Beside setting up the collector configuration file, more things have to be configured to enable all functions of the ParaStation GridMonitor.
The collector requires a valid license provided in the file
/etc/pscollect/license
.
Licenses may be obtained from
<support@par-tec.com>
.
All remote agents, launched by the collector, will by default be
started using ssh.
Therefore, the user pscd
on the collector
node must be able to login to each cluster node without
providing a password or pass phrase.
To enable this "auto-login", a ssh
key,
e.g. ~/.ssh/id_dsa.pub
, of user
pscd
on the collector node must be added
to its own ~/.ssh/authorized_keys
file and the
authorized_keys
of this user on all other
nodes.
Typically, BMCs share one of the network interfaces with the
server.
Quite often, the BMC is only reachable from other nodes, not
from the local host itself.
To circumvent this problem, the collector may be advised to use
the local kernel interface to read data from the local BMC.
To do so, the special name localhost
must
be used as BMC name when calling the
impi_host()
mapping function.
Refer to the section called “Configuring the collector – step 6” for details.
To enable the collector to read data for the local node using the
command ipmitool, the collector must be able
to access the IPMI device /dev/ipmi0
.
In order to do so, the following lines must be added to the
file /etc/sudoers
:
Defaults:pscd passwd_tries=0, !syslog pscd ALL=NOPASSWD: /usr/bin/ipmitool -A none \ -I open -H localhost [a-zA-Z/]*, /usr/bin/ipmitool -A none -I open \ -H localhost -S [a-zA-Z /]*
Example 4.22. Enabling access to local IPMI data
To enable the collector to read and monitor disk parameters using
S.M.A.R.T, the command smartctl, launched
by the agent process, must be able to access the raw disk
devices, like /dev/hda
. To do so, the
following lines must be added to the file
/etc/sudoers
:
Defaults:pscd passwd_tries=0, !syslog pscd ALL=NOPASSWD: /usr/sbin/smartctl \ -[iAH] /dev/[sh]d[a-z], /usr/sbin/smartctl \ -d ata -[iAH] /dev/[sh]d[a-z]
Example 4.23. Enabling access to disk parameters
This has to be done on all nodes of the cluster.
For particular disk controllers or drives, the smartctl may hang your system. Please test carefully! Not all drives or controllers support S.M.A.R.T.