On Friday 13 September 2002 02:57, Eric.Doutreleau AT int-evry DOT fr 
wrote:
>I have a strange problem with amanda
>
>my server is a Redhat linux 7.3 machine with amanda 2.4.3b3
>my client is a redhat7.3 machine with a stock amanda package from
> redhat version 2.4.2p2
>
>In my report for this client some backup work and some don't
>
>I got the message
>could not connect to molure for some of the partition.
>
>Indeed when i ran amcheck this machine coul'nt be check and i got
> the same message.
>
>When i log on this machine all is running except the fact than
> xinetd didn't spawn the amandad process.
>
>when i do a ps i found an amandad process
>
>15933 ?        Z      0:00 [amandad <defunct>]
>
>this process is a son of the xinetd process
>
>but when i look at amandad file related to this process all seems
> ok
>
>Here is the end of this file
>
>got packet:
>--------
>Amanda 2.4 REQ HANDLE 000-F8B70708 SEQ 1031857221
>SECURITY USER dumpy
>SERVICE sendbackup
>OPTIONS hostname=molure;
>GNUTAR /var/spool/imap13 0 1970:1:1:0:0:0 OPTIONS
>
>|;bsd-auth;compress-fast;index;exclude-list=/usr/local/lib/amanda/
>|exclude.gtar;
>
>--------
>
>sending ack:
>----
>Amanda 2.4 ACK HANDLE 000-F8B70708 SEQ 1031857221
>----
>
>bsd security: remote host backup.int-evry.fr user dumpy local user
> amanda amandahosts security check passed
>amandad: running service "/usr/lib/amanda/sendbackup"
>amandad: sending REP packet:
>----
>Amanda 2.4 REP HANDLE 000-F8B70708 SEQ 1031857221
>CONNECT DATA 57421 MESG 57422 INDEX 57423
>OPTIONS ;compress-fast;bsd-auth;index;
>----
>
>amandad: got packet:
>----
>Amanda 2.4 ACK HANDLE 000-F8B70708 SEQ 1031857221
>----
>
>amandad: pid 15933 finish time Thu Sep 12 22:17:32 2002
>
>
>But the prcess stuck seems to prevent the launch of the following
> backup of different partition on the same machine?
>
>if i restart xinetd all run fine again
>but at the next backup i got again this problem.
>
>Has someone already seen that kind of problem?
>
>Thanks in advance for any help

The last time I had a "zombie" process like that, which was blocking 
any new incarnations, I failed to 'kill number' it several times 
and finally had to reboot, but that was probably 20 kernel versions 
back up the log.  Thats probably what you'll have to do to the 
machine with the zombie process.  You might even consider a fresher 
kernel version.

-- 
Cheers, Gene
AMD K6-III@500mhz 320M
Athlon1600XP@1400mhz  512M
99.15% setiathome rank, not too shabby for a WV hillbilly