On Sun, Oct 18, 2009 at 8:50 PM, Stefan Fritsch <[email protected]> wrote: > On Thursday 15 October 2009, Dick Davies wrote: >> In any event, does it made sense to use something other than the >> inode as the key into the lockDB - the URI for example? > > Is the performance improvement of inode keyed locking so large that it > is worth the hassle? If mod_dav_fs used filename keyed locking > entirely, there would be an easy way to make file replacement by PUT > atomic (see PR 39815). The current behaviour of deleting the old and > the new file when the PUT fails is really bad.
Agreed. I can see how inodes are faster, but I'd like an alternative option - maybe a similar config option to the ETags tunable ; if the DBM key was mtime+inode number my problem would go away, for example. [ I know it may sounds like 'Doctor it hurts when I do this' wrt. other processes editing files underneath mod_dav, but it's a fairly common thing to do in most installs I've seen. It's certainly the first time I've heard that DAV requires exclusive access to a filesystem to avoid strange behaviour. ]
