DataStax Help Center

Repair service incorrectly reporting timeout warnings in OpsCenter releases prior to 5.2.3

Summary

In releases prior to 5.2.3 some users might see incorrect repair service timeouts warnings.

Symptoms

The following types of errors might be seen in the repair service log:

2015-07-08 05:52:12+0000 [MyCluster] ERROR: Repair task (<Node 10.1.2.3='-502648577219449610'>, [-9185345399602795050, 9217528375793162274]) timed out after 3600 seconds.
2015-07-08 05:52:12+0000 [MyCluster] ERROR: Task (<Node 10.1.2.3='-502648577219449610'>, [-9185345399602795050, 9217528375793162274]) has failed 9 times.
2015-07-08 05:52:12+0000 [MyCluster] ERROR: 9 errors have occurred out of 100 allowed.
2015-07-08 05:52:12+0000 [MyCluster]  INFO: Adding repair task to end of queue
2015-07-08 05:52:12+0000 [MyCluster]  INFO: Beginning repair of range [-9185345399602795050, 9217528375793162274] on node 10.1.2.3: ks=my_keyspace, cfs=[u'my_keyspace', u'my_keyspace_users_keyspaces']

 

Cause

There was a known issue in outlined in the following internal jira:

OPSC-5955 - Repair service reporting timeouts when actually repairs are completing ok

Workaround

Checking the actual node logs for repair failure messages will suffice until it is possible to upgrade. Using manual repair around the ring is also an alternative.

Solution

Upgrade to OpsCenter 5.2.3

 

Was this article helpful?
1 out of 1 found this helpful
Have more questions? Submit a request

Comments

Powered by Zendesk