[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16655114#comment-16655114 ] ASF GitHub Bot commented on IGNITE-9082: Github user asfgit closed the pull request at: https://github.com/apache/ignite/pull/4809 > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.8 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16655028#comment-16655028 ] Alexei Scherbakov commented on IGNITE-9082: --- Blockers do not seem to be related to the commit. Ready for merge. > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.8 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[
https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16655027#comment-16655027
]
Ignite TC Bot commented on IGNITE-9082:
---
{panel:title=Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}Hadoop{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=2108685]]
* IgniteHadoopFileSystemClientBasedPrimarySelfTest.testClientReconnect (last
started)
{color:#d04437}Inspections (AOP){color} [[tests 0 Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=2108750]]
{color:#d04437}Inspections (Core){color} [[tests 0 Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=2108751]]
{color:#d04437}Platform .NET (Coverage)*{color} [[tests
45|https://ci.ignite.apache.org/viewLog.html?buildId=2108766]]
* exe: SqlDmlTest.TestLogWarning - 1,0% fails in last 100 master runs.
* exe: CacheLinqTest.TestDateTime - 1,0% fails in last 100 master runs.
* exe: CacheLinqTest.TestDateTime - 1,0% fails in last 100 master runs.
* exe: CacheLinqTest.TestDateTime - 1,0% fails in last 100 master runs.
* exe: MessagingTest.TestMessageDataTypes - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestCallJavaService - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestCallJavaServiceDynamicProxy - 1,0% fails in last 100
master runs.
* exe: ServicesTest.TestCancel - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestCancelException - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestDeployAllInputErrors - 1,0% fails in last 100 master
runs.
* exe: ServicesTest.TestDeployClusterSingleton - 1,0% fails in last 100 master
runs.
* exe: ServicesTest.TestDeployKeyAffinitySingleton - 1,0% fails in last 100
master runs.
* exe: ServicesTest.TestDeployKeyAffinitySingletonBinarizable - 1,0% fails in
last 100 master runs.
* exe: ServicesTest.TestDeployMultiple - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestDeployNodeSingleton - 1,0% fails in last 100 master
runs.
* exe: ServicesTest.TestDeploymentExceptionSerializable - 1,0% fails in last
100 master runs.
* exe: ServicesTest.TestExecuteException - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestFooterSetting - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestGetDynamicServiceProxy - 1,0% fails in last 100 master
runs.
* exe: ServicesTest.TestGetDynamicServiceProxyLocal - 1,0% fails in last 100
master runs.
* exe: ServicesTest.TestMarshalExceptionOnRead - 1,0% fails in last 100 master
runs.
* exe: ServicesTest.TestMarshalExceptionOnWrite - 1,0% fails in last 100 master
runs.
* exe: ServicesTest.TestServiceDescriptors - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestWithKeepBinaryBoth - 1,0% fails in last 100 master runs.
* exe: ServicesTest.TestWithKeepBinaryClient - 1,0% fails in last 100 master
runs.
* exe: ServicesTestFullFooter.TestDeploy(False) - 1,0% fails in last 100 master
runs.
* exe: ServicesTestFullFooter.TestDeploy(True) - 1,0% fails in last 100 master
runs.
* exe: ServicesTestFullFooter.TestDeployAll(False) - 1,0% fails in last 100
master runs.
* exe: ServicesTestFullFooter.TestDeployAll(True) - 1,0% fails in last 100
master runs.
* exe: ServicesTestFullFooter.TestDeployAllException(False) - 1,0% fails in
last 100 master runs.
* exe: ServicesTestFullFooter.TestDeployAllException(True) - 1,0% fails in last
100 master runs.
* exe: ServicesTestFullFooter.TestDeployException(False) - 1,0% fails in last
100 master runs.
* exe: ServicesTestFullFooter.TestDeployException(True) - 1,0% fails in last
100 master runs.
* exe: ServicesTestFullFooter.TestDeployMultipleException(False) - 1,0% fails
in last 100 master runs.
