[
https://issues.apache.org/jira/browse/IGNITE-7090?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16335794#comment-16335794
]
Vladisav Jelisavcic commented on IGNITE-7090:
---------------------------------------------
Hi Tim, I reviewed the updated pull request. Another thing came to my mind,
this solution will only work if the node 1 that created the first semaphore
instance can execute transaction after it fails, we should cover a more dire
scenario when it can't. In this case the node 2 which creates the second
semaphore instance has to do the recovery.
Since we already have a mapping of all nodes that are waiting on semaphore in
GridCacheSemaphoreState we can check this when creating the new instance (in
DataStructuresProcessor#semaphore() method). What do you think?
> Semaphore Stuck when no acquirers to assign permit
> --------------------------------------------------
>
> Key: IGNITE-7090
> URL: https://issues.apache.org/jira/browse/IGNITE-7090
> Project: Ignite
> Issue Type: Bug
> Components: cache, data structures
> Affects Versions: 2.1, 2.4
> Reporter: Tim Onyschak
> Priority: Major
> Fix For: 2.5
>
> Attachments: SemaphoreFailoverNoWaitingAcquirerTest.java
>
>
> If no acquirers are available to take permit of semaphore, the permit never
> gets release and any further acquirerers will wait forever.
> On node shut down DataStructuresProcessor.dsMap gets cleared out prior to
> event listener being able to execute onNodeRemoved, hence owner is never
> cleared out if it was unable to pass to a different acquirer.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
