After a rolling restart of 26 indexers, I show 162 indexes in state: Cannot fix search count as the bucket hasn't rolled yet.
I get this state every time I perform this rolling restart, which is done during off hours to reduce overhead.
according to the docs http://docs.splunk.com/Documentation/Splunk/6.4.3/Indexer/Backupindexeddata#Rolling_buckets_manually... where you find the example command, it is done per index. But I'm sure you can script it and loop over all indexes if needed.
Hope this helps ...
Thanks for the answer. Yes, I got the command from the same link. I wanted to check if there is a Splunk command to move all the index data at once.
If not then Splunk should come up with one 🙂
If you force a restart of all peers, during the proccess all hot buckets will roll to warm.
So you can use the rolling restart command: