On a related note ..
What is the easiest way to debug sendsize ?
Im a bit stumped with the following.
sendsize used to work OK for weeks (with the little patch that calls
calcsize iso gnutar) .
As of yesterday sendsize for my Linux boxen acts up weird.
For one machine it only returns the level 0 dump estimate.
On another machine it returns only the level 0 estimate for disk 1, but
gives correct estimates for disk 2 ..
The 3rd linuxbox givbes correct results for all disks ..
The solaris box (that sues gnutar) has rthe same patch and works fine for
all disks.
The sendsize.*.debug on the dustpuppy (whcih has one correct one fail) sez:
sendsize: debug 1 pid 23298 ruid 0 euid 0 start time Mon Feb 26 20:00:04 2001
/volume/bin/sendsize: version 2.4.2p1
calculating for amname 'sda4', dirname '/localdata'
calculating for amname 'sdb1', dirname '/localdatb'
calcsize: running cmd: calcsize GNUTAR sda4 /localdata 0 0
calcsize: running cmd: calcsize GNUTAR sdb1 /localdatb 0 0 1 982782460
sendsize: pid 23298 finish time Mon Feb 26 20:00:56 2001
Now I had an aborted amanda run yesterday afternoon ( i had added some
entries to disklist and was running an amanda trial run to see if
everything worked )
Could it be that this has anything to do with this ?
Currently listening to: 01-mr_self_destruct
Gerhard, <@jasongeo.com> == The Acoustic Motorbiker ==
--
__O If your watch is wound, wound to run, it will
=`\<, If your time is due, due to come, it will
(=)/(=) Living this life, is like trying to learn latin
in a chines firedrill
