[1353] in Moira
Re: Moira Daily Report
daemon@ATHENA.MIT.EDU (K. Ramm)
Tue Feb 15 17:09:14 2000
To: Garry Zacheiss <zacheiss@MIT.EDU>
Cc: moiradev@MIT.EDU
From: kcr@MIT.EDU (K. Ramm)
Date: 15 Feb 2000 17:09:10 -0500
In-Reply-To: Garry Zacheiss's message of "Mon, 14 Feb 2000 18:30:15 -0500"
Message-ID: <uus3dqu86ex.fsf@EMACS-MAKES-A-COMPUTER-SLOW.MIT.EDU>
I mostly avoided this problem in hesiod.sh with a
exec >/tmp/moira_update.log 2>&1
near the beginning so that the output went to a file rather than the
console. It made remote debugging easier, too. Maybe we should do that in
all our update scripts.
kcr
Garry Zacheiss <zacheiss@MIT.EDU> writes:
> Hi Tom,
>
> This showed up in our mail this morning:
>
> >> **** dcm is still running
> >> ** moira3.43928 IMAGERY.MIT.EDU.moira_update 8760 0 8760 0 ESTABLISHED
> >>
>
> Moira's nightly update process appears to have hung last
> night while talking to imagery. On the Suns we maintain, we tend to see
> this mode of lossage when a Sun is booted without a keyboard/monitor so
> that it's using a serial console, but fails to have its serial console
> attached for a long time, causing further attempts to write to
> /dev/console to hang. Could you take a look at imagery for us?
>
> Garry
