[ 
https://issues.apache.org/jira/browse/IGNITE-13051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17138584#comment-17138584
 ] 

Maksim Timonin edited comment on IGNITE-13051 at 6/17/20, 4:07 PM:
-------------------------------------------------------------------

[~ascherbakov] hi! thanks for comments!
 # This is not Free Switch Exchange specific bug. The reason is that MVCC flag 
is not set correctly in 3 places (my test cases cover that):
 ## when client topology is not configured with mvcc CacheConfiguration;
 ## when cluster_state_change event appeared for client topology;
 ## when cache joined to client topology after node already started.

            I can add tests that covers the free-switch-exchange too and the 
tests will pass. But do we really need them as they are already exists in your 
13003 PR.

     2. Yes, this is a point. The reason I did it is an attempt to fix flag 
setting for all possible cases. But I think you're correct here and split this 
change on 3 places as I described above;

     3.  Unfortunately naming _Future_ was introduced before me, I can't fix 
without breaking code ownership. I'll remove all my code from this place and 
somebody could fix the naming in future in any related tasks.

 

Add screenshot with passed tests and enabled MVCC to attachments

 


was (Author: timonin.maksim):
[~ascherbakov] hi! thanks for comments!
 # This is not Free Switch Exchange specific bug. The reason is that MVCC flag 
is not set correctly in 3 places (my test cases cover that):
 ## when client topology is not configured with mvcc CacheConfiguration;
 ## when cluster_state_change event appeared for client topology;
 ## when cache joined to client topology after node already started.

            I can add tests that covers the free-switch-exchange too and the 
tests will pass. But do we really need them as they are already exists in your 
13003 PR.

     2. Yes, this is a point. The reason I did it is an attempt to fix flag 
setting for all possible cases. But I think you're correct here and split this 
change on 3 places as I described above;

     3.  Unfortunately naming _Future_ was introduced before me, I can't fix 
without breaking code ownership. I'll remove all my code from this place and 
somebody could fix the naming in future in any related tasks.

 

Add screenshot with passed tests and enabled MVCC to attachments

 

> Optimized affinity switch on baseline node left is broken for client topology 
> and MVCC
> --------------------------------------------------------------------------------------
>
>                 Key: IGNITE-13051
>                 URL: https://issues.apache.org/jira/browse/IGNITE-13051
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Alexey Scherbakov
>            Assignee: Maksim Timonin
>            Priority: Critical
>             Fix For: 2.9
>
>         Attachments: reproducer.patch, Снимок экрана 2020-06-03 в 
> 19.41.53.png, Снимок экрана 2020-06-03 в 19.42.44.png, Снимок экрана 
> 2020-06-17 в 19.03.09.png
>
>          Time Spent: 2.5h
>  Remaining Estimate: 0h
>
> If a node contains only client cache topology with MVCC enabled, PME will 
> hang after changes introduced in IGNITE-12617.
> Reproduced by 
> CachePartitionLossWithRestartsTest.testPartitionLossDetectionOnClientTopology[0
>  false false -1] and enabled MVCC.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to