>>>>>>>>> Update JDBC metadata to match actual capabilities
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13427
> > > >>>>>>>>>>> The local metastorage system view fails if unmarshallable
> > > >>> values
> > > >>>>>>> present
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13401
> > > >>>>>>>>>>> Unsupported protocol version exception when getting cache
> > > >>>>>>> configuration
> > > >>>>>>>>>>> from Java thin client
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13388
> > > >>>>>>>>>>> apache-ignite deb package depends on a non-existent package
> > > >>> and
> > > >>>>>> can't
> > > >>>>>>> be
> > > >>>>>>>>>>> installed on Debian 10
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13397
> > > >>>>>>>>>>> NPE in logSupplierDone(UUID nodeId)
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13296
> > > >>>>>>>>>>> .NET: TransactionImpl finalizer can crash the process
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13382
> > > >>>>>>>>>>> DurableBackgroundTask can abandon incomplete task
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-12509
> > > >>>>>>>>>>> CACHE_REBALANCE_STOPPED event raises for wrong caches in
> > > >> case
> > > >>> of
> > > >>>>>>>>>> specified
> > > >>>>>>>>>>> RebalanceDelay
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13072
> > > >>>>>>>>>>> Synchronization problems when different classloaders are
> > > >> used
> > > >>> for
> > > >>>>>>>>>>> deployment of same class
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13379
> > > >>>>>>>>>>> Exception occur on SQL caches when client reconnect
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13363
> > > >>>>>>>>>>> GridDhtCacheEntry::toString locks
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13373
> > > >>>>>>>>>>> WAL segmentns do not released on
> > > >> releaseHistoryForPreloading()
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13540
> > > >>>>>>>>>>> Exchange worker, waiting for new task from queue,
> considered
> > > >>> as
> > > >>>>>>> blocked.
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13491
> > > >>>>>>>>>>> Fix incorrect topology snapshot logger output about
> > > >>> coordinator
> > > >>>>>>> change.
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13439
> > > >>>>>>>>>>> Printing detailed classpath slowdowns node initialization
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> WDYT?
> > > >>>>>>>>>>
> > > >>>>>>>
> > > >>>>>>>
> > > >>>>>>
> > > >>>>>
> > > >>>
> > > >>
> > >
> > >
>
Re: 2.9.1 release proposal
Posted by Maxim Muzafarov <mm...@apache.org>.
Folks, Alexey,
Sure, I forgot about the Christmas holidays. I don't think we will be
able to complete our release procedures with-in the next few weeks, so
let's shift the dates a bit late. Let's prepare everything prior to
the holidays and set the TC.Bot for the stabilization period on
holidays to catch all the flaky issues.
Code Freeze: December 25th, 2020
Release Date: January 18th, 2021
WDYT?
On Thu, 5 Nov 2020 at 13:42, Alexey Goncharuk
<al...@gmail.com> wrote:
>
> Maxim,
>
> Should we shift the dates so they are not too close to state holidays in
> various countries? I'm concerned we won't be getting much activity around
> holidays and people who would otherwise spend some time on testing the
> release will not be able to do so.
>
> чт, 5 нояб. 2020 г. в 12:21, Maxim Muzafarov <mm...@apache.org>:
>
> > Folks,
> >
> > What may be the comfortable dates of the 2.10 release to finish all
> > the related development activities?
> > I suggest release Apache Ignite 2.10 prior to the end of this year and
> > focus on the 3.0 activities in 2021 with only bug-fix releases for the
> > 2.10.x.
> >
> > I suggest the following release dates for the 2.10:
> >
> > Code Freeze: December 16th, 2020
> > Release Date: December 30th, 2020
> >
> > On Tue, 3 Nov 2020 at 18:15, Yaroslav Molochkov <mo...@gmail.com>
> > wrote:
> > >
> > > Nikolay, hi!
> > >
> > > Done.
> > >
> > > On Tue, Nov 3, 2020 at 10:48 AM Nikolay Izhikov <ni...@apache.org>
> > wrote:
> > >
> > > > Hello, Yaroslav.
> > > >
> > > > Looks like we have fixVersion 2.9.1 in Jira.
> > > > Let’s mark all tickets with label «2.9.1-rc1» with fixVersion=2.9.1 and
> > > > use fixVersion instead of label further.
> > > >
> > > >
> > > >
> > https://issues.apache.org/jira/issues/?jql=project%20%3D%20IGNITE%20AND%20fixVersion%20%3D%202.9.1
> > > >
> > > > > 2 нояб. 2020 г., в 19:42, Yaroslav Molochkov <mo...@gmail.com>
> > > > написал(а):
> > > > >
> > > > > Guys,
> > > > >
> > > > > should you agree that issues with the 2.9.1-rc tag are good enough
> > > > > for a maintenance release, i'd like to give it a go.
> > > > >
> > > > > On Thu, Oct 29, 2020 at 5:33 PM Alexey Zinoviev <
> > zaleslaw.sin@gmail.com>
> > > > > wrote:
> > > > >
> > > > >> Let's discuss the possible planning dates for feature freeze for
> > 2.10,
> > > > for
> > > > >> example? Do you have any plans or ideas?
> > > > >>
> > > > >> чт, 29 окт. 2020 г. в 17:24, Andrey Gura <ag...@apache.org>:
> > > > >>
> > > > >>> Hi,
> > > > >>>
> > > > >>> I agree with Alexey. We should release 2.9.1 and start preparing
> > for
> > > > >>> the 2.10 release.
> > > > >>>
> > > > >>> 2.x releases usually take a lot of time. So we can release 2.9.1,
> > and
> > > > >>> even 2.9.2 before 2.10.
> > > > >>>
> > > > >>> On Thu, Oct 29, 2020 at 12:02 PM Alexey Goncharuk
> > > > >>> <al...@gmail.com> wrote:
> > > > >>>>
> > > > >>>> Hello folks,
> > > > >>>>
> > > > >>>> I think we should start both 2.9.1 and 2.10. In practice,
> > maintenance
> > > > >>>> release contains only critical and usability bugfixes (for
> > example, I
> > > > >>> would
> > > > >>>> include this ticket [1] to include in 2.9.1 as it prevents users
> > from
> > > > >>> using
> > > > >>>> tracing) and is released much faster than a minor release.
> > > > >>>>
> > > > >>>> [1] https://issues.apache.org/jira/browse/IGNITE-13640
> > > > >>>>
> > > > >>>> вт, 27 окт. 2020 г. в 21:41, Guru Stron <
> > gurustronpublic@gmail.com>:
> > > > >>>>
> > > > >>>>> Hello,
> > > > >>>>>
> > > > >>>>> Agree with Pavel.
> > > > >>>>>
> > > > >>>>> On Tue, 27 Oct 2020 at 19:22, Pavel Tupitsyn <
> > ptupitsyn@apache.org>
> > > > >>> wrote:
> > > > >>>>>
> > > > >>>>>> Igniters,
> > > > >>>>>>
> > > > >>>>>> I think we should plan 2.10 instead of 2.9.1.
> > > > >>>>>> ignite-2.9 branch was cut 4 months ago, a bunch of new features
> > are
> > > > >>>>> waiting
> > > > >>>>>> to be released.
> > > > >>>>>>
> > > > >>>>>> On Tue, Oct 27, 2020 at 4:23 AM 18624049226 <
> > 18624049226@163.com>
> > > > >>> wrote:
> > > > >>>>>>
> > > > >>>>>>> Hello,
> > > > >>>>>>>
> > > > >>>>>>> I suggest that the remaining document issue in version 2.9.0
> > can
> > > > >> be
> > > > >>>>>>> solved in version 2.9.1, and it is not a good practice to
> > > > >> postpone
> > > > >>> to
> > > > >>>>>>> version 2.10.
