I'm running a 5-node Riak cluster at version 1.1.2, which started
behaving strangely this morning: the main symptom seemed to be almost
all map-reduce queries timing out cluster-wide.

The only other visible problem I was able to find was that 2 out of
the 5 nodes had stopped writing to their logs back in August (!).
Meaning that none of the files in /var/log/riak had been written to
since then, except for one of the erlang.log files which was still
getting ALIVE messages from heart.

I restarted one of the two nodes which had stopped logging, and the
map-reduce timeouts stopped immediately. So I'm wondering if the two
issues could be related somehow, and whether this sounds like anything
that's been fixed since 1.1.2?

(Yeah, I realize the advice is probably to upgrade in any case, but
I'm risk-averse.)

Mike

_______________________________________________
riak-users mailing list
[email protected]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com

Reply via email to