Selfmonitoring

CA UIM Self monitoring probe. This probe has been created to do self monitoring of CA UIM Hubs and robots.

Warning this probe only work on hub, the probe do local checkup with nimRequest only (dont use to do remote checkup).

Features

• NisBridge Checkup (With HA Support)
• Probe on hub checkup with HA Support (Callback & Down state).
• Hub robots intermediate and spooler checkup.
• (remote) UMP monitoring
• Distsrv deployment monitoring.
• Hub uptime (when a hub restarted)

Feel free to PR new monitoring

Installation and configuration guide

First of all, dont use nim_login and nim_password if you package the probe. Use these fields when you run the script manually on the system.

Dont forget you need perluim R4.0 framework for this probe. Find the framework HERE

Setup section

Section	Key	Values	Description
setup	domain	string	CA UIM Domain
setup	audit	1 - 0	When audit is set to 1, the probe does not generate new alarms (cool to test in production the first time).
setup	callback_retry_count	number	Number of retries of primary callbacks (getrobots and probeslist).
setup	output_directory	string	the name of output directory.
setup	output_cache_time	number	the cache time in second for output directory.

Configuration section

Section	Key	Values	Description
configuration	check_nisbridge	yes - no	Check Nis_bridge state (Support HA)
configuration	priority_on_ha	yes - no	HA To rewrite 'alarm_on_probe_deactivated' to 1 on every probe (if ha_superiority is set to 'yes')
configuration/alarms	intermediate	1 - 0	Launch alarms when we detect intermediate robot.
configuration/alarms	spooler	1 - 0	Launch alarm when callback get_info fail on one robot spooler.
configuration	check_hubuptime	yes - no	Check for hub restart and generate a new alarm
configuration	uptime_seconds	number	Uptime in second for check_hubuptime

UMP_Monitoring

Setup ump monitoring on your primary and secondary hub. Just put the name of yours UMP in the servers key.

<ump_monitoring>
    servers = ump1robotname,ump2robotname
    alarm_callback = ump_failcallback
    alarm_probelist = ump_probelist_fail
</ump_monitoring>

TIPS : This section is optional

Deployment monitoring

Monitoring of distsrv deployment (checking the numbers of deployment and if deployments time are under a defined threshold).

<deployment_monitoring>
    job_time_threshold = 600
    max_jobs = 5000
</deployment_monitoring>

TIPS : This section is optional

Probes_monitoring

Setup your probes here. Callback is an optional key (no callback is the equivalent of probe down/up checkup). Set alarm_on_probe_deactivated to 0 if you want to not launch a alarm when the probe is offline.

Warning : alarm_on_probe_deactivated is rewrited by HA connected 0 if you have priority_on_ha set to yes.

Set ha_superiority to no if you dont want HA to rewrite the configuration of this profile. (At root level if you dont want to rewrite for all profiles).

Exemples

<probes_monitoring>
    <discovery_server>
        callback = get_device_statistics
        alarm_on_probe_deactivated = 0
        ha_superiority = yes
    </discovery_server>
    <alarm_enrichment>
        callback = getStatistics
        alarm_on_probe_deactivated = 1
        ha_superiority = yes
    </alarm_enrichment>
    <prop_processor>
        alarm_on_probe_deactivated = 1
        ha_superiority = yes
    </prop_processor>
    <nas>
        callback = get_info
        alarm_on_probe_deactivated = 1
        ha_superiority = yes
        <check_keys>
            <pub_subscribers>
                <0>
                    name = NiS-Bridge
                    queue_len = <<100000
                </0>
            </pub_subcribers>
        </check_keys>
        check_alarm_name = checkconfig_nisbridge
    </nas>
</probes_monitoring>

Alarms configuration (not updated).

Alarms message are configurable in the alarms_messages section. Variable are setted in the Script (so refer to this guide to use variables).

Alarms default fields name

Alarms default fields name
origin
domain
source
dev_id
usertag2
usertag1
supp_key
probe
robot
rc

Alarms custom field(s)

Callback	Variables
callback_fail	$callback, $probeName, $hubname
probe_offline	$probeName, $hubname
spooler_fail	$hubname
intermediate_robot	$hubname
nisbridge	$hubname, $robotname, $nis, $ha
distsrv_deployment	$jobid, $pkgName, $started, $diff, $hubname, $robotName
distsrv_maxjobs	$max, $count, $hubname, $robotName
ump_probelist_fail	$robotname, $umpName
ump_failcallback	$robotname, $umpName
hub_restart	$second, $hubName

You can add easily your own variables in the code, just search for callback name. You will find a code like this :

my $probe_offline = $alarm_manager->get('probe_offline');
my ($RC_ALARM,$AlarmID) = $probe_offline->call({ 
    probe => "$probe->{name}", 
    hubname => "$hub->{name}",
    customVar => "toto"
});

Juste add customVar like you want, and in the message use $customVar to print "toto". Feel free to pull-request new variables.