Many Thanks to Ron Vasey (vasey@issi.com). He had two suggestions.
1) not using -n option of rsh (are you sure you got `em all? ;^)
[We had done this and we had all of them]
2) using the n option of rdump (to notify local users -- sigh!).
*** No. 2 was the winner of the prize. removing the "n" option from the
dump/rdump command fixed the problem. Backups have been running fine
since Wednesday of last week.
Thanks also to:
highscc!jb@gdls.com (Jon Bowman)
shankar@ulysses.att.com (Shankar Ishwar)
rhaddick@us.oracle.com (Robert Haddick)
hellerpd@aa.ab.com (Dave(PD) Heller)
who's responses were also appreciated.
**Original Message**
:Hello,
: I am running a backup script from a sunos4.1.1 machine. There are about:
:a dozen workstations being backed up using "rsh" or "rsh-n" to those machines.
:The "rsh then does a rdump to a remote drive. The workstations vary between
:Sparc 1+s to a Sparc 10/51. The Sparc 10/51 is runing Solaris 2.3.
:
: The problem I am having is with four Sparc 10/41s running 4.1.3 (Patches
:100444-43, 100726-11, and 100744-01 for OS and OPENWIN). The dump completes on
:the machine, /etc/dumpdates is updated, but the shell process does not
:terminate. If I kill the process, the original script will continue to the next
:partition or machine that is to be backed up. This problem is intermittent, but
:only happens on these Sparc 10/41s running 4.1.3 (all with large swap and local
:memory). This becomes more complicated with the response that Sun Support has
:given stating that they really do not support "rdump".
:
: Any suggestions are appreciated.
:P.S.
: The original script's output is redirected to a file ad the log shows
:that the dump has completed too.
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:08:58 CDT