[
https://issues.apache.org/jira/browse/IGNITE-6767?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16277368#comment-16277368
]
ASF GitHub Bot commented on IGNITE-6767:
----------------------------------------
GitHub user tonyschak reopened a pull request:
https://github.com/apache/ignite/pull/2939
IGNITE-6767 reset of the topVer on loading of the entry
I added though the loadedValue method should reset the topVer since it
could be inval
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/tonyschak/ignite IGNITE-6767
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/ignite/pull/2939.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 #2939
----
commit 84523afcec51c56f0a23ac2d248012841bce11aa
Author: Tim Onyschak <[email protected]>
Date: 2017-10-26T18:59:33Z
IGNITE-6767 reset of the topVer on loading of the entry
commit 1efbd512374a96852e8f92f980c23df543acd322
Author: Tim Onyschak <[email protected]>
Date: 2017-11-08T17:25:20Z
IGNITE-6767 updated from comments by Vyacheslav Koptilin
commit 6b7918446eaeef7ad801250db69fd4f28e477d3e
Author: Tim Onyschak <[email protected]>
Date: 2017-11-08T17:33:53Z
IGNITE-6767 Removed topUpdate from original fix, will be taken care of with
!valid() check
----
> NearCache#localPeek(key, NEAR) always returns 'null' if the node, which owns
> the primary partition for the given key, left the cluster.
> ---------------------------------------------------------------------------------------------------------------------------------------
>
> Key: IGNITE-6767
> URL: https://issues.apache.org/jira/browse/IGNITE-6767
> Project: Ignite
> Issue Type: Bug
> Components: cache
> Affects Versions: 2.1
> Reporter: Vyacheslav Koptilin
> Fix For: 2.4
>
> Attachments: GridCacheNearClientHitTest.java
>
>
> {{IgniteCache.localPeek(key, PeekMode.NEAR)}} always returns 'null' if the
> node, which owns the primary partition for the given key, left the cluster,
> even if {{IgniteCache.get(key)}} was called.
> How to reproduce:
> # start two server nodes.
> # create partitioned, atomic cache and populate data.
> # start client node with near cache configured.
> # perform {{IgniteCache.get(key)}} and check that
> {{IgniteCache.localPeek(key, PeekMode.NEAR)}} returns not null value.
> # stop server node which owns primary partition for the given {{key}}.
> # perform {{IgniteCache.get(key)}}.
> # after that {{IgniteCache.localPeek(key, PeekMode.NEAR)}} always returns
> null value.
> This issue was reported on the user-list:
> http://apache-ignite-users.70518.x6.nabble.com/Near-Cache-Topoolgy-change-causes-NearCache-to-always-miss-tt17539.html
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
