Our backup server has spontaneously stopped doing backups of itself, after
working fine for several months. it backs up (most) other clients fine; but
the daily mail message says:
localhost /var/tmp lev 0 FAILED [Request to localhost timed out.]
localhost /var/lib/rpm lev 0 FAILED [Request to localhost timed out.]
localhost /etc lev 0 FAILED [Request to localhost timed out.]
the disklist entry for localhost is:
localhost /etc localhost-nocomp-5am -1 local
localhost /var/lib/rpm tar-comp-high -1 local #(whoops, forgot to change the
dumptype last night).
localhost /var/tmp localhost-nocomp-5am -1 local
and the amanda.conf entries for those dumptypes are:
define dumptype global {
comment "Global definitions"
# This is quite useful for setting global parameters, so you don't have
# to type them everywhere. All dumptype definitions in this sample file
# do include these definitions, either directly or indirectly.
# There's nothing special about the name `global'; if you create any
# dumptype that does not contain the word `global' or the name of any
# other dumptype that contains it, these definitions won't apply.
# Note that these definitions may be overridden in other
# dumptypes, if the redefinitions appear *after* the `global'
# dumptype name.
# You may want to use this for globally enabling or disabling
# indexing, recording, etc. Some examples:
index yes
# record no
}
define dumptype tar-comp-high {
comment "very important partitions on fast machines"
global
program "GNUTAR"
compress client best
priority high
}
define dumptype localhost-nocomp-5am {
comment "localhost, so don't compress. don't use the holdingdisk, so
it won't beat up our one hdd"
global
program "GNUTAR"
compress none
holdingdisk no
starttime 500 # for some reason this doesn't work
priority low
}
I've been messing around with options lately; so if some things don't make
sense, that's probably why. please tell me what could be better, tho. :)
this thing *had* been working fine for some months; the only thing that
changed, was that one client machine was reinstalled/upgraded, and had
screwed-up NFS mounts for a while. that's been fixed now, but it seems like
it was interfering with amanda backups for a while, on those clients that
had non-functioning/stale NFS mounts from that box.
there's plenty of disk space.
the debug files in /tmp/amanda seem fine and normal; the amandad and runtar
files exist and don't have any errors. the 'sendbackup' debug file doesn't
exist, tho. :(
(and yes, the sendbackup binary exists and has the same permissions as a
working system does).
looks like there's only one 'amandad' connect notice in /var/log/secure,
rather than one for each disk to be backed up, like on other systems. I
don't see any errors in syslog or other logs, tho.
where do I continue debugging?
Carl Soderstrom.
--
Network Engineer
Real-Time Enterprises
(952) 943-8700
