DataStax Help Center

Nodetool status showing incorrect data load values after incremental repair

Summary

When running incremental repair, the output of nodetool status may show incorrect data load values

Symptoms

After running incremental repairs, the output of nodetool status is observed to show varying values of data load which seem incorrect.

The problem can be verified by using du, the result of which will show the actual data size is markedly different to what nodetool status reports

 

nodetool status my_keyspace 
Datacenter: DC1
===============
Status=Up/Down
|/ State=Normal/Leaving/Joining/Moving
-- Address Load Tokens Owns (effective) Host ID Rack
UN 10.1.2.1 356.26 GB 256 53.4% 1a2d5a21-cfad-4499-a1f5-92bf20c58e7b RAC1
UN 10.1.2.2 437.72 GB 256 50.1% 2289dd18-15d6-46ce-bf56-4ef7a27462b0 RAC1
UN 10.1.2.3 252.51 GB 256 49.9% 4ad481c7-d185-4faf-9817-7cb36c95322a RAC1
UN 10.1.2.4 372.33 GB 256 47.7% 0ace0d32-d041-4bb7-99ab-389c3a06d67d RAC1
UN 10.1.2.5 702.85 GB 256 48.9% e6d9f5ba-3774-48bb-8186-be07ef9041fb RAC1
UN 10.1.2.6 383.21 GB 256 50.1% fb38e223-e2de-4c99-a8b4-ff134a9c3279 RAC1

And running du from node 10.1.2.2

du -sh *
8.0K dse_perf
20K dse_system
236G my_keyspace 
16K lost+found
1.3G OpsCenter
2.5M system
132K system_auth
1.3M system_traces 

Cause

The problem is caused by CASSANDRA-10430 which is a duplicate of CASSANDRA-10831

Workaround

If the node is restarted the correct values will show, however this is not an ideal workaround but useful to verify the problem is just a visual issue.

Solution

Upgrade to DSE4.7.7, DSE4.8.4 or later

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

Comments

Powered by Zendesk