> > > > >>>>>>>
> > > > >>>>>>> 在 2020/10/27 上午2:00, Maxim Muzafarov 写道:
> > > > >>>>>>>> Hello,
> > > > >>>>>>>>
> > > > >>>>>>>> +1
> > > > >>>>>>>> Should we start the discussion about the release leader and
> > > > >>> release
> > > > >>>>>>> dates?
> > > > >>>>>>>>
> > > > >>>>>>>> On Tue, 20 Oct 2020 at 10:12, Anton Vinogradov <av@apache.org
> > >
> > > > >>>>> wrote:
> > > > >>>>>>>>> +1 to start the 2.9.1 release process once as soon as 2.9
> > > > >>> released.
> > > > >>>>>>>>>
> > > > >>>>>>>>> On Mon, Oct 19, 2020 at 8:49 PM Nikolay Izhikov <
> > > > >>>>> nizhikov@apache.org>
> > > > >>>>>>> wrote:
> > > > >>>>>>>>>
> > > > >>>>>>>>>> Hello, Yaroslav.
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> I support the idea.
> > > > >>>>>>>>>> But, we should carefully work with the release scope.
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> IGNITE-13477 - fix for a SQL tracing that will be available
> > > > >>> only in
> > > > >>>>>>> 2.10
> > > > >>>>>>>>>> IGNITE-13427 - fix for a new system view that exists in
> > > > >> master
> > > > >>>>> only,
> > > > >>>>>> we
> > > > >>>>>>>>>> need to include IGNITE-13409
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> Other tickets should be checked, also.
> > > > >>>>>>>>>> Is this a real fix of a released bug or we fix some issue we
> > > > >>> bring
> > > > >>>>>> with
> > > > >>>>>>>>>> the brand new contribution.
> > > > >>>>>>>>>>
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> I propose to include the following tickets, also:
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> CMD tools improvements:
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> IGNITE-13488 - Command to print metric value
> > > > >>>>>>>>>> IGNITE-13426 - Command to print system view content
> > > > >>>>>>>>>> IGNITE-13422 - Parameter to explicitly enable experimental
> > > > >>> commands
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> IGNITE-13380 - Output IgniteSystemProperties via ignite.sh
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> New system views:
> > > > >>>>>>>>>>
> > > > >>>>>>>>>> IGNITE-13409 Metastorage and DistributedMetastorage viewы.
> > > > >>>>>>>>>> IGNITE-13408 BinaryMetadata view.
> > > > >>>>>>>>>>
> > > > >>>>>>>>>>
> > > > >>>>>>>>>>> 19 окт. 2020 г., в 18:20, Yaroslav Molochkov <
> > > > >>>>> molochkovyn@gmail.com
> > > > >>>>>>>
> > > > >>>>>>>>>> написал(а):
> > > > >>>>>>>>>>> Hello, Igniters!
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> I've compiled a list of tickets that, i think, deserve to
> > be
> > > > >>>>>> released
> > > > >>>>>>> in
> > > > >>>>>>>>>> a
> > > > >>>>>>>>>>> minor Ignite release (meaning in 2.9.1) after 2.9. Here
> > they
> > > > >>> are:
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13569
> > > > >>>>>>>>>>> disable archiving + walCompactionEnabled probably broke
> > > > >>> reading
> > > > >>>>> from
> > > > >>>>>>> wal
> > > > >>>>>>>>>> on
> > > > >>>>>>>>>>> server restart
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13418
> > > > >>>>>>>>>>> Deadlock on multiple cache delete
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13563
> > > > >>>>>>>>>>> Deserializing IBinaryObject containing an IBinaryObject
> > > > >> field
> > > > >>>>> fails
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13575
> > > > >>>>>>>>>>> Invalid blocking section in GridNioWorker and
> > > > >>> GridNioClientWorker
> > > > >>>>>>> leads
> > > > >>>>>>>>>> to
> > > > >>>>>>>>>>> false positive blocking thread detection
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13458
> > > > >>>>>>>>>>> RebalancingPartitionsTotal metrics
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13536
> > > > >>>>>>>>>>> .NET: Child processes become zombies when persistence is
> > > > >> used
> > > > >>> with
> > > > >>>>>>>>>>> direct-io on Linux
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13500
> > > > >>>>>>>>>>> Checkpoint read lock fail if it is taking under write lock
> > > > >>> during
> > > > >>>>>> the
> > > > >>>>>>>>>>> stopping node
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13431
> > > > >>>>>>>>>>> NPE during Cassandra Store initialization with PRIMITIVE
> > > > >>> strategy
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13417
> > > > >>>>>>>>>>> Cache Interceptors deserialization on client nodes
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13495
> > > > >>>>>>>>>>> ZookeeperDiscoverySpiMBeanImpl#getCoordinator can return
> > > > >>> invalid
> > > > >>>>>> node
> > > > >>>>>>> as
> > > > >>>>>>>>>>> coordinator
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13479
> > > > >>>>>>>>>>> Both ignite.sh and control.sh use the same JVM_OPTS.
> > > > >>> Control.sh
> > > > >>>>>>> doesn't
> > > > >>>>>>>>>>> start if JMX port was set
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-11312
> > > > >>>>>>>>>>> JDBC: Thin driver reports incorrect property names
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13462
> > > > >>>>>>>>>>> .NET: Thin client Dispose hangs when continuous query is
> > > > >>> active on
> > > > >>>>>>> .NET
> > > > >>>>>>>>>>> Core 3.x
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13484
> > > > >>>>>>>>>>> C++ odbc-example losing some values if run with 1
> > additional
> > > > >>> node
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13477
> > > > >>>>>>>>>>> Fix NPE in SQL tracing implementation.
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13435
> > > > >>>>>>>>>>> Fixing some unrecorded issues command warm-up control.sh
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13403
> > > > >>>>>>>>>>> Update JDBC metadata to match actual capabilities
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13427
> > > > >>>>>>>>>>> The local metastorage system view fails if unmarshallable
> > > > >>> values
> > > > >>>>>>> present
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13401
> > > > >>>>>>>>>>> Unsupported protocol version exception when getting cache
> > > > >>>>>>> configuration
> > > > >>>>>>>>>>> from Java thin client
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13388
> > > > >>>>>>>>>>> apache-ignite deb package depends on a non-existent package
> > > > >>> and
> > > > >>>>>> can't
> > > > >>>>>>> be
> > > > >>>>>>>>>>> installed on Debian 10
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13397
> > > > >>>>>>>>>>> NPE in logSupplierDone(UUID nodeId)
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13296
> > > > >>>>>>>>>>> .NET: TransactionImpl finalizer can crash the process
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13382
> > > > >>>>>>>>>>> DurableBackgroundTask can abandon incomplete task
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-12509
> > > > >>>>>>>>>>> CACHE_REBALANCE_STOPPED event raises for wrong caches in
> > > > >> case
> > > > >>> of
> > > > >>>>>>>>>> specified
> > > > >>>>>>>>>>> RebalanceDelay
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13072
> > > > >>>>>>>>>>> Synchronization problems when different classloaders are
> > > > >> used
> > > > >>> for
> > > > >>>>>>>>>>> deployment of same class
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13379
> > > > >>>>>>>>>>> Exception occur on SQL caches when client reconnect
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13363
> > > > >>>>>>>>>>> GridDhtCacheEntry::toString locks
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13373
> > > > >>>>>>>>>>> WAL segmentns do not released on
> > > > >> releaseHistoryForPreloading()
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13540
> > > > >>>>>>>>>>> Exchange worker, waiting for new task from queue,
> > considered
> > > > >>> as
> > > > >>>>>>> blocked.
