[Planetlab-users] suspended slices

[Larry Peterson]

Many of you are now (painfully) aware that your slices
are suspended if your nodes have been down for an extended
period of time, and the PlanetLab support staff has had
no luck in prodding someone at your site to fix the problem.

To help you understand what state PLC thinks your nodes are
in, so you can be proactive in keeping your nodes healthy,
you just need to look at the same information we do, which
means querying CoMon. CoMon is generally available at

   http://comon.cs.princeton.edu

but for specific queries...

Nodes that are up:
http://summer.cs.princeton.edu/status/tabulator.cgi? 
table=table_nodeviewshort&select='resptime%20%3E%200'

Nodes are sick (we nag, but don't suspend -- yet):

Bad DNS:
http://summer.cs.princeton.edu/status/tabulator.cgi? 
table=table_nodeview&select='dns1udp%20%3E%2080%20&&%20dns2udp%20%3E% 
2080'

Clock drift:
http://summer.cs.princeton.edu/status/tabulator.cgi? 
table=table_nodeview&select='drift%20%3E%201m'

Bad disks:
http://summer.cs.princeton.edu/status/tabulator.cgi? 
table=table_nodeview&select='filerw%3E0'

If your nodes aren't on the above lists, then we treat
them as down. To see the set of down nodes:
http://summer.cs.princeton.edu/status/tabulator.cgi? 
table=table_nodeviewshort&select='resptime%20==%200'

Larry