Included probe: QueueCheck 2.9.9.2 can (try to) help you with:
- monitor all Hub and Nas queues in your environment
- calculate message rate between polls
- send alarms when thresholds' are breached
- send alarm when GET queue is not connected
- send alarm when ATTACH queue has no consumers
- clear alarm when needed
- send email when thresholds are breached
- more granular configuration settings compared with the old LUA script
- create qos metrics to create dashboard overviews, CABI/Jaspersoft report, Metric View or PRD report
- sample dashboards are included
Sample start dashboard:
Sample dashboard: Overview (with queue connect status, queueing last status, max/min last hour, qos rate last status, min/max last hour, # polls and Lastpoll date&time). If the lastpoll is more that 60 minutes ago it will be colored orange.
This view should give you all important information in 1 screen.
Sample dashboard: hub status: (with last status, max last 6/12 hours, uptime, version, #polls and lastpoll)
Sample dashboard : Hub & Nas queues (with last status number metrics queued, max last 6/12 hours)
Sample dashboard: queue rate (number of metrics processed by each queue since previous poll)
Sample dashboