centreon / centreon-plugins

Collection of standard plugins to discover and gather cloud-to-edge metrics and status across your whole IT infrastructure.
https://www.centreon.com
Apache License 2.0
312 stars 275 forks source link

Pacemaker unable to ignore resources or set counter threshold #2588

Closed ghost closed 3 years ago

ghost commented 3 years ago

Hi,

The current plugin cannot ignore a resource, a status or an counter.

Example: -bash-4.2$ /usr/lib/centreon/plugins//centreon_pacemaker_ssh.pl --plugin=apps::pacemaker::local::plugin --mode=crm --hostname='10.10.41.17' --ssh-backend='libssh' --ssh-username='xxxx' --ignore-failed-actions='fence' --verbose CRITICAL: Resource 'xxxx01' is stopped - Resource 'xxxx02' is stopped WARNING: xxxx01 failure detected, fail-count=1000000 - xxxx02 failure detected, fail-count=1000000 - xxxx01 failure detected, fail-count=1000000 - xxxx02 failure detected, fail-count=1000000 | Resource 'lv_xxxxclusterdisk01' is started on node 'xxxx04' Resource 'fs_xxxx04clusterdisk01' is started on node 'xxxx04' Resource 'fence01' is stopped Resource 'fence02' is stopped Resource 'aaa14' is started on node 'xxxx04' Resource 'bbb14' is started on node 'xxxx04' Resource 'ccc14' is started on node 'xxxx04' Resource 'ddd141' is started on node 'xxxx04' Resource 'eee142' is started on node 'xxxx04' Resource 'fff143' is started on node 'xxxx04' Resource 'ggg144' is started on node 'xxxx04' Resource 'hhh' is started on node 'xxxx04'

Expected result: Ignoring fence resources and set threshold on fail-count counter

garnier-quentin commented 3 years ago

I think i will update that mode.

bmareau commented 3 years ago

Hi If we can add this enhancement request also :

In centreon_pacemaker_ssh.pl (line 554), you manage the case if only if the crm displays (unmanaged) FAILED as shown in your part of the code below :

elsif ($line =~ /\s*(\S+?)\s+ \(.*\)\:\s+\w+\s+\w+\s+\(unmanaged\)\s+FAILED/) {
# Check Unmanaged
$self->{output}->output_add(severity => 'CRITICAL',
short_msg => "$1 unmanaged FAILED");
}

But I have an application that displays the following state in my crm :

postgresql-9.4 (lsb:postgresql-9.4): FAILED sofit-pgsql1.grdf.fr (unmanaged)

FAILED at the beginning and then (unmanaged) at the end is not managed by the plugin. In your plugin it is (unmanaged) at the beginning and then FAILED at the end.

Regards

garnier-quentin commented 3 years ago

You can test with last git version. Now there are many more options!