[2425] in Release_7.7_team
Re: Minutes of 2000-09-13 release team meeting
daemon@ATHENA.MIT.EDU (Robert A Basch)
Thu Sep 14 16:32:02 2000
Message-Id: <200009142031.QAA196430@aupair.mit.edu>
To: Greg Hudson <ghudson@MIT.EDU>
cc: release-team@MIT.EDU
In-Reply-To: Your message of "Wed, 13 Sep 2000 13:57:00 EDT."
<200009131757.NAA27579@small-gods.mit.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 14 Sep 2000 16:31:57 -0400
From: Robert A Basch <rbasch@MIT.EDU>
> portmap has been dying on some of our SGI machines, probably due to
> scans. This causes fam error messages in the console and also
> prevents Frame from running properly. There is some speculation that
> it might be internal net-security scans which are killing portmap, but
> we don't know that for sure.
>
> We don't have a fix yet, but SGI's initial response is that it might
> be a known bug. We'll see.
An update...
We've confirmed that it was a scan by net-security that killed portmap;
I've relayed the relevant info to SGI.
I downloaded the latest maintenance release (6.5.9) from SGI, and found
that the problem does seem to be fixed there. Unfortunately, the fix seems
to be in the C library, not the portmap executable itself. Also, running
portmap with the "-A" option, to restrict access to local clients only,
does *not* seem to be a workaround. So, getting a patch from SGI seems
to be the only feasible solution; I've been told that such a patch will
be forthcoming.
Bob
