[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-18 Thread ASF subversion and git services (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16404115#comment-16404115
 ] 

ASF subversion and git services commented on SOLR-12083:


Commit 8b014ef41e7bff2d621d759bf7e37781c2ad1d09 in lucene-solr's branch 
refs/heads/branch_7_3 from [~varun_saxena]
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=8b014ef ]

SOLR-12083: Fix RealTime GET to work on a cluster running CDCR when using 
Solr's in-place updates

(cherry picked from commit 57524f1)


> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-16 Thread ASF subversion and git services (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16402531#comment-16402531
 ] 

ASF subversion and git services commented on SOLR-12083:


Commit c4d0223ad40d36fd908bb0d3b291763425fe69b4 in lucene-solr's branch 
refs/heads/master from [~varun_saxena]
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=c4d0223 ]

SOLR-12083: Fix PeerSync, Leader Election failures and CDCR checkpoint 
inconsistencies on a cluster running CDCR


> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-16 Thread ASF subversion and git services (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16402532#comment-16402532
 ] 

ASF subversion and git services commented on SOLR-12083:


Commit 033afbfaad0fc0b0a48967765cddf9e2b455 in lucene-solr's branch 
refs/heads/branch_7x from [~varun_saxena]
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=033afbf ]

SOLR-12083: Fix PeerSync, Leader Election failures and CDCR checkpoint 
inconsistencies on a cluster running CDCR

(cherry picked from commit c4d0223)


> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-14 Thread Varun Thacker (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16399224#comment-16399224
 ] 

Varun Thacker commented on SOLR-12083:
--

Until INFRA-15850 is resolved the user tagged with the commit will not be me 

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-14 Thread ASF subversion and git services (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16399218#comment-16399218
 ] 

ASF subversion and git services commented on SOLR-12083:


Commit f8bbfcdc75af2fe9cfbd6e507fba81d720406402 in lucene-solr's branch 
refs/heads/branch_7x from [~varun_saxena]
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=f8bbfcd ]

SOLR-12083: Fix RealTime GET to work on a cluster running CDCR when using 
Solr's in-place updates

(cherry picked from commit 57524f1)


> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-14 Thread ASF subversion and git services (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16399216#comment-16399216
 ] 

ASF subversion and git services commented on SOLR-12083:


Commit 57524f1d4179f3ab57ffa53ba8f5e4dd1e198a11 in lucene-solr's branch 
refs/heads/master from [~varun_saxena]
[ https://git-wip-us.apache.org/repos/asf?p=lucene-solr.git;h=57524f1 ]

SOLR-12083: Fix RealTime GET to work on a cluster running CDCR when using 
Solr's in-place updates


> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-14 Thread Varun Thacker (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16399205#comment-16399205
 ] 

Varun Thacker commented on SOLR-12083:
--

precommit, all tests, beasted 10 rounds of TestInPlaceUpdatesDistrib and 
TestRealTimeGet all passed

Committing shortly

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-14 Thread Varun Thacker (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16399065#comment-16399065
 ] 

Varun Thacker commented on SOLR-12083:
--

Final patch which I think is ready! Thanks Amrit

Running tests and precommit now. I plan on committing it to master and branch7x 
within the next few hours. The current CHANGES entry for the change is under 
7.3 

If Jenkins is happy today I'll check with the RM if it's okay to backport it to 
the release branch. If it's too late then i'll move the CHANGES to 7.4 on 
master and branch7x

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> SOLR-12083.patch, SOLR-12083.patch, add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-13 Thread Amrit Sarkar (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16398144#comment-16398144
 ] 

Amrit Sarkar commented on SOLR-12083:
-

[~varunthacker], thank you for the feedback.

bq. Any reason why we don't randomzie the update log in 
TestInPlaceUpdatesDistrib like we do in the other approach?
Randomization of UpdateLog should be part of underlying test framework instead 
of developer setting it in tests, as every component should work with 
CdcrUpdateLog, which works with UpdateLog, unless you are testing UpdateLog 
itself. Like in the patch, we enforce 'solr.UpdateLog' for UpdateLogTest. 
Everytime now a new test is written and 'solrconfig-tlog.xml' is used, it will 
automatically test both for UpdateLog and CdcrUpdateLog without the need of 
developer of mentioning it specially.

bq. Can we log in randomizeUpdateLogClsProp which update log is being used
Will do that.

bq. A small refactor could be instead of using a sysprop and then reading the 
sysprop in SolrTestCaseJ4 to call randomizeUpdateLogClsProp , why not just make 
randomizeUpdateLogClsProp and clearUpdateLogClsProp public and call them 
directly in the beforeClass/afterClass method of TestInPlaceUpdatesDistrib / 
UpdateLogTest . What do you think?
I can make them public, but in {{TestInPlaceUpdatesDistrib}}, there is no need 
to specify as this randomization should be smart.

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> When we were adding bi-directional sync support in CDCR ( SOLR-11003 ) we 
> changed the CDCR Update Log codec to write an extra bits. 
> When we use the RealTimeGet component on a cluster running CDCR and have 
> in-place updates in the update log we will falsely trip an assert thus 
> causing the request to fail
> Here's the proposed change
> {code:java}
> - assert entry.size() == 5;
> + if (ulog instanceof CdcrUpdateLog) {
> +   assert entry.size() == 6;
> + }
> + else {
> +   assert entry.size() == 5;
> + }{code}
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-13 Thread Varun Thacker (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16398107#comment-16398107
 ] 

Varun Thacker commented on SOLR-12083:
--

Thanks Amrit for working on this!