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13491
> > > > >>>>>>>>>>> Fix incorrect topology snapshot logger output about
> > > > >>> coordinator
> > > > >>>>>>> change.
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> IGNITE-13439
> > > > >>>>>>>>>>> Printing detailed classpath slowdowns node initialization
> > > > >>>>>>>>>>>
> > > > >>>>>>>>>>> WDYT?
> > > > >>>>>>>>>>
> > > > >>>>>>>
> > > > >>>>>>>
> > > > >>>>>>
> > > > >>>>>
> > > > >>>
> > > > >>
> > > >
> > > >
> >
Re: 2.9.1 release proposal
Posted by Alexey Goncharuk <al...@gmail.com>.
Maxim,
Should we shift the dates so they are not too close to state holidays in
various countries? I'm concerned we won't be getting much activity around
holidays and people who would otherwise spend some time on testing the
release will not be able to do so.
чт, 5 нояб. 2020 г. в 12:21, Maxim Muzafarov <mm...@apache.org>:
> Folks,
>
> What may be the comfortable dates of the 2.10 release to finish all
> the related development activities?
> I suggest release Apache Ignite 2.10 prior to the end of this year and
> focus on the 3.0 activities in 2021 with only bug-fix releases for the
> 2.10.x.
>
> I suggest the following release dates for the 2.10:
>
> Code Freeze: December 16th, 2020
> Release Date: December 30th, 2020
>
> On Tue, 3 Nov 2020 at 18:15, Yaroslav Molochkov <mo...@gmail.com>
> wrote:
> >
> > Nikolay, hi!
> >
> > Done.
> >
> > On Tue, Nov 3, 2020 at 10:48 AM Nikolay Izhikov <ni...@apache.org>
> wrote:
> >
> > > Hello, Yaroslav.
> > >
> > > Looks like we have fixVersion 2.9.1 in Jira.
> > > Let’s mark all tickets with label «2.9.1-rc1» with fixVersion=2.9.1 and
> > > use fixVersion instead of label further.
> > >
> > >
> > >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20IGNITE%20AND%20fixVersion%20%3D%202.9.1
> > >
> > > > 2 нояб. 2020 г., в 19:42, Yaroslav Molochkov <mo...@gmail.com>
> > > написал(а):
> > > >
> > > > Guys,
> > > >
> > > > should you agree that issues with the 2.9.1-rc tag are good enough
> > > > for a maintenance release, i'd like to give it a go.
> > > >
> > > > On Thu, Oct 29, 2020 at 5:33 PM Alexey Zinoviev <
> zaleslaw.sin@gmail.com>
> > > > wrote:
> > > >
> > > >> Let's discuss the possible planning dates for feature freeze for
> 2.10,
> > > for
> > > >> example? Do you have any plans or ideas?
> > > >>
> > > >> чт, 29 окт. 2020 г. в 17:24, Andrey Gura <ag...@apache.org>:
> > > >>
> > > >>> Hi,
> > > >>>
> > > >>> I agree with Alexey. We should release 2.9.1 and start preparing
> for
> > > >>> the 2.10 release.
> > > >>>
> > > >>> 2.x releases usually take a lot of time. So we can release 2.9.1,
> and
> > > >>> even 2.9.2 before 2.10.
> > > >>>
> > > >>> On Thu, Oct 29, 2020 at 12:02 PM Alexey Goncharuk
> > > >>> <al...@gmail.com> wrote:
> > > >>>>
> > > >>>> Hello folks,
> > > >>>>
> > > >>>> I think we should start both 2.9.1 and 2.10. In practice,
> maintenance
> > > >>>> release contains only critical and usability bugfixes (for
> example, I
> > > >>> would
> > > >>>> include this ticket [1] to include in 2.9.1 as it prevents users
> from
> > > >>> using
> > > >>>> tracing) and is released much faster than a minor release.
> > > >>>>
> > > >>>> [1] https://issues.apache.org/jira/browse/IGNITE-13640
> > > >>>>
> > > >>>> вт, 27 окт. 2020 г. в 21:41, Guru Stron <
> gurustronpublic@gmail.com>:
> > > >>>>
> > > >>>>> Hello,
> > > >>>>>
> > > >>>>> Agree with Pavel.
> > > >>>>>
> > > >>>>> On Tue, 27 Oct 2020 at 19:22, Pavel Tupitsyn <
> ptupitsyn@apache.org>
> > > >>> wrote:
> > > >>>>>
> > > >>>>>> Igniters,
> > > >>>>>>
> > > >>>>>> I think we should plan 2.10 instead of 2.9.1.
> > > >>>>>> ignite-2.9 branch was cut 4 months ago, a bunch of new features
> are
> > > >>>>> waiting
> > > >>>>>> to be released.
> > > >>>>>>
> > > >>>>>> On Tue, Oct 27, 2020 at 4:23 AM 18624049226 <
> 18624049226@163.com>
> > > >>> wrote:
> > > >>>>>>
> > > >>>>>>> Hello,
> > > >>>>>>>
> > > >>>>>>> I suggest that the remaining document issue in version 2.9.0
> can
> > > >> be
> > > >>>>>>> solved in version 2.9.1, and it is not a good practice to
> > > >> postpone
> > > >>> to
> > > >>>>>>> version 2.10.
> > > >>>>>>>
> > > >>>>>>> 在 2020/10/27 上午2:00, Maxim Muzafarov 写道:
> > > >>>>>>>> Hello,
> > > >>>>>>>>
> > > >>>>>>>> +1
> > > >>>>>>>> Should we start the discussion about the release leader and
> > > >>> release
> > > >>>>>>> dates?
> > > >>>>>>>>
> > > >>>>>>>> On Tue, 20 Oct 2020 at 10:12, Anton Vinogradov <av@apache.org
> >
> > > >>>>> wrote:
> > > >>>>>>>>> +1 to start the 2.9.1 release process once as soon as 2.9
> > > >>> released.
> > > >>>>>>>>>
> > > >>>>>>>>> On Mon, Oct 19, 2020 at 8:49 PM Nikolay Izhikov <
> > > >>>>> nizhikov@apache.org>
> > > >>>>>>> wrote:
> > > >>>>>>>>>
> > > >>>>>>>>>> Hello, Yaroslav.
> > > >>>>>>>>>>
> > > >>>>>>>>>> I support the idea.
> > > >>>>>>>>>> But, we should carefully work with the release scope.
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13477 - fix for a SQL tracing that will be available
> > > >>> only in
> > > >>>>>>> 2.10
> > > >>>>>>>>>> IGNITE-13427 - fix for a new system view that exists in
> > > >> master
> > > >>>>> only,
> > > >>>>>> we
> > > >>>>>>>>>> need to include IGNITE-13409
> > > >>>>>>>>>>
> > > >>>>>>>>>> Other tickets should be checked, also.
> > > >>>>>>>>>> Is this a real fix of a released bug or we fix some issue we
> > > >>> bring
> > > >>>>>> with
> > > >>>>>>>>>> the brand new contribution.
> > > >>>>>>>>>>
> > > >>>>>>>>>>
> > > >>>>>>>>>> I propose to include the following tickets, also:
> > > >>>>>>>>>>
> > > >>>>>>>>>> CMD tools improvements:
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13488 - Command to print metric value
> > > >>>>>>>>>> IGNITE-13426 - Command to print system view content
> > > >>>>>>>>>> IGNITE-13422 - Parameter to explicitly enable experimental
> > > >>> commands
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13380 - Output IgniteSystemProperties via ignite.sh
> > > >>>>>>>>>>
> > > >>>>>>>>>> New system views:
> > > >>>>>>>>>>
> > > >>>>>>>>>> IGNITE-13409 Metastorage and DistributedMetastorage viewы.
