Hi Sage, On Wed, 2011-02-16 at 17:54 -0700, Sage Weil wrote: > On Wed, 16 Feb 2011, Sage Weil wrote: > > shouldn't affect anything. We may have missed something.. do you have a > > log showing this in action? > > Obviously yes, looking at your original email. :) At the beginning of > each log line we include a thread id. What would be really helpful would > be to narrow down where in OSD::heartbeat_entry() and heartbeat() things > are blocking, either based on the existing output, or by adding additional > dout lines at interesting points in time. I'll take a deeper look at my existing logs with that in mind; let me know if you'd like me to send you some. I have also been looking at map_lock, as it seems to be shared between the heartbeat and map update threads. Would instrumenting acquiring/releasing that lock be helpful? Is there some other lock that may be more fruitful to instrument? I can reproduce pretty reliably, so adding instrumentation is no problem. -- Jim > > sage > -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