I kind of like the SOLR-12083-A-within-test-framework.patch approach better

Couple of things about the SOLR-12083-A-within-test-framework.patch patch
 * Any reason why we don't randomzie the update log in 
TestInPlaceUpdatesDistrib like we do in the other approach?
 * Can we log in randomizeUpdateLogClsProp which update log is being used
 * A small refactor could be instead of using a sysprop and then reading the 
sysprop in SolrTestCaseJ4 to call randomizeUpdateLogClsProp , why not just make 
randomizeUpdateLogClsProp and clearUpdateLogClsProp public and call them 
directly in the beforeClass/afterClass method of TestInPlaceUpdatesDistrib / 
UpdateLogTest . What do you think?

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> On the lines of SOLR-12063: Bidirectional support introduced serious bugs and 
> here RealTimeGetComponent is broken. 
> As we have added additional flag to each {{tlog}} entry, the following 
> assertions fail when Cdcr enabled:
> {code}
> if (oper == UpdateLog.UPDATE_INPLACE) {
>  assert entry.size() == 5;
> }
> {code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-13 Thread Amrit Sarkar (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16397501#comment-16397501
 ] 

Amrit Sarkar commented on SOLR-12083:
-

I have uploaded three patches with different designs:

1. add_support_for_random_ulog_in_tests.patch : This patch has nothing to do 
with the problem here but randomizing UpdateLog b/w 'UpdateLog' and 
'CdcrUpdateLog'. We don't need to do 'HdfsUpdateLog' as it gets picked up 
anyway when directoryFactory is HDFSDirectoryFactory.

2. SOLR-12083-B-wo-test-framework.patch : This patch fixes the problem by 
randomising the UpdateLog b/w 'UpdateLog' and 'CdcrUpdateLog' within the 
{{TestInPlaceUpdatesDistrib}} which validates the fix I have put in place.

3. SOLR-12083-A-within-test-framework.patch : This patch extends the 
'add_support_for_random_ulog_in_tests' with the fix in place.

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083-A-within-test-framework.patch, 
> SOLR-12083-B-wo-test-framework.patch, SOLR-12083.patch, SOLR-12083.patch, 
> add_support_for_random_ulog_in_tests.patch
>
>
> On the lines of SOLR-12063: Bidirectional support introduced serious bugs and 
> here RealTimeGetComponent is broken. 
> As we have added additional flag to each {{tlog}} entry, the following 
> assertions fail when Cdcr enabled:
> {code}
> if (oper == UpdateLog.UPDATE_INPLACE) {
>  assert entry.size() == 5;
> }
> {code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-13 Thread Amrit Sarkar (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16396833#comment-16396833
 ] 

Amrit Sarkar commented on SOLR-12083:
-

If we can randomise between:
{code}

  ${solr.ulog.dir:}
  ${solr.ulog.maxNumLogsToKeep:10}
  ${solr.ulog.numRecordsToKeep:100}
  ${solr.ulog.tlogDfsReplication:2}

{code}
{code}

  ${solr.ulog.dir:}
  ${solr.ulog.maxNumLogsToKeep:10}
  ${solr.ulog.numRecordsToKeep:100}
  ${solr.ulog.tlogDfsReplication:2}

{code}
in our solrconfig's. In-place property substitution doesn't work with XML 
arguments, at least I don't know how to do it. like:
{code}

{code}
[~varunthacker]: unfortunately {{TestRealTimeGet}} doesn't fail when we have 
CdcrUpdateLog, but {{TestInPlaceUpdatesDistrib}} do, which essentially fails 
for RealTimeGetComponent. Please find the updated patch.

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083.patch, SOLR-12083.patch
>
>
> On the lines of SOLR-12063: Bidirectional support introduced serious bugs and 
> here RealTimeGetComponent is broken. 
> As we have added additional flag to each {{tlog}} entry, the following 
> assertions fail when Cdcr enabled:
> {code}
> if (oper == UpdateLog.UPDATE_INPLACE) {
>  assert entry.size() == 5;
> }
> {code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Commented] (SOLR-12083) RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled

2018-03-12 Thread Varun Thacker (JIRA)

[ 
https://issues.apache.org/jira/browse/SOLR-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16396602#comment-16396602
 ] 

Varun Thacker commented on SOLR-12083:
--

I'll think about this more but what if we starting randomizing between the 
TransactionLog and  CdcrTransactionLog for some of our tests. 

For example in TestRealTimeGet what if we randomized between TransactionLog and 
CdcrTransactionLog as the implementation . That would test both implementations 
for real time GET's 

> RealTimeGetComponent fails for INPLACE_UPDATE when Cdcr enabled 
> 
>
> Key: SOLR-12083
> URL: https://issues.apache.org/jira/browse/SOLR-12083
> Project: Solr
>  Issue Type: Bug
>  Security Level: Public(Default Security Level. Issues are Public) 
>  Components: CDCR
>Affects Versions: 7.2, 7.2.1, 7.3
>Reporter: Amrit Sarkar
>Assignee: Varun Thacker
>Priority: Major
> Attachments: SOLR-12083.patch
>
>
> On the lines of SOLR-12063: Bidirectional support introduced serious bugs and 
> here RealTimeGetComponent is broken. 
> As we have added additional flag to each {{tlog}} entry, the following 
> assertions fail when Cdcr enabled:
> {code}
> if (oper == UpdateLog.UPDATE_INPLACE) {
>  assert entry.size() == 5;
> }
> {code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org