|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |
Mime-Version: 1.0 (Apple Message framework v1084) Content-Type: text/plain; charset=us-ascii From: Jonathan Reed <jdreed@MIT.EDU> In-Reply-To: <alpine.GSO.1.10.1104121750030.19944@multics.mit.edu> Date: Tue, 12 Apr 2011 17:54:11 -0400 Cc: source-commits@MIT.EDU Message-Id: <3228E2A1-1C65-4B32-B60B-8772A86BD45E@MIT.EDU> To: Benjamin Kaduk <kaduk@MIT.EDU> Content-Transfer-Encoding: 8bit >> CLUSTER= >> -eval "$(getcluster -b $(lsb_release -sr))" >> +if [ -s /var/run/athena-clusterinfo.sh ]; then >> + . /var/run/athena-clusterinfo.sh >> +fi > > Does there want to be an 'else' branch here? > > -Ben I don't think so. Unless we want to run getcluster for every athinfo query, but that seems wrong. Basically, if clusterinfo isn't there, it'll fall back to looking at metapackges. If we think the need to remote-diagnose a machine with broken clusterinfo and broken metapackages outweighs the loss from running "getcluster" on every cluster-only athinfo query, we could do that. Or we could run save-cluster-info, I guess. -Jon
|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |