[
https://issues.apache.org/jira/browse/GEODE-7085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16911720#comment-16911720
]
ASF subversion and git services commented on GEODE-7085:
--------------------------------------------------------
Commit f17931bf541fc0255112f713931388d9ee0bbc30 in geode's branch
refs/heads/develop from Dan Smith
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=f17931b ]
GEODE-7085: Ensure bitset is flushed in all code paths
In recordVersion, there was a code path where we would not call
flushBitSetDuringRecording before calling setVersionInBitSet. Moving the
flush to the top of the method to ensure that the bit set is always
flushed, and we never end up with a case where we try to set too large
of a bit.
By code inspection, we determined that initializeFrom was what got us
into the state that triggered this code path - it could leave
bitSetVersion at a small number while setting version to a large number.
Fixing initializeFrom so that it correctly sets bitSetVersion.
Co-Authored-By: Ernest Burghardt <eburgha...@pivotal.io>
> Cannot recover from disk store if region version is greater than
> Integer.MAX_VALUE
> ----------------------------------------------------------------------------------
>
> Key: GEODE-7085
> URL: https://issues.apache.org/jira/browse/GEODE-7085
> Project: Geode
> Issue Type: Bug
> Components: membership, persistence
> Reporter: Dan Smith
> Assignee: Dan Smith
> Priority: Major
> Fix For: 1.11.0
>
> Time Spent: 4h 20m
> Remaining Estimate: 0h
>
> We hit an issue where a member failed to recover due to a
> IndexOutOfBoundsException while recording a version during recovery.
> Looking closer, it looks like the issue is due to the fact that a
> RegionVersionHolder cannot record a version greater than Integer.MAX_VALUE if
> it just just constructed.
> When we are recovering from disk, the first thing we read from is the .drf
> files. The first thing in those drf files is RVV information. We read the RVV
> records and call recordRecoveredGCVersion.
> When that call gets down inside RegionVersionHolder.recordVersion, there is
> some logic that is supposed to flush out the bitSet and advance the
> bitSetVersion. Unfortunately it looks like flushBitSetDuringRecording is not
> actually doing that. So if version we read from disk is greater than
> Integer.MAX_VALUE, we wrap around and try to set a negative index in the
> bitset.
> I can reproduce this with a unit test of RegionVersionVector that records a
> version greater than Integer.MAX_VALUE. I’m looking into how to fix the
> flushBitSetDuringRecording method.
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
