You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Ignite TC Bot (Jira)" <ji...@apache.org> on 2022/04/06 08:28:00 UTC
[jira] [Commented] (IGNITE-16118) Near cache entry can miss a TTL update if an NPE occurs on the primary node.
[ https://issues.apache.org/jira/browse/IGNITE-16118?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17517927#comment-17517927 ]
Ignite TC Bot commented on IGNITE-16118:
----------------------------------------
{panel:title=Branch: [pull/9865/head] Base: [master] : No blockers found!|borderStyle=dashed|borderColor=#ccc|titleBGColor=#D6F7C1}{panel}
{panel:title=Branch: [pull/9865/head] Base: [master] : No new tests found!|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}{panel}
[TeamCity *--> Run :: All* Results|https://ci.ignite.apache.org/viewLog.html?buildId=6507178&buildTypeId=IgniteTests24Java8_RunAll]
> Near cache entry can miss a TTL update if an NPE occurs on the primary node.
> ----------------------------------------------------------------------------
>
> Key: IGNITE-16118
> URL: https://issues.apache.org/jira/browse/IGNITE-16118
> Project: Ignite
> Issue Type: Bug
> Reporter: Pavel Pereslegin
> Assignee: Pavel Pereslegin
> Priority: Minor
> Labels: ise
> Fix For: 2.13
>
> Attachments: IgniteCacheAtomicExpiryPolicyReadTest.java, ttl-updates.png
>
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> In a rare case, you might observe a confusing {{GridCacheEntryRemovedException "Failed to send TTL update request"}} in logs while *reading a non-expired* cache value.
> {noformat}
> [ERROR][sys-#258%expiry.EntryRemovedOnReadTest2%][root] <default> Failed to send TTL update request.
> org.apache.ignite.internal.processors.cache.GridCacheEntryRemovedException
> at org.apache.ignite.internal.processors.cache.GridCacheMapEntry.checkObsolete(GridCacheMapEntry.java:3052)
> at org.apache.ignite.internal.processors.cache.distributed.dht.GridDhtCacheEntry.checkReadersLocked(GridDhtCacheEntry.java:732)
> at org.apache.ignite.internal.processors.cache.distributed.dht.GridDhtCacheEntry.checkReaders(GridDhtCacheEntry.java:708)
> at org.apache.ignite.internal.processors.cache.distributed.dht.GridDhtCacheEntry.readers(GridDhtCacheEntry.java:416)
> at org.apache.ignite.internal.processors.cache.distributed.dht.GridDhtCacheAdapter$8.run(GridDhtCacheAdapter.java:1122)
> at org.apache.ignite.internal.util.IgniteUtils.wrapThreadLoader(IgniteUtils.java:7329)
> at org.apache.ignite.internal.processors.closure.GridClosureProcessor$1.body(GridClosureProcessor.java:827)
> at org.apache.ignite.internal.util.worker.GridWorker.run(GridWorker.java:125)
> at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748)
> {noformat}
> It looks like there is a race between {{sendTtlUpdateRequest(UUID, GridCacheTtlUpdateRequest)}} and heap entry eviction.
> This situation is currently not handled correctly, resulting in a hidden NullPointerException that aborts TTL updates.
> Due to the nature of the TTL updates in Ignite, this problem can only occur when the TTL is updated from the backup node.
> !ttl-updates.png!
> Thus, in a very rare case, we may notice a lack of updates in the near-cache and/or on "non-initiator" backups
--
This message was sent by Atlassian Jira
(v8.20.1#820001)