[jira] [Updated] (ACCUMULO-3160) ExamplesIT usage of MAC is sub-par

["Josh Elser (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/ACCUMULO-3160?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Josh Elser updated ACCUMULO-3160:
---------------------------------
    Fix Version/s: 1.7.0

> ExamplesIT usage of MAC is sub-par
> ----------------------------------
>
>                 Key: ACCUMULO-3160
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-3160
>             Project: Accumulo
>          Issue Type: Improvement
>          Components: test
>    Affects Versions: 1.6.0
>            Reporter: Josh Elser
>            Assignee: Josh Elser
>            Priority: Minor
>             Fix For: 1.6.2, 1.7.0
>
>
> ExamplesIT does a couple of screwy things. It tries to share a single MAC instance, assuming, to make the test go faster. This has the downside of any test case that causes the MAC to get into a bad state to hose all of the other tests.
> For example, on my personal jenkins instance, the tserver ran out of heap, and it hung all of the other tests. 
> There's also some larger issue happening where the test cases aren't actually exiting the process for each test which result in a bunch of the Ingest/Query processes that get started for the test to never get cleaned up
> {noformat}
> # jps -m 
> 3460 Main org.apache.accumulo.examples.simple.dirlist.Ingest -i miniInstance -z localhost:21593 -u root -p testRootPassword1 --dirTable dirTable --indexTable indexTable --dataTable dataTable --vis A|B --chunkSize 10000 ....
> 3258 Main org.apache.accumulo.examples.simple.client.RowOperations -i miniInstance -z localhost:21593 -u root -p testRootPassword1
> 2413 Main org.apache.accumulo.master.Master
> 3496 Main org.apache.accumulo.examples.simple.mapreduce.TeraSortIngest --count 1000000 -nk 10 -xk 10 -nv 10 -xv 10 -t sorted -i miniInstance -z localhost:21593 -u root -p testRootPassword1 --splits 4
> 3397 Main org.apache.accumulo.examples.simple.isolation.InterferenceTest -i miniInstance -z localhost:21593 -u root -p testRootPassword1 -t itest2 --iterations 100000
> 2894 Main org.apache.accumulo.examples.simple.isolation.InterferenceTest -i miniInstance -z localhost:21593 -u root -p testRootPassword1 -t itest1 --iterations 100000 --isolated
> 2319 Main org.apache.accumulo.tserver.TabletServer
> 3291 Main org.apache.accumulo.tracer.TraceServer
> 2322 Main org.apache.accumulo.tserver.TabletServer
> 3598 Main org.apache.accumulo.server.util.Admin stopAll
> 3353 Main org.apache.accumulo.examples.simple.client.ReadWriteExample -i miniInstance -z localhost:21593 -u root -p testRootPassword1 --auths A,B --table test2 --createtable -c --debug
> 2416 Main org.apache.accumulo.gc.SimpleGarbageCollector
> 3571 Main org.apache.accumulo.examples.simple.helloworld.InsertWithBatchWriter -i miniInstance -z localhost:21593 -u root -p testRootPassword1 -t helloBatch
> {noformat}
> I'm not sure why the processes aren't getting cleaned up (no destroy in a finally when the test times out?), nor 100% sure why the stopAll appears to hang everything (timeouts not applied to After annotations?).



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)