> > > >>>>>>>>>> IGNITE-13408 BinaryMetadata view.
> > > >>>>>>>>>>
> > > >>>>>>>>>>
> > > >>>>>>>>>>> 19 окт. 2020 г., в 18:20, Yaroslav Molochkov <
> > > >>>>> molochkovyn@gmail.com
> > > >>>>>>>
> > > >>>>>>>>>> написал(а):
> > > >>>>>>>>>>> Hello, Igniters!
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> I've compiled a list of tickets that, i think, deserve to
> be
> > > >>>>>> released
> > > >>>>>>> in
> > > >>>>>>>>>> a
> > > >>>>>>>>>>> minor Ignite release (meaning in 2.9.1) after 2.9. Here
> they
> > > >>> are:
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13569
> > > >>>>>>>>>>> disable archiving + walCompactionEnabled probably broke
> > > >>> reading
> > > >>>>> from
> > > >>>>>>> wal
> > > >>>>>>>>>> on
> > > >>>>>>>>>>> server restart
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13418
> > > >>>>>>>>>>> Deadlock on multiple cache delete
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13563
> > > >>>>>>>>>>> Deserializing IBinaryObject containing an IBinaryObject
> > > >> field
> > > >>>>> fails
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13575
> > > >>>>>>>>>>> Invalid blocking section in GridNioWorker and
> > > >>> GridNioClientWorker
> > > >>>>>>> leads
> > > >>>>>>>>>> to
> > > >>>>>>>>>>> false positive blocking thread detection
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13458
> > > >>>>>>>>>>> RebalancingPartitionsTotal metrics
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13536
> > > >>>>>>>>>>> .NET: Child processes become zombies when persistence is
> > > >> used
> > > >>> with
> > > >>>>>>>>>>> direct-io on Linux
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13500
> > > >>>>>>>>>>> Checkpoint read lock fail if it is taking under write lock
> > > >>> during
> > > >>>>>> the
> > > >>>>>>>>>>> stopping node
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13431
> > > >>>>>>>>>>> NPE during Cassandra Store initialization with PRIMITIVE
> > > >>> strategy
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13417
> > > >>>>>>>>>>> Cache Interceptors deserialization on client nodes
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13495
> > > >>>>>>>>>>> ZookeeperDiscoverySpiMBeanImpl#getCoordinator can return
> > > >>> invalid
> > > >>>>>> node
> > > >>>>>>> as
> > > >>>>>>>>>>> coordinator
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13479
> > > >>>>>>>>>>> Both ignite.sh and control.sh use the same JVM_OPTS.
> > > >>> Control.sh
> > > >>>>>>> doesn't
> > > >>>>>>>>>>> start if JMX port was set
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-11312
> > > >>>>>>>>>>> JDBC: Thin driver reports incorrect property names
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13462
> > > >>>>>>>>>>> .NET: Thin client Dispose hangs when continuous query is
> > > >>> active on
> > > >>>>>>> .NET
> > > >>>>>>>>>>> Core 3.x
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13484
> > > >>>>>>>>>>> C++ odbc-example losing some values if run with 1
> additional
> > > >>> node
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13477
> > > >>>>>>>>>>> Fix NPE in SQL tracing implementation.
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13435
> > > >>>>>>>>>>> Fixing some unrecorded issues command warm-up control.sh
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13403
> > > >>>>>>>>>>> Update JDBC metadata to match actual capabilities
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13427
> > > >>>>>>>>>>> The local metastorage system view fails if unmarshallable
> > > >>> values
> > > >>>>>>> present
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13401
> > > >>>>>>>>>>> Unsupported protocol version exception when getting cache
> > > >>>>>>> configuration
> > > >>>>>>>>>>> from Java thin client
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13388
> > > >>>>>>>>>>> apache-ignite deb package depends on a non-existent package
> > > >>> and
> > > >>>>>> can't
> > > >>>>>>> be
> > > >>>>>>>>>>> installed on Debian 10
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13397
> > > >>>>>>>>>>> NPE in logSupplierDone(UUID nodeId)
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13296
> > > >>>>>>>>>>> .NET: TransactionImpl finalizer can crash the process
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13382
> > > >>>>>>>>>>> DurableBackgroundTask can abandon incomplete task
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-12509
> > > >>>>>>>>>>> CACHE_REBALANCE_STOPPED event raises for wrong caches in
> > > >> case
> > > >>> of
> > > >>>>>>>>>> specified
> > > >>>>>>>>>>> RebalanceDelay
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13072
> > > >>>>>>>>>>> Synchronization problems when different classloaders are
> > > >> used
> > > >>> for
> > > >>>>>>>>>>> deployment of same class
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13379
> > > >>>>>>>>>>> Exception occur on SQL caches when client reconnect
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13363
> > > >>>>>>>>>>> GridDhtCacheEntry::toString locks
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13373
> > > >>>>>>>>>>> WAL segmentns do not released on
> > > >> releaseHistoryForPreloading()
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13540
> > > >>>>>>>>>>> Exchange worker, waiting for new task from queue,
> considered
> > > >>> as
> > > >>>>>>> blocked.
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13491
> > > >>>>>>>>>>> Fix incorrect topology snapshot logger output about
> > > >>> coordinator
> > > >>>>>>> change.
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> IGNITE-13439
> > > >>>>>>>>>>> Printing detailed classpath slowdowns node initialization
> > > >>>>>>>>>>>
> > > >>>>>>>>>>> WDYT?
> > > >>>>>>>>>>
> > > >>>>>>>
> > > >>>>>>>
> > > >>>>>>
> > > >>>>>
> > > >>>
> > > >>
> > >
> > >
>
Re: 2.9.1 release proposal
Posted by Maxim Muzafarov <mm...@apache.org>.
Folks,
What may be the comfortable dates of the 2.10 release to finish all
the related development activities?
I suggest release Apache Ignite 2.10 prior to the end of this year and
focus on the 3.0 activities in 2021 with only bug-fix releases for the
2.10.x.
I suggest the following release dates for the 2.10:
Code Freeze: December 16th, 2020
Release Date: December 30th, 2020
On Tue, 3 Nov 2020 at 18:15, Yaroslav Molochkov <mo...@gmail.com> wrote:
>
> Nikolay, hi!
>
> Done.
>
> On Tue, Nov 3, 2020 at 10:48 AM Nikolay Izhikov <ni...@apache.org> wrote:
>
> > Hello, Yaroslav.
> >
> > Looks like we have fixVersion 2.9.1 in Jira.
> > Let’s mark all tickets with label «2.9.1-rc1» with fixVersion=2.9.1 and
> > use fixVersion instead of label further.
> >
> >
> > https://issues.apache.org/jira/issues/?jql=project%20%3D%20IGNITE%20AND%20fixVersion%20%3D%202.9.1
> >
> > > 2 нояб. 2020 г., в 19:42, Yaroslav Molochkov <mo...@gmail.com>
> > написал(а):
> > >
> > > Guys,
> > >
> > > should you agree that issues with the 2.9.1-rc tag are good enough
> > > for a maintenance release, i'd like to give it a go.
> > >
> > > On Thu, Oct 29, 2020 at 5:33 PM Alexey Zinoviev <za...@gmail.com>
> > > wrote:
> > >
> > >> Let's discuss the possible planning dates for feature freeze for 2.10,
> > for
> > >> example? Do you have any plans or ideas?
> > >>
> > >> чт, 29 окт. 2020 г. в 17:24, Andrey Gura <ag...@apache.org>:
> > >>
> > >>> Hi,
> > >>>
> > >>> I agree with Alexey. We should release 2.9.1 and start preparing for
> > >>> the 2.10 release.
