SUMMARY: Error code 1 when patching on Solaris 10?

From: Christopher L. Barnard <>
Date: Tue Apr 03 2007 - 16:23:47 EDT
I asked:

> I have a modified install_cluster that I built back in the days of
> Solaris 2.6 that deletes patches with an error code 2 or 8 so that when
> the cluster is done, the only patches left are the ones that had issues.
> This worked fine with Solaris 2.6, 7, 8, and 9.  With Solaris 10, sun changed
> the installpatch shell script to a binary.  This should not have mattered,
> but a lot of patches now fail with the error code of 1.  According to the
> patchadd shell script, error code 1 is a patchadd usage error. (!??!!!)
> I rather doubt that Sun would release the vast majority of their patches in
> the 10_Recommended cluster with patch spinning errors.  Can anyone identify
> why about 70% of my patches fail with this error?  Some do error with a
> 2 or 8, so that functionality is still preserved in Solaris 10.  I never
> had an error code 1 on any previous OS installation.
> TIA, and I will summarize.

The answer:

Sun changed the way patchadd works to support multiple zones on a Solaris 10
server.  If you don't have multiple zones on a server (like me) you can
add the -t flag to patchadd to get the former usage from the script.  With
the -t flag added, all is working properly.

Thanks to
Martin Paul <>
Tom Crummey <>

as well as

Michel Nguyen <>
Michael Hocke <>
Rob Windsor <>
brian zuk <>
Anthony D'Atri <>

| Christopher L. Barnard         O     When I was a boy I was told that |
|         / \    anybody could become president.  |
| (312) 347-4901               O---O   Now I'm beginning to believe it. |
|                --Clarence Darrow |
+----------PGP public key available via finger or PGP keyserver---------+
sunmanagers mailing list
Received on Tue Apr 3 16:25:15 2007

This archive was generated by hypermail 2.1.8 : Thu Mar 03 2016 - 06:44:05 EST