Thanks to all that responded. Original question at the end.
alan_at_nabeth.cxo.dec.com
>I'd suggesting checking to see if there are any patches
> of interest from Services.
>
> Dump (and probably vdump) start a number of processes
> to actually backup the data; one parent process and a
> bunch of children to handle the various parts of the
> backup. What could be happening is that one of the
> children is writing the "raw_getproplist" message as
> a result of some error and then dies. The parent
> notices, doesn't gracefully deal with the child having
> died and exits itself.
>
> Look around to see if there is a core dump that might
> be the child. If the original program isn't stripped
> and you have the license to use dbx, that might offer
> a clue what it was doing. With that you might be able
> to do some sort of work-around.
>
I did not find a core file and we are at patch level 2 on all of our
systems and this was the only one giving the problem.
By the way I was backing this system up to upgrade to patch kit 3.
Robert Yohn
>how much free space to you have on /usr? If it is a pretty full
filesystem, you will not have any free space >for the os to put some temp
files while it backs up.
/usr was 75% full on a 2GB partition
>Also, have you tried to backup /usr in multiuser mode? I'm running the
same setup as you, AS4100, 4.0d with >PK6 and DAT tape drive. I haven't
experienced any of these problems, but if I think of anything else I'll
>let you know.
dump still failed in multiuser mode
Darryl Cook
>I ran into a similar problem with dump. My machine would go along fine
>and finish with no errors but upon checking it left off files near the end
>of the backup process. I put out a post similar to yours and got replies
>that basically said there is a limit to the number of files in one file
>system that dump will back up. I switched to vdump and have not had any
>problems since. Hope this helps. BTW vdump works the same way as dump
>but seems to be more reliable.
>
This worked until I tried to do a vrestore from OS cd. The vrestore would
crash.
The way I was able to get this to work was do a vdump to tape, then
shutdown the system move the disk to another system and do a vrestore. I
then put the drive back in the original system and was now able to do a
regular dump.
I am trying to dumps on an AlphaServer 4100 running 4.0d.
I boot the system to single user mode, run bcheckrc, update and swapon -a.
I then do dump -0f /dev/rmt0h /usr
The dump goes along as though all is well until th last time remaining
statement is printed at which point it print 0:00 time left and then
raw_getproplist: No such process followed by dump:SIGTERM received and then
dump:The ENTIRE dump is aborted.
I am using a tlz09-va and have changed it with a working one.
I am using dds2 dat tapes and have tried multiples.
I even booted off of the 4.0d cd and got the same errors except this time
it said no such file or directory instead of no such process.
I have searched the archive but have not found anything like this.
Received on Mon Apr 17 2000 - 15:49:38 NZST