* exe: ServicesTestFullFooter.TestDeployMultipleException(True) - 1,0% fails in
last 100 master runs.
* exe: ServicesTestFullFooter.TestDuckTyping(False) - 1,0% fails in last 100
master runs.
* exe: ServicesTestFullFooter.TestDuckTyping(True) - 1,0% fails in last 100
master runs.
* exe: ServicesTestFullFooter.TestGetServiceProxy(False) - 1,0% fails in last
100 master runs.
* exe: ServicesTestFullFooter.TestGetServiceProxy(True) - 1,0% fails in last
100 master runs.
{panel}
[TeamCity Run
All|http://ci.ignite.apache.org/viewLog.html?buildId=2108775buildTypeId=IgniteTests24Java8_RunAll]
> Throwing checked exception during tx commit without node stopping leads to
> data corruption
> --
>
> Key: IGNITE-9082
> URL: https://issues.apache.org/jira/browse/IGNITE-9082
> Project: Ignite
> Issue Type: Bug
> Components: cache
>Affects Versions: 2.4, 2.5, 2.6
>Reporter: Pavel Kovalenko
>Assignee: Alexei Scherbakov
>Priority: Major
> Fix For: 2.8
>
>
> If we get checked exception during
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16639470#comment-16639470 ] Alexei Scherbakov commented on IGNITE-9082: --- [~vozerov], I have plans to finish the ticket in next couple of days. > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16637370#comment-16637370 ] Vladimir Ozerov commented on IGNITE-9082: - [~ascherbakov], [~Jokser], [~agoncharuk], Igniters, Do we have any progress with this ticket? Should we move it to AI 2.8? > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[
https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16631521#comment-16631521
]
Ignite TC Bot commented on IGNITE-9082:
---
{panel:title=Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}Queries 1{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=1956721]]
* IgniteCacheCollocatedQuerySelfTest.testColocatedQueryRight (last started)
{color:#d04437}Java Client{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=1956658]]
* RedisProtocolStringSelfTest.testStrlen (last started)
{color:#d04437}PDS 1{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=1956694]]
* IgniteClusterActivateDeactivateTestWithPersistence.testDeactivateFailover1
(last started)
{color:#d04437}Cache 1{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=1956715]]
* GridCacheColocatedTxExceptionSelfTest.testPutMultipleKeysTx (last started)
{color:#d04437}Cache (Failover) 1{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=1956707]]
* GridCacheNearRemoveFailureTest.testPutAndRemovePessimisticTx (last started)
{color:#d04437}_Licenses Headers_{color} [[tests 0 Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=1956705]]
{color:#d04437}MVCC Cache{color} [[tests
2|https://ci.ignite.apache.org/viewLog.html?buildId=1956735]]
* IgniteCacheMvccTestSuite:
CacheMvccReplicatedCoordinatorFailoverTest.testAccountsTxGet_Server_Backups0_CoordinatorFails_Persistence
- 0,0% fails in last 100 master runs.
* IgniteCacheMvccTestSuite:
CacheMvccReplicatedCoordinatorFailoverTest.testAccountsTxGet_SingleNode_CoordinatorFails
- 0,0% fails in last 100 master runs.
{color:#d04437}PDS (Direct IO) 2{color} [[tests
3|https://ci.ignite.apache.org/viewLog.html?buildId=1956691]]
* IgnitePdsNativeIoTestSuite2:
IgniteWalFlushLogOnlyWithMmapBufferSelfTest.testFailWhileStart - 0,0% fails in
last 100 master runs.
{panel}
[TeamCity Run
All|http://ci.ignite.apache.org/viewLog.html?buildId=1956738buildTypeId=IgniteTests24Java8_RunAll]
> Throwing checked exception during tx commit without node stopping leads to
> data corruption
> --
>
> Key: IGNITE-9082
> URL: https://issues.apache.org/jira/browse/IGNITE-9082
> Project: Ignite
> Issue Type: Bug
> Components: cache
>Affects Versions: 2.4, 2.5, 2.6
>Reporter: Pavel Kovalenko
>Assignee: Alexei Scherbakov
>Priority: Major
> Fix For: 2.7
>
>
> If we get checked exception during tx commit on a primary node and this
> exception is not supposed to be handled as NodeStopping OR doesn't lead to
> node stop using Failure Handler, in this case, we may get data loss on a node
> which is a backup node for this tx.
