MDS: stuck in clientreplay
I can't find any hints, except that it is in fact asking to stay in clientreplay
#2 Updated by Sam Lang almost 8 years ago
- Status changed from New to Duplicate
This is a duplicate of #4742. It looks like setattr is the culprit. I was able to generate a core file of the mds while it was in this state, and the only request sitting in mds->mdcache->active_requests is the setattr which the client is waiting for (and already has an unsafe reply to). I have the dump of the mds cache as well, all that it shows for the inode the setattr is operating on is that its dirty.
#3 Updated by Greg Farnum almost 8 years ago
Interesting; on #4742 it was clearly waiting on a request because it kept saying "still have 1 active replay requests" as soon as I turned on logging, whereas these two didn't. That log line is output from MDS::_dispatch() so I guess maybe #4742 got another request that kept retrying, whereas these ones didn't? Or perhaps something else which I can't guess the significance of.