[
https://issues.apache.org/jira/browse/ZOOKEEPER-4684?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Kezhu Wang resolved ZOOKEEPER-4684.
-----------------------------------
Resolution: Not A Problem
{quote}
I think there are several potential causes for this error. One possibility is
that there is a permission issue with the directory or file, which is
preventing ZooKeeper from renaming the temporary file. Another possibility is
that the disk is full or experiencing other issues, preventing the file rename
from occurring.
{quote}
There is nothing other than crash ZooKeeper can do. Also, 3.4 is end of life.
> Unable to load database on disk exception occurs when the cluster startup
> --------------------------------------------------------------------------
>
> Key: ZOOKEEPER-4684
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-4684
> Project: ZooKeeper
> Issue Type: Bug
> Affects Versions: 3.4.5, 3.4.6, 3.4.9, 3.4.10
> Reporter: ZheyuanLin
> Priority: Major
>
> I found a cluster startup error where two of the three nodes failed to start
> properly due to this exception, so the cluster couldn't hold elections,
> eventually crashing:
> {code:java}
> 2023-03-16 09:08:31,945 [myid:3] - INFO [main:QuorumPeer@429] - currentEpoch
> not found! Creating with a reasonable default of 0. This should only happen
> when you are upgrading your installation
> 2023-03-16 09:08:31,962 [myid:3] - INFO [main:QuorumPeer@444] -
> acceptedEpoch not found! Creating with a reasonable default of 0. This should
> only happen when you are upgrading your installation
> 2023-03-16 09:08:31,972 [myid:3] - ERROR [main:QuorumPeer@453] - Unable to
> load database on disk
> java.io.IOException: Could not rename temporary file
> /zookeeper/zookeeper-3.4.5/zkdata/version-2/currentEpoch.tmp to
> /zookeeper/zookeeper-3.4.5/zkdata/version-2/currentEpoch
> at
> org.apache.zookeeper.common.AtomicFileOutputStream.close(AtomicFileOutputStream.java:82)
> at
> org.apache.zookeeper.server.quorum.QuorumPeer.writeLongToFile(QuorumPeer.java:1117)
> at
> org.apache.zookeeper.server.quorum.QuorumPeer.loadDataBase(QuorumPeer.java:447)
> at
> org.apache.zookeeper.server.quorum.QuorumPeer.start(QuorumPeer.java:409)
> at
> org.apache.zookeeper.server.quorum.QuorumPeerMain.runFromConfig(QuorumPeerMain.java:151)
> at
> org.apache.zookeeper.server.quorum.QuorumPeerMain.initializeAndRun(QuorumPeerMain.java:111)
> at
> org.apache.zookeeper.server.quorum.QuorumPeerMain.main(QuorumPeerMain.java:78)
> 2023-03-16 09:08:31,973 [myid:3] - ERROR [main:QuorumPeerMain@89] -
> Unexpected exception, exiting abnormally
> java.lang.RuntimeException: Unable to run quorum server
> at
> org.apache.zookeeper.server.quorum.QuorumPeer.loadDataBase(QuorumPeer.java:454)
> at
> org.apache.zookeeper.server.quorum.QuorumPeer.start(QuorumPeer.java:409)
> at
> org.apache.zookeeper.server.quorum.QuorumPeerMain.runFromConfig(QuorumPeerMain.java:151)
> at
> org.apache.zookeeper.server.quorum.QuorumPeerMain.initializeAndRun(QuorumPeerMain.java:111)
> at
> org.apache.zookeeper.server.quorum.QuorumPeerMain.main(QuorumPeerMain.java:78)
> Caused by: java.io.IOException: Could not rename temporary file
> /zookeeper/zookeeper-3.4.5/zkdata/version-2/currentEpoch.tmp to
> /zookeeper/zookeeper-3.4.5/zkdata/version-2/currentEpoch
> at
> org.apache.zookeeper.common.AtomicFileOutputStream.close(AtomicFileOutputStream.java:82)
> at
> org.apache.zookeeper.server.quorum.QuorumPeer.writeLongToFile(QuorumPeer.java:1117)
> at
> org.apache.zookeeper.server.quorum.QuorumPeer.loadDataBase(QuorumPeer.java:447)
> ... 4 more
> {code}
> I almost found this exception on version 3.4.X, and the cluster starts
> normally on version 3.5.0 and above.
> I think there are several potential causes for this error. One possibility is
> that there is a permission issue with the directory or file, which is
> preventing ZooKeeper from renaming the temporary file. Another possibility is
> that the disk is full or experiencing other issues, preventing the file
> rename from occurring.
> Hope someone can come and help me solve it, thanks a lot.
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
