[1923] in SIPB-AFS-requests
Re: running /usr/local/etc/elcsd on charon,rtfm,rosebud,ronald-ann
daemon@ATHENA.MIT.EDU (mhpower@MIT.EDU)
Thu Apr 6 22:52:43 1995
From: mhpower@MIT.EDU
To: jweiss@MIT.EDU
Cc: charon-maintainers@MIT.EDU, rtfm-maintainers@MIT.EDU, sipb-afsreq@MIT.EDU
In-Reply-To: "[1922] in SIPB-AFS-requests",
"[1232] in Charon_Maintainers_Archive",
"[3784] in RTFM_Maintainers_Archive"
Date: Thu, 06 Apr 95 22:50:54 EDT
>This bug was fixed in ultrix 4.3. I have built a 4.3 binary (with the
>athena patches). The binary is running on a couple of machines and
>has survived satan attacks.
As far as I can tell, the elcsd in Ultrix 4.3 and 4.4 still accepts
tcp connections on port 704 regardless of whether LOGREM is set in
/etc/elcsd.conf (which I seem to have written as /etc/elscd.conf in my
previous message). I think it's cleaner to just not listen on tcp port
704 if no one is supposed to be sending traffic there (LOGREM would
typically be set only on central elcsd servers that intentionally
receive information via tcp from other machines -- this is presumably
the reason for the "sed -n -e 's/^5/3/'" command in the ops.add
script). Given that we'd be running a version of elcsd different from
the standard release anyway, I'd prefer to stay with the version I
built. For example, unless someone has read the 4.3 source code, we
really don't know if sending some particular data to port 704 might
again trigger the cpu-time problem or some other problem...
Matt
