DJ Lucas wrote: > Bruce Dubbs wrote: >> [EMAIL PROTECTED] wrote: >>> Author: dj >>> Date: 2008-03-15 00:47:29 -0600 (Sat, 15 Mar 2008) >>> New Revision: 8496 >>> >>> Modified: >>> trunk/bootscripts/contrib/lsb-v3/Changelog >>> trunk/bootscripts/contrib/lsb-v3/lsb/init-functions >>> Log: >>> contrib/lsb-v3 - added missing stale pid file checking for pidofproc >> >> Do we need to add an errata entry for LFS 6.3 to address this? >> >> -- Bruce > > :-) Not for the contrib/lsb-v3 bootscripts, but yes, we need to correct > the pid/pidfile handling in the current bootscripts and roll a new > tarball. I don't have the stock scripts installed anymore, but I can > setup to use both so that I can test. It might take me a little longer > than the fix above did, as I haven't touched the old scripts in quite > some time, and it'll likely be a pretty big change. > > Actually, just copy pidofproc from my scripts and drop them into > current, and just continue to always spit out a list or nothing and we > should be good, but something is telling me it's not quite that > simple...oh yeah, pidlist is reused. Shouldn't be too hard to work > around. Anyway, I'll test that tomorrow evening unless somebody else > wants to try it first.
Actually, I just looked at this. We'd have to change the functions file in 12 places to account for the different behavior (no -s (silent) flag). No LFS or BLFS bootscripts use the -s flag. I'll hack up a tenative patch real quick, but do know that it's 3:00AM here and will be completely untested. If somebody can reproduce the problem easily, and test the patch, that would be great, otherwise I can test it tomorrow. -- DJ Lucas -- http://linuxfromscratch.org/mailman/listinfo/lfs-book FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