> Possible solution:
> If we get any checked or unchecked exception during tx commit we should stop
> this node after that to prevent further data loss.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16628907#comment-16628907 ] Pavel Kovalenko commented on IGNITE-9082: - [~ascherbakov] Thank you for contribution. No objections regarding PR. But I noticed several suspicious failures in MTCGA analysys (linked to ticket). Please re-trigger failing suites to make sure that new changes don't affect master branch state. > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16628881#comment-16628881 ] Alexei Scherbakov commented on IGNITE-9082: --- [~Jokser], I've fixed review comments, please take a look again. > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16627000#comment-16627000 ] Alexei Scherbakov commented on IGNITE-9082: --- Done https://reviews.ignite.apache.org/ignite/review/IGNT-CR-874 > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16626036#comment-16626036 ] Pavel Kovalenko commented on IGNITE-9082: - [~ascherbakov] Could you please create a review for this change at reviews.ignite.apache.org ? > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16623897#comment-16623897 ] Alexei Scherbakov commented on IGNITE-9082: --- [~NIzhikov], PR is ready so chances are high. > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16623886#comment-16623886 ] ASF GitHub Bot commented on IGNITE-9082: GitHub user ascherbakoff opened a pull request: https://github.com/apache/ignite/pull/4809 IGNITE-9082 Throwing checked exception during tx commit without node stopping leads to data corruption You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite ignite-9082 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/ignite/pull/4809.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #4809 commit 07d90f0554a20b405d3787f19ee3e9f1bb7ec550 Author: Aleksei Scherbakov Date: 2018-09-19T14:04:46Z IGNITE-9082 wip. commit fafefd6cb413567eb7cceef37d21d0f9f7b847ab Author: ascherbakoff Date: 2018-09-19T19:25:25Z IGNITE-9082 wip. commit 2ab602bb007147f1bc8594ed82c89e1c6d778427 Author: Aleksei Scherbakov Date: 2018-09-20T15:48:25Z IGNITE-9082 wip. commit b230793f02707a9a1de8b599452c3514c2494256 Author: Aleksei Scherbakov Date: 2018-09-20T17:24:05Z IGNITE-9082 wip. commit 03e82b17efb73646ba130e98d9e10dd1df834431 Author: Aleksei Scherbakov Date: 2018-09-21T17:02:09Z IGNITE-9082 wip. commit 1df3684ec8528ddbb4ed1c7710d62a2914e374a3 Author: Aleksei Scherbakov Date: 2018-09-21T17:04:02Z Merge branch 'master' of https://github.com/apache/ignite into ignite-9082 > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Commented] (IGNITE-9082) Throwing checked exception during tx commit without node stopping leads to data corruption
[ https://issues.apache.org/jira/browse/IGNITE-9082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16623588#comment-16623588 ] Nikolay Izhikov commented on IGNITE-9082: - [~ascherbakov] Do we have a chance to resolve this ticket until the code freeze of 2.7? > Throwing checked exception during tx commit without node stopping leads to > data corruption > -- > > Key: IGNITE-9082 > URL: https://issues.apache.org/jira/browse/IGNITE-9082 > Project: Ignite > Issue Type: Bug > Components: cache >Affects Versions: 2.4, 2.5, 2.6 >Reporter: Pavel Kovalenko >Assignee: Alexei Scherbakov >Priority: Major > Fix For: 2.7 > > > If we get checked exception during tx commit on a primary node and this > exception is not supposed to be handled as NodeStopping OR doesn't lead to > node stop using Failure Handler, in this case, we may get data loss on a node > which is a backup node for this tx. > Possible solution: > If we get any checked or unchecked exception during tx commit we should stop > this node after that to prevent further data loss. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