> > >>>
> > >>> 2.x releases usually take a lot of time. So we can release 2.9.1, and
> > >>> even 2.9.2 before 2.10.
> > >>>
> > >>> On Thu, Oct 29, 2020 at 12:02 PM Alexey Goncharuk
> > >>> <al...@gmail.com> wrote:
> > >>>>
> > >>>> Hello folks,
> > >>>>
> > >>>> I think we should start both 2.9.1 and 2.10. In practice, maintenance
> > >>>> release contains only critical and usability bugfixes (for example, I
> > >>> would
> > >>>> include this ticket [1] to include in 2.9.1 as it prevents users from
> > >>> using
> > >>>> tracing) and is released much faster than a minor release.
> > >>>>
> > >>>> [1] https://issues.apache.org/jira/browse/IGNITE-13640
> > >>>>
> > >>>> вт, 27 окт. 2020 г. в 21:41, Guru Stron <gu...@gmail.com>:
> > >>>>
> > >>>>> Hello,
> > >>>>>
> > >>>>> Agree with Pavel.
> > >>>>>
> > >>>>> On Tue, 27 Oct 2020 at 19:22, Pavel Tupitsyn <pt...@apache.org>
> > >>> wrote:
> > >>>>>
> > >>>>>> Igniters,
> > >>>>>>
> > >>>>>> I think we should plan 2.10 instead of 2.9.1.
> > >>>>>> ignite-2.9 branch was cut 4 months ago, a bunch of new features are
> > >>>>> waiting
> > >>>>>> to be released.
> > >>>>>>
> > >>>>>> On Tue, Oct 27, 2020 at 4:23 AM 18624049226 <18...@163.com>
> > >>> wrote:
> > >>>>>>
> > >>>>>>> Hello,
> > >>>>>>>
> > >>>>>>> I suggest that the remaining document issue in version 2.9.0 can
> > >> be
> > >>>>>>> solved in version 2.9.1, and it is not a good practice to
> > >> postpone
> > >>> to
> > >>>>>>> version 2.10.
> > >>>>>>>
> > >>>>>>> 在 2020/10/27 上午2:00, Maxim Muzafarov 写道:
> > >>>>>>>> Hello,
> > >>>>>>>>
> > >>>>>>>> +1
> > >>>>>>>> Should we start the discussion about the release leader and
> > >>> release
> > >>>>>>> dates?
> > >>>>>>>>
> > >>>>>>>> On Tue, 20 Oct 2020 at 10:12, Anton Vinogradov <av...@apache.org>
> > >>>>> wrote:
> > >>>>>>>>> +1 to start the 2.9.1 release process once as soon as 2.9
> > >>> released.
> > >>>>>>>>>
> > >>>>>>>>> On Mon, Oct 19, 2020 at 8:49 PM Nikolay Izhikov <
> > >>>>> nizhikov@apache.org>
> > >>>>>>> wrote:
> > >>>>>>>>>
> > >>>>>>>>>> Hello, Yaroslav.
> > >>>>>>>>>>
> > >>>>>>>>>> I support the idea.
> > >>>>>>>>>> But, we should carefully work with the release scope.
> > >>>>>>>>>>
> > >>>>>>>>>> IGNITE-13477 - fix for a SQL tracing that will be available
> > >>> only in
> > >>>>>>> 2.10
> > >>>>>>>>>> IGNITE-13427 - fix for a new system view that exists in
> > >> master
> > >>>>> only,
> > >>>>>> we
> > >>>>>>>>>> need to include IGNITE-13409
> > >>>>>>>>>>
> > >>>>>>>>>> Other tickets should be checked, also.
> > >>>>>>>>>> Is this a real fix of a released bug or we fix some issue we
> > >>> bring
> > >>>>>> with
> > >>>>>>>>>> the brand new contribution.
> > >>>>>>>>>>
> > >>>>>>>>>>
> > >>>>>>>>>> I propose to include the following tickets, also:
> > >>>>>>>>>>
> > >>>>>>>>>> CMD tools improvements:
> > >>>>>>>>>>
> > >>>>>>>>>> IGNITE-13488 - Command to print metric value
> > >>>>>>>>>> IGNITE-13426 - Command to print system view content
> > >>>>>>>>>> IGNITE-13422 - Parameter to explicitly enable experimental
> > >>> commands
> > >>>>>>>>>>
> > >>>>>>>>>> IGNITE-13380 - Output IgniteSystemProperties via ignite.sh
> > >>>>>>>>>>
> > >>>>>>>>>> New system views:
> > >>>>>>>>>>
> > >>>>>>>>>> IGNITE-13409 Metastorage and DistributedMetastorage viewы.
> > >>>>>>>>>> IGNITE-13408 BinaryMetadata view.
> > >>>>>>>>>>
> > >>>>>>>>>>
> > >>>>>>>>>>> 19 окт. 2020 г., в 18:20, Yaroslav Molochkov <
> > >>>>> molochkovyn@gmail.com
> > >>>>>>>
> > >>>>>>>>>> написал(а):
> > >>>>>>>>>>> Hello, Igniters!
> > >>>>>>>>>>>
> > >>>>>>>>>>> I've compiled a list of tickets that, i think, deserve to be
> > >>>>>> released
> > >>>>>>> in
> > >>>>>>>>>> a
> > >>>>>>>>>>> minor Ignite release (meaning in 2.9.1) after 2.9. Here they
> > >>> are:
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13569
> > >>>>>>>>>>> disable archiving + walCompactionEnabled probably broke
> > >>> reading
> > >>>>> from
> > >>>>>>> wal
> > >>>>>>>>>> on
> > >>>>>>>>>>> server restart
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13418
> > >>>>>>>>>>> Deadlock on multiple cache delete
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13563
> > >>>>>>>>>>> Deserializing IBinaryObject containing an IBinaryObject
> > >> field
> > >>>>> fails
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13575
> > >>>>>>>>>>> Invalid blocking section in GridNioWorker and
> > >>> GridNioClientWorker
> > >>>>>>> leads
> > >>>>>>>>>> to
> > >>>>>>>>>>> false positive blocking thread detection
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13458
> > >>>>>>>>>>> RebalancingPartitionsTotal metrics
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13536
> > >>>>>>>>>>> .NET: Child processes become zombies when persistence is
> > >> used
> > >>> with
> > >>>>>>>>>>> direct-io on Linux
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13500
> > >>>>>>>>>>> Checkpoint read lock fail if it is taking under write lock
> > >>> during
> > >>>>>> the
> > >>>>>>>>>>> stopping node
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13431
> > >>>>>>>>>>> NPE during Cassandra Store initialization with PRIMITIVE
> > >>> strategy
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13417
> > >>>>>>>>>>> Cache Interceptors deserialization on client nodes
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13495
> > >>>>>>>>>>> ZookeeperDiscoverySpiMBeanImpl#getCoordinator can return
> > >>> invalid
> > >>>>>> node
> > >>>>>>> as
> > >>>>>>>>>>> coordinator
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13479
> > >>>>>>>>>>> Both ignite.sh and control.sh use the same JVM_OPTS.
> > >>> Control.sh
> > >>>>>>> doesn't
> > >>>>>>>>>>> start if JMX port was set
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-11312
> > >>>>>>>>>>> JDBC: Thin driver reports incorrect property names
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13462
> > >>>>>>>>>>> .NET: Thin client Dispose hangs when continuous query is
> > >>> active on
> > >>>>>>> .NET
> > >>>>>>>>>>> Core 3.x
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13484
> > >>>>>>>>>>> C++ odbc-example losing some values if run with 1 additional
> > >>> node
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13477
> > >>>>>>>>>>> Fix NPE in SQL tracing implementation.
