[00:49:21] --- Stephan Wiesand has become available [06:05:27] --- mvitale has become available [06:27:48] --- meffie has become available [06:29:44] --- deason has become available [07:32:58] --- Stephan Wiesand has left [09:46:29] --- jwelsh has become available [10:22:24] --- jwelsh has left [10:22:32] --- meffie has left [10:22:43] --- m.meffie has become available [10:58:52] --- squinney has become available [11:51:31] --- Simon Wilkinson has become available [12:25:47] I have a two-day-old-process running on whole-enchilada.mit.edu that took an flock on /afs/sipb.mit.edu/project/feed/web_scripts/tt-rss/lock/update_daemon.lock. This lock is visible from other local processes, but not from processes on any remote machine. (Other flocks acquired on the same machine at the same time are still visible remotely.) Should I expect this to work? Any debugging ideas? [12:25:48] --- Marc Dionne has left: Lost connection [12:33:10] --- squinney has left [12:37:31] --- Simon Wilkinson has left [12:39:19] the fileserver only honors locks for.... 5 minutes at a time, I think? if the client couldn't reach the fileserver for that long, maybe it had issues acquiring it again [12:41:12] briefly looking, it looks like we only ExtendLock in the background, so if we lost the lock, we wouldn't get it back because we don't try to SetLock again [12:41:59] maybe I'm missing some detail where it works anyway, but that would be my guess [12:56:42] --- squinney has become available [13:10:10] --- squinney has left [13:43:32] Derrick, I don't understand what you want me to do. Remove the cherry-picked line and add a comment that this is applied to master after going into 1.6.2? [13:44:20] i think that would make more sense. [13:46:05] Okay. [13:47:03] Is it worth including the (truncated) hash of the older commit? [13:51:02] I'd like it in there in some capacity... [13:56:08] Have a patchset 2. [15:21:48] --- mvitale has left [15:22:38] --- m.meffie has left [15:23:56] --- deason has left [20:15:48] --- ballbery has become available [22:04:06] --- squinney has become available [23:11:29] --- squinney has left [23:56:45] --- Stephan Wiesand has become available