Thanks to some help off the list I think that the problems with
amd hanging in 'siobi' state is solved. Now that it's been reliable enough
to put into production, I'm seeing some interesting problems. After it's
been running for 24 hours, then the web traffic is turned off I get some
mounted directories that won't unmount. I can't unmount them with amq, or
umount.
On a tip I loaded up lsof to see if there were any files open on
those systems, here is what I found (sorry about the formatting).
31# mount
/dev/wd0s1a on / (local, writes: sync 341 async 4211)
/dev/wd0s1f on /usr (local, writes: sync 5195 async 34714)
/dev/wd0s1e on /var (local, writes: sync 296 async 6940)
procfs on /proc (local)
pid157@hostname:/usr/amd/Hold on /usr/amd/Hold
pid157@hostname:/usr/amd/Interfaces on /usr/amd/Interfaces
Web001:/Space/209.132.28 on /usr/amd/realmounts/Web001/Space/209.132.28
Web001:/Space/209.132.30 on /usr/amd/realmounts/Web001/Space/209.132.30
Web005:/Space/209.132.45 on /usr/amd/realmounts/Web005/Space/209.132.45
Web005:/Space/209.132.46 on /usr/amd/realmounts/Web005/Space/209.132.46
Web006:/Space/209.132.48 on /usr/amd/realmounts/Web006/Space/209.132.48
Web006:/Space/209.132.51 on /usr/amd/realmounts/Web006/Space/209.132.51
Web007:/Space/209.132.52 on /usr/amd/realmounts/Web007/Space/209.132.52
Web007:/Space/209.132.53 on /usr/amd/realmounts/Web007/Space/209.132.53
Web006:/Space/209.132.50 on /usr/amd/realmounts/Web006/Space/209.132.50
22# /usr/local/sbin/lsof -N -R
COMMAND PID PPID USER FD TYPE DEVICE SIZE/OFF NODE
NAME
httpd 58179 13616 simplenet 4r VREG 255,12189698 193 572486
/usr/amd/realmounts/Web006/Space/209.132.50/176/Server/Documents/.htaccess
httpd 58202 13616 simplenet cwd VDIR 255,4784130 96 844036
/usr/amd/realmounts/Web001/Space/209.132.28/145/Server/Documents/error
httpd 58402 13616 simplenet cwd VDIR 255,4784130 96 844036
/usr/amd/realmounts/Web001/Space/209.132.28/145/Server/Documents/error
PID 13616 was an httpd process, as were the other PID's. After I killed
that process the files were no longer open according to lsof, however the
filesystems were still mounted, and I am still unable to umount them by
hand. All of the hosts that these "stuck" filesystems are on are up, and
nfs-mountable from the sun systems in our office.
Any comments or suggestions on things to try are welcome. One item
that may be of interest is that while I am running portmapper (amd won't
run without it) I'm not running rpc.statd, since it didn't seem like it
was needed. I'm thinking that I'll start it up though, and see if that
helps.
Thanks,
Doug
--
On account of being a democracy and run by the people, we are the only
nation in the world that has to keep a government four years, no matter
what it does.
-- Will Rogers
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