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13435
> > >>>>>>>>>>> Fixing some unrecorded issues command warm-up control.sh
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13403
> > >>>>>>>>>>> Update JDBC metadata to match actual capabilities
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13427
> > >>>>>>>>>>> The local metastorage system view fails if unmarshallable
> > >>> values
> > >>>>>>> present
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13401
> > >>>>>>>>>>> Unsupported protocol version exception when getting cache
> > >>>>>>> configuration
> > >>>>>>>>>>> from Java thin client
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13388
> > >>>>>>>>>>> apache-ignite deb package depends on a non-existent package
> > >>> and
> > >>>>>> can't
> > >>>>>>> be
> > >>>>>>>>>>> installed on Debian 10
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13397
> > >>>>>>>>>>> NPE in logSupplierDone(UUID nodeId)
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13296
> > >>>>>>>>>>> .NET: TransactionImpl finalizer can crash the process
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13382
> > >>>>>>>>>>> DurableBackgroundTask can abandon incomplete task
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-12509
> > >>>>>>>>>>> CACHE_REBALANCE_STOPPED event raises for wrong caches in
> > >> case
> > >>> of
> > >>>>>>>>>> specified
> > >>>>>>>>>>> RebalanceDelay
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13072
> > >>>>>>>>>>> Synchronization problems when different classloaders are
> > >> used
> > >>> for
> > >>>>>>>>>>> deployment of same class
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13379
> > >>>>>>>>>>> Exception occur on SQL caches when client reconnect
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13363
> > >>>>>>>>>>> GridDhtCacheEntry::toString locks
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13373
> > >>>>>>>>>>> WAL segmentns do not released on
> > >> releaseHistoryForPreloading()
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13540
> > >>>>>>>>>>> Exchange worker, waiting for new task from queue, considered
> > >>> as
> > >>>>>>> blocked.
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13491
> > >>>>>>>>>>> Fix incorrect topology snapshot logger output about
> > >>> coordinator
> > >>>>>>> change.
> > >>>>>>>>>>>
> > >>>>>>>>>>> IGNITE-13439
> > >>>>>>>>>>> Printing detailed classpath slowdowns node initialization
> > >>>>>>>>>>>
> > >>>>>>>>>>> WDYT?
> > >>>>>>>>>>
> > >>>>>>>
> > >>>>>>>
> > >>>>>>
> > >>>>>
> > >>>
> > >>
> >
> >
Re: 2.9.1 release proposal
Posted by Yaroslav Molochkov <mo...@gmail.com>.
Nikolay, hi!
Done.
On Tue, Nov 3, 2020 at 10:48 AM Nikolay Izhikov <ni...@apache.org> wrote:
> Hello, Yaroslav.
>
> Looks like we have fixVersion 2.9.1 in Jira.
> Let’s mark all tickets with label «2.9.1-rc1» with fixVersion=2.9.1 and
> use fixVersion instead of label further.
>
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20IGNITE%20AND%20fixVersion%20%3D%202.9.1
>
> > 2 нояб. 2020 г., в 19:42, Yaroslav Molochkov <mo...@gmail.com>
> написал(а):
> >
> > Guys,
> >
> > should you agree that issues with the 2.9.1-rc tag are good enough
> > for a maintenance release, i'd like to give it a go.
> >
> > On Thu, Oct 29, 2020 at 5:33 PM Alexey Zinoviev <za...@gmail.com>
> > wrote:
> >
> >> Let's discuss the possible planning dates for feature freeze for 2.10,
> for
> >> example? Do you have any plans or ideas?
> >>
> >> чт, 29 окт. 2020 г. в 17:24, Andrey Gura <ag...@apache.org>:
> >>
> >>> Hi,
> >>>
> >>> I agree with Alexey. We should release 2.9.1 and start preparing for
> >>> the 2.10 release.
> >>>
> >>> 2.x releases usually take a lot of time. So we can release 2.9.1, and
> >>> even 2.9.2 before 2.10.
> >>>
> >>> On Thu, Oct 29, 2020 at 12:02 PM Alexey Goncharuk
> >>> <al...@gmail.com> wrote:
> >>>>
> >>>> Hello folks,
> >>>>
> >>>> I think we should start both 2.9.1 and 2.10. In practice, maintenance
> >>>> release contains only critical and usability bugfixes (for example, I
> >>> would
> >>>> include this ticket [1] to include in 2.9.1 as it prevents users from
> >>> using
> >>>> tracing) and is released much faster than a minor release.
> >>>>
> >>>> [1] https://issues.apache.org/jira/browse/IGNITE-13640
> >>>>
> >>>> вт, 27 окт. 2020 г. в 21:41, Guru Stron <gu...@gmail.com>:
> >>>>
> >>>>> Hello,
> >>>>>
> >>>>> Agree with Pavel.
> >>>>>
> >>>>> On Tue, 27 Oct 2020 at 19:22, Pavel Tupitsyn <pt...@apache.org>
> >>> wrote:
> >>>>>
> >>>>>> Igniters,
> >>>>>>
> >>>>>> I think we should plan 2.10 instead of 2.9.1.
> >>>>>> ignite-2.9 branch was cut 4 months ago, a bunch of new features are
> >>>>> waiting
> >>>>>> to be released.
> >>>>>>
> >>>>>> On Tue, Oct 27, 2020 at 4:23 AM 18624049226 <18...@163.com>
> >>> wrote:
> >>>>>>
> >>>>>>> Hello,
> >>>>>>>
> >>>>>>> I suggest that the remaining document issue in version 2.9.0 can
> >> be
> >>>>>>> solved in version 2.9.1, and it is not a good practice to
> >> postpone
> >>> to
> >>>>>>> version 2.10.
> >>>>>>>
> >>>>>>> 在 2020/10/27 上午2:00, Maxim Muzafarov 写道:
> >>>>>>>> Hello,
> >>>>>>>>
> >>>>>>>> +1
> >>>>>>>> Should we start the discussion about the release leader and
> >>> release
> >>>>>>> dates?
> >>>>>>>>
> >>>>>>>> On Tue, 20 Oct 2020 at 10:12, Anton Vinogradov <av...@apache.org>
> >>>>> wrote:
> >>>>>>>>> +1 to start the 2.9.1 release process once as soon as 2.9
> >>> released.
> >>>>>>>>>
> >>>>>>>>> On Mon, Oct 19, 2020 at 8:49 PM Nikolay Izhikov <
> >>>>> nizhikov@apache.org>
> >>>>>>> wrote:
> >>>>>>>>>
> >>>>>>>>>> Hello, Yaroslav.
> >>>>>>>>>>
> >>>>>>>>>> I support the idea.
> >>>>>>>>>> But, we should carefully work with the release scope.
> >>>>>>>>>>
> >>>>>>>>>> IGNITE-13477 - fix for a SQL tracing that will be available
> >>> only in
> >>>>>>> 2.10
> >>>>>>>>>> IGNITE-13427 - fix for a new system view that exists in
> >> master
> >>>>> only,
> >>>>>> we
> >>>>>>>>>> need to include IGNITE-13409
> >>>>>>>>>>
> >>>>>>>>>> Other tickets should be checked, also.
> >>>>>>>>>> Is this a real fix of a released bug or we fix some issue we
> >>> bring
> >>>>>> with
> >>>>>>>>>> the brand new contribution.
