Documentation—clusterlogin
NAME
clusterlogin - log into a cluster node with most available resources
SYNOPSIS
The clusterlogin utility polls all nodes and, based on a user-defined resource, logs the user via rlogin, or however else you like, into the highest ranking node for a given benchmark.
Logging in Based on Load Average
By default, the clusterlogin will log the user into the least loaded node, as calculated by the 1 minute load average available from /proc/loadavg.
> clusterlogin Punching into best host for resource [load 0.00]: 0of2 Executing rlogin 0of2 Last login: Tue Oct 9 22:07:45 from 4of0
Logging in by CPU, Memory and Number of Users
You can log into a node with the most responsive CPU, fastest memory, I/O or fewest number of users using
> clusterlogin -s "cpu" Punching into best host for resource [cpu 0.545866]: 8of0 Executing rlogin 8of0 > clusterlogin -s "mem" Punching into best host for resource [mem 0.140143]: 1of0 Executing rlogin 1of0
> ./clusterlogin -s "io" Punching into best host for resource [b_io 0.630308]: 0of3 Executing rlogin 0of3
> clusterlogin -s "nusers" Punching into best host for resource [nusers 0]: 4of1 Executing rlogin 4of1
Adjusting Timeout
> clusterlogin -t 15
Setting Port and/or Broadcast
> clusterlogin -p 8095 -b 10.1.2.255
DESCRIPTION
This is part of the clusterpunch system. Each node that you wish to monitor must be running the clusterpunch.shutdown daemon, documented in the clusterpunch.shutdown. Once the server is running, you can poll it and send commands using the API documented in clusterpunchclient.pm, or through the various utilities, such as this one.
SEE ALSO
Daemons
clusterpunchserver, clusterpunch.start, clusterpunch.shutdown
API
clusterpunch.pm
Utilities
benchdriver, clusterbench, clusterlogin, clusternodecount, clustersnapshot
CHANGES
AUTHOR
Martin Krzywinski (martink@bcgsc.ca)
$Id: clusterlogin,v 1.6 2003/01/26 08:03:56 martink Exp $