VERY high load on server

[Mike Heller]

I have amanda running on serveral servers and last night I tried to back 
up one more to the tape server.  When I arrived this morning, the 
backups were still running and the new server had an extremely high load 
on it.  It's a RedHat Linux 9.0 server and the load was over 520 (quad 
Xeon system).  There were about 1500 processes with "amanda" as the 
user, some are listed below:

amanda    4030  4028  3 01:03 ?        00:17:00 /sbin/restore -tvf -
amanda    4031  4028  0 01:03 ?        00:01:47 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4046     1  0 01:05 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4048  4046  3 01:05 ?        00:18:16 /sbin/restore -tvf -
amanda    4049  4046  0 01:05 ?        00:01:56 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4057     1  0 01:05 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4059  4057  3 01:05 ?        00:17:47 /sbin/restore -tvf -
amanda    4060  4057  0 01:05 ?        00:01:53 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4077     1  0 01:06 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4079  4077  3 01:06 ?        00:16:05 /sbin/restore -tvf -
amanda    4080  4077  0 01:06 ?        00:01:42 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4107     1  0 01:08 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4108  4107  3 01:08 ?        00:14:35 /sbin/restore -tvf -
amanda    4114  4107  0 01:08 ?        00:01:34 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4121     1  0 01:09 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4123  4121  3 01:09 ?        00:14:04 /sbin/restore -tvf -
amanda    4128  4121  0 01:09 ?        00:01:32 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4134     1  0 01:09 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4136  4134  2 01:09 ?        00:13:13 /sbin/restore -tvf -
amanda    4137  4134  0 01:09 ?        00:01:24 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4145     1  0 01:09 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4147  4145  2 01:09 ?        00:12:26 /sbin/restore -tvf -
amanda    4150  4145  0 01:09 ?        00:01:19 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4169     1  0 01:11 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4170  4169  2 01:11 ?        00:13:14 /sbin/restore -tvf -
amanda    4172  4169  0 01:11 ?        00:01:25 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4181     1  0 01:11 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4186  4181  2 01:11 ?        00:11:49 /sbin/restore -tvf -
amanda    4187  4181  0 01:11 ?        00:01:15 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4210     1  0 01:12 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4212  4210  2 01:12 ?        00:11:08 /sbin/restore -tvf -
amanda    4213  4210  0 01:12 ?        00:01:09 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4227     1  0 01:14 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4228  4227  2 01:14 ?        00:12:04 /sbin/restore -tvf -
amanda    4229  4227  0 01:14 ?        00:01:19 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4239     1  0 01:14 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4241  4239  2 01:14 ?        00:12:08 /sbin/restore -tvf -
amanda    4242  4239  0 01:14 ?        00:01:17 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\
amanda    4259     1  0 01:15 ?        00:00:00 sh -c /sbin/restore -tvf 
- 2>&1
amanda    4260  4259  2 01:15 ?        00:11:56 /sbin/restore -tvf -
amanda    4262  4259  0 01:15 ?        00:01:16 sed -e ?s/^leaf[ 
?]*[0-9]*[ ?]*\

That is only a small sample.  I have not seen this on any other server 
before and I'm not sure what is causing it.  Tonight I'm going to try 
and backup another directory from that server that is less than 10 MB in 
size to see if it completes without problems.

Can anyone tell me what may be happening?  Would filling the tape on the 
tape server cause this?

Any help appreciated,
Mike