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>> I propose to include the following tickets, also:
> >>>>>>>>>>
> >>>>>>>>>> CMD tools improvements:
> >>>>>>>>>>
> >>>>>>>>>> IGNITE-13488 - Command to print metric value
> >>>>>>>>>> IGNITE-13426 - Command to print system view content
> >>>>>>>>>> IGNITE-13422 - Parameter to explicitly enable experimental
> >>> commands
> >>>>>>>>>>
> >>>>>>>>>> IGNITE-13380 - Output IgniteSystemProperties via ignite.sh
> >>>>>>>>>>
> >>>>>>>>>> New system views:
> >>>>>>>>>>
> >>>>>>>>>> IGNITE-13409 Metastorage and DistributedMetastorage viewы.
> >>>>>>>>>> IGNITE-13408 BinaryMetadata view.
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>>> 19 окт. 2020 г., в 18:20, Yaroslav Molochkov <
> >>>>> molochkovyn@gmail.com
> >>>>>>>
> >>>>>>>>>> написал(а):
> >>>>>>>>>>> Hello, Igniters!
> >>>>>>>>>>>
> >>>>>>>>>>> I've compiled a list of tickets that, i think, deserve to be
> >>>>>> released
> >>>>>>> in
> >>>>>>>>>> a
> >>>>>>>>>>> minor Ignite release (meaning in 2.9.1) after 2.9. Here they
> >>> are:
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13569
> >>>>>>>>>>> disable archiving + walCompactionEnabled probably broke
> >>> reading
> >>>>> from
> >>>>>>> wal
> >>>>>>>>>> on
> >>>>>>>>>>> server restart
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13418
> >>>>>>>>>>> Deadlock on multiple cache delete
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13563
> >>>>>>>>>>> Deserializing IBinaryObject containing an IBinaryObject
> >> field
> >>>>> fails
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13575
> >>>>>>>>>>> Invalid blocking section in GridNioWorker and
> >>> GridNioClientWorker
> >>>>>>> leads
> >>>>>>>>>> to
> >>>>>>>>>>> false positive blocking thread detection
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13458
> >>>>>>>>>>> RebalancingPartitionsTotal metrics
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13536
> >>>>>>>>>>> .NET: Child processes become zombies when persistence is
> >> used
> >>> with
> >>>>>>>>>>> direct-io on Linux
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13500
> >>>>>>>>>>> Checkpoint read lock fail if it is taking under write lock
> >>> during
> >>>>>> the
> >>>>>>>>>>> stopping node
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13431
> >>>>>>>>>>> NPE during Cassandra Store initialization with PRIMITIVE
> >>> strategy
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13417
> >>>>>>>>>>> Cache Interceptors deserialization on client nodes
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13495
> >>>>>>>>>>> ZookeeperDiscoverySpiMBeanImpl#getCoordinator can return
> >>> invalid
> >>>>>> node
> >>>>>>> as
> >>>>>>>>>>> coordinator
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13479
> >>>>>>>>>>> Both ignite.sh and control.sh use the same JVM_OPTS.
> >>> Control.sh
> >>>>>>> doesn't
> >>>>>>>>>>> start if JMX port was set
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-11312
> >>>>>>>>>>> JDBC: Thin driver reports incorrect property names
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13462
> >>>>>>>>>>> .NET: Thin client Dispose hangs when continuous query is
> >>> active on
> >>>>>>> .NET
> >>>>>>>>>>> Core 3.x
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13484
> >>>>>>>>>>> C++ odbc-example losing some values if run with 1 additional
> >>> node
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13477
> >>>>>>>>>>> Fix NPE in SQL tracing implementation.
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13435
> >>>>>>>>>>> Fixing some unrecorded issues command warm-up control.sh
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13403
> >>>>>>>>>>> Update JDBC metadata to match actual capabilities
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13427
> >>>>>>>>>>> The local metastorage system view fails if unmarshallable
> >>> values
> >>>>>>> present
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13401
> >>>>>>>>>>> Unsupported protocol version exception when getting cache
> >>>>>>> configuration
> >>>>>>>>>>> from Java thin client
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13388
> >>>>>>>>>>> apache-ignite deb package depends on a non-existent package
> >>> and
> >>>>>> can't
> >>>>>>> be
> >>>>>>>>>>> installed on Debian 10
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13397
> >>>>>>>>>>> NPE in logSupplierDone(UUID nodeId)
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13296
> >>>>>>>>>>> .NET: TransactionImpl finalizer can crash the process
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13382
> >>>>>>>>>>> DurableBackgroundTask can abandon incomplete task
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-12509
> >>>>>>>>>>> CACHE_REBALANCE_STOPPED event raises for wrong caches in
> >> case
> >>> of
> >>>>>>>>>> specified
> >>>>>>>>>>> RebalanceDelay
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13072
> >>>>>>>>>>> Synchronization problems when different classloaders are
> >> used
> >>> for
> >>>>>>>>>>> deployment of same class
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13379
> >>>>>>>>>>> Exception occur on SQL caches when client reconnect
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13363
> >>>>>>>>>>> GridDhtCacheEntry::toString locks
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13373
> >>>>>>>>>>> WAL segmentns do not released on
> >> releaseHistoryForPreloading()
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13540
> >>>>>>>>>>> Exchange worker, waiting for new task from queue, considered
> >>> as
> >>>>>>> blocked.
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13491
> >>>>>>>>>>> Fix incorrect topology snapshot logger output about
> >>> coordinator
> >>>>>>> change.
> >>>>>>>>>>>
> >>>>>>>>>>> IGNITE-13439
> >>>>>>>>>>> Printing detailed classpath slowdowns node initialization
> >>>>>>>>>>>
> >>>>>>>>>>> WDYT?
> >>>>>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>
> >>>
> >>
>
>
Re: 2.9.1 release proposal
Posted by Nikolay Izhikov <ni...@apache.org>.
Hello, Yaroslav.
Looks like we have fixVersion 2.9.1 in Jira.
Let’s mark all tickets with label «2.9.1-rc1» with fixVersion=2.9.1 and use fixVersion instead of label further.
https://issues.apache.org/jira/issues/?jql=project%20%3D%20IGNITE%20AND%20fixVersion%20%3D%202.9.1
> 2 нояб. 2020 г., в 19:42, Yaroslav Molochkov <mo...@gmail.com> написал(а):
>
> Guys,
>
> should you agree that issues with the 2.9.1-rc tag are good enough
> for a maintenance release, i'd like to give it a go.
>
> On Thu, Oct 29, 2020 at 5:33 PM Alexey Zinoviev <za...@gmail.com>
> wrote:
>
>> Let's discuss the possible planning dates for feature freeze for 2.10, for
>> example? Do you have any plans or ideas?
>>
>> чт, 29 окт. 2020 г. в 17:24, Andrey Gura <ag...@apache.org>:
>>
>>> Hi,
>>>
>>> I agree with Alexey. We should release 2.9.1 and start preparing for
>>> the 2.10 release.
>>>
>>> 2.x releases usually take a lot of time. So we can release 2.9.1, and
>>> even 2.9.2 before 2.10.
>>>
>>> On Thu, Oct 29, 2020 at 12:02 PM Alexey Goncharuk
>>> <al...@gmail.com> wrote:
>>>>
>>>> Hello folks,
>>>>
>>>> I think we should start both 2.9.1 and 2.10. In practice, maintenance
>>>> release contains only critical and usability bugfixes (for example, I
>>> would
>>>> include this ticket [1] to include in 2.9.1 as it prevents users from
>>> using
>>>> tracing) and is released much faster than a minor release.
>>>>
>>>> [1] https://issues.apache.org/jira/browse/IGNITE-13640
>>>>
>>>> вт, 27 окт. 2020 г. в 21:41, Guru Stron <gu...@gmail.com>:
>>>>
>>>>> Hello,
>>>>>
>>>>> Agree with Pavel.
