[2977] in SIPB bug reports
Re: Zephyr-to-BITNET gateway broken
daemon@ATHENA.MIT.EDU (mhpower@stan.mit.edu)
Mon Jun 22 13:47:15 1992
Date: Mon, 22 Jun 1992 13:43:53 -0400
From: mhpower@stan.mit.edu
To: marc@Athena.MIT.EDU
Cc: bug-sipb@Athena.MIT.EDU
In-Reply-To: [2975] in SIPB bug reports
>This is a MITVMA problem, and there's nothing we can do about it?
I think so. In this case, connectivity was restored at 4 A.M., which
is a restart time for some of the MITVMA network software. Also, the
MITVMA maintainers manually monitor the connection on at least a daily
basis, so they might sometimes restart from their side at some random
time, too.
>Is there anything we can do when shutting down cleanly (which was not
>the case this time; console logins were failing) to avoid this
>problem?
I believe this time charon just crashed with "panic: protection
fault"; this certainly would prevent console logins...
There are two possibilities: either MITVMA isn't accepting any NJE
line restart requests, or it's not accepting MITSIPB's request since
it's confused about the state of the line.
If the latter, yes, it might help if we can try to shutdown our side
before rebooting charon.
If you're able to login as root, run /usr/local/nje/shutdown-both. If
you're not able to login, but think Zephyr might still be working, you
can try doing a
bwrite njed_ctl@mitsipb
and put some reason for the shutdown in the text of the message (if
you want; this is just for the log, it doesn't try to parse your
reason and haggle with you). I've added marc.root@ATHENA.MIT.EDU to
the acl for this.
If charon recovers, you should login as root and run
/usr/local/nje/restart-both. Because of this, no one should be on the
Zephyr shutdown acl who doesn't have root access anyway.
In any case, it's probably not a good idea to do a shutdown as a test,
since there's no guarantee that we'll be able to reconnect right away.
Matt
