[6941] in Release_7.7_team
Re: Did the install-debathena.sh script change?
daemon@ATHENA.MIT.EDU (andrew m. boardman)
Fri Sep 3 16:45:52 2010
Message-Id: <201009032045.o83KjeQC007018@pothole.mit.edu>
To: Myron Freeman <fletch1@eecs.MIT.EDU>
cc: Jonathan Reed <jdreed@MIT.EDU>,
"release-team@mit.edu" <release-team@MIT.EDU>,
"debathena@mit.edu" <debathena@MIT.EDU>, Oliver Thomas <othomas@MIT.EDU>,
amb@MIT.EDU
In-Reply-To: Your message of "Fri, 03 Sep 2010 16:27:30 EDT."
<921D4FFE-2F02-4B92-92C0-FED3491EAD19@eecs.mit.edu>
Date: Fri, 03 Sep 2010 16:45:40 -0400
From: "andrew m. boardman" <amb@MIT.EDU>
> Does the install-debathena.sh script log to a file someplace? I might
> be able to figure out something from that.
It does not by default (though it's high on the "to do" list), but if you
ran, say, "./install-debathena.sh 2>&1|tee /tmp/dainstall.log" I think
we'd all be very interested in seeing the output.
Since what the install script is fundamentally doing is setting up the
new apt repository and then running approximately the same aptitude
command which works when run by hand, the failure is quite curious.
We've seen something similar when an implied dependency of one of the
packages we install was missing (causing an overall failure of the
overall metapackage install the first time around, but since something
else along the way had required the dependency it worked the second time
around), but are not aware of any current problems.
One more question: what's your primary apt repository on these machines?
(Basically, in /etc/apt/sources.list, what's the URL in the
"deb <url> lucid main restricted" line which is probably the first line
in the file that's not blank or a comment?)
andrew
