Hello Managers,
My origianl question...
Hardware : Enterprise 450
O/S : Solaris 2.6 Unpatched.
Netscape enterprise 3.0 server, unpatched.
Dear Managers,
We seem to be experiencing problems stopping and starting the
Netscape enterprise 3.0 server process.
We are running the daemon chrooted into the document root directory,
which ( after alot of messing around with paths ) works fine, but when
we execute the stop script, only one of the two server processes dies,
the other is left in the process table, and will not respond to a kill,
or even a kill -9. Trying to start the server processes again in this
state, unsurprisingly, yields the 'address already in use message' . The
only solution to this seems is a reboot....Hmm, that's bad news, because
my average uptime for this machine is only a few days, but all my other
machines manage at least 100 days. I've looked at the patch descriptions
for both the Netscape enterprise server, and Solaris, and none of them
relate to this kind of problem...( though if anyone knows any better,
let me know....!!!)
Even though there is more mention of anything releated to this in the
recommend patches docs,
the fix was to install 2.6 Recommended patches, and upgrade to
enterprise 3.0k.
I noticed that the parts of the kernel and libthread.so.x are patched, I
suspect that these were the cause of the problems, since the jumbo patch
was mostly patching stuff that I don't use or need.
Kind regards,
Simon.
Thanks to Nicholas Misika.
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:39 CDT