Hi Willy, On Sat, Apr 13, 2019 at 08:01:53AM +0200, Willy Tarreau wrote: > Did you issue one of the commands that tries to be alone, thus "show sess" > or "show fd" ? It's possible that you were having only one thread blocked > initially and that with the command that was waiting for all threads to > stop, at some point they all wake up to wait and all eat your CPU.
nice reminder for those two commands. we however don't control all the commands which are done through the unix socket as we have our provisioner using it every n seconds, and also another process collecting the stats. We however don't issue "show sess" or "show fd" to my knowledge. > Thus the real issue to figure is why one thread was blocked. > This is a spinlock so apparently it is one of the culprits. The problem > is that I see no other place where this lock is taken and not restored > in this code part. So it might have happened in another function. If > you ever manage so see this again and obtain a core, I'm interested in > seeing all threads' backtraces. In the mean time I'm carefully rechecking > all locked functions. Yes I will be more carefull next time we trigger this one if we do with last HEAD. I would like to automate a bit more the collecting of such debug info when we trigger such rare events. Thanks for the answer anyway, I was reporting this as a "warning" to collect our possible known issues, even if it was a very partial bug report. -- William