>>>>>
>>>>> On Tue, 27 Oct 2020 at 19:22, Pavel Tupitsyn <pt...@apache.org>
>>> wrote:
>>>>>
>>>>>> Igniters,
>>>>>>
>>>>>> I think we should plan 2.10 instead of 2.9.1.
>>>>>> ignite-2.9 branch was cut 4 months ago, a bunch of new features are
>>>>> waiting
>>>>>> to be released.
>>>>>>
>>>>>> On Tue, Oct 27, 2020 at 4:23 AM 18624049226 <18...@163.com>
>>> wrote:
>>>>>>
>>>>>>> Hello,
>>>>>>>
>>>>>>> I suggest that the remaining document issue in version 2.9.0 can
>> be
>>>>>>> solved in version 2.9.1, and it is not a good practice to
>> postpone
>>> to
>>>>>>> version 2.10.
>>>>>>>
>>>>>>> 在 2020/10/27 上午2:00, Maxim Muzafarov 写道:
>>>>>>>> Hello,
>>>>>>>>
>>>>>>>> +1
>>>>>>>> Should we start the discussion about the release leader and
>>> release
>>>>>>> dates?
>>>>>>>>
>>>>>>>> On Tue, 20 Oct 2020 at 10:12, Anton Vinogradov <av...@apache.org>
>>>>> wrote:
>>>>>>>>> +1 to start the 2.9.1 release process once as soon as 2.9
>>> released.
>>>>>>>>>
>>>>>>>>> On Mon, Oct 19, 2020 at 8:49 PM Nikolay Izhikov <
>>>>> nizhikov@apache.org>
>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> Hello, Yaroslav.
>>>>>>>>>>
>>>>>>>>>> I support the idea.
>>>>>>>>>> But, we should carefully work with the release scope.
>>>>>>>>>>
>>>>>>>>>> IGNITE-13477 - fix for a SQL tracing that will be available
>>> only in
>>>>>>> 2.10
>>>>>>>>>> IGNITE-13427 - fix for a new system view that exists in
>> master
>>>>> only,
>>>>>> we
>>>>>>>>>> need to include IGNITE-13409
>>>>>>>>>>
>>>>>>>>>> Other tickets should be checked, also.
>>>>>>>>>> Is this a real fix of a released bug or we fix some issue we
>>> bring
>>>>>> with
>>>>>>>>>> the brand new contribution.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> I propose to include the following tickets, also:
>>>>>>>>>>
>>>>>>>>>> CMD tools improvements:
>>>>>>>>>>
>>>>>>>>>> IGNITE-13488 - Command to print metric value
>>>>>>>>>> IGNITE-13426 - Command to print system view content
>>>>>>>>>> IGNITE-13422 - Parameter to explicitly enable experimental
>>> commands
>>>>>>>>>>
>>>>>>>>>> IGNITE-13380 - Output IgniteSystemProperties via ignite.sh
>>>>>>>>>>
>>>>>>>>>> New system views:
>>>>>>>>>>
>>>>>>>>>> IGNITE-13409 Metastorage and DistributedMetastorage viewы.
>>>>>>>>>> IGNITE-13408 BinaryMetadata view.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>> 19 окт. 2020 г., в 18:20, Yaroslav Molochkov <
>>>>> molochkovyn@gmail.com
>>>>>>>
>>>>>>>>>> написал(а):
>>>>>>>>>>> Hello, Igniters!
>>>>>>>>>>>
>>>>>>>>>>> I've compiled a list of tickets that, i think, deserve to be
>>>>>> released
>>>>>>> in
>>>>>>>>>> a
>>>>>>>>>>> minor Ignite release (meaning in 2.9.1) after 2.9. Here they
>>> are:
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13569
>>>>>>>>>>> disable archiving + walCompactionEnabled probably broke
>>> reading
>>>>> from
>>>>>>> wal
>>>>>>>>>> on
>>>>>>>>>>> server restart
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13418
>>>>>>>>>>> Deadlock on multiple cache delete
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13563
>>>>>>>>>>> Deserializing IBinaryObject containing an IBinaryObject
>> field
>>>>> fails
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13575
>>>>>>>>>>> Invalid blocking section in GridNioWorker and
>>> GridNioClientWorker
>>>>>>> leads
>>>>>>>>>> to
>>>>>>>>>>> false positive blocking thread detection
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13458
>>>>>>>>>>> RebalancingPartitionsTotal metrics
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13536
>>>>>>>>>>> .NET: Child processes become zombies when persistence is
>> used
>>> with
>>>>>>>>>>> direct-io on Linux
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13500
>>>>>>>>>>> Checkpoint read lock fail if it is taking under write lock
>>> during
>>>>>> the
>>>>>>>>>>> stopping node
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13431
>>>>>>>>>>> NPE during Cassandra Store initialization with PRIMITIVE
>>> strategy
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13417
>>>>>>>>>>> Cache Interceptors deserialization on client nodes
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13495
>>>>>>>>>>> ZookeeperDiscoverySpiMBeanImpl#getCoordinator can return
>>> invalid
>>>>>> node
>>>>>>> as
>>>>>>>>>>> coordinator
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13479
>>>>>>>>>>> Both ignite.sh and control.sh use the same JVM_OPTS.
>>> Control.sh
>>>>>>> doesn't
>>>>>>>>>>> start if JMX port was set
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-11312
>>>>>>>>>>> JDBC: Thin driver reports incorrect property names
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13462
>>>>>>>>>>> .NET: Thin client Dispose hangs when continuous query is
>>> active on
>>>>>>> .NET
>>>>>>>>>>> Core 3.x
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13484
>>>>>>>>>>> C++ odbc-example losing some values if run with 1 additional
>>> node
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13477
>>>>>>>>>>> Fix NPE in SQL tracing implementation.
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13435
>>>>>>>>>>> Fixing some unrecorded issues command warm-up control.sh
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13403
>>>>>>>>>>> Update JDBC metadata to match actual capabilities
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13427
>>>>>>>>>>> The local metastorage system view fails if unmarshallable
>>> values
>>>>>>> present
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13401
>>>>>>>>>>> Unsupported protocol version exception when getting cache
>>>>>>> configuration
>>>>>>>>>>> from Java thin client
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13388
>>>>>>>>>>> apache-ignite deb package depends on a non-existent package
>>> and
>>>>>> can't
>>>>>>> be
>>>>>>>>>>> installed on Debian 10
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13397
>>>>>>>>>>> NPE in logSupplierDone(UUID nodeId)
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13296
>>>>>>>>>>> .NET: TransactionImpl finalizer can crash the process
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13382
>>>>>>>>>>> DurableBackgroundTask can abandon incomplete task
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-12509
>>>>>>>>>>> CACHE_REBALANCE_STOPPED event raises for wrong caches in
>> case
>>> of
>>>>>>>>>> specified
>>>>>>>>>>> RebalanceDelay
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13072
>>>>>>>>>>> Synchronization problems when different classloaders are
>> used
>>> for
>>>>>>>>>>> deployment of same class
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13379
>>>>>>>>>>> Exception occur on SQL caches when client reconnect
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13363
>>>>>>>>>>> GridDhtCacheEntry::toString locks
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13373
>>>>>>>>>>> WAL segmentns do not released on
>> releaseHistoryForPreloading()
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13540
>>>>>>>>>>> Exchange worker, waiting for new task from queue, considered
>>> as
>>>>>>> blocked.
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13491
>>>>>>>>>>> Fix incorrect topology snapshot logger output about
>>> coordinator
>>>>>>> change.
>>>>>>>>>>>
>>>>>>>>>>> IGNITE-13439
>>>>>>>>>>> Printing detailed classpath slowdowns node initialization
>>>>>>>>>>>
>>>>>>>>>>> WDYT?
>>>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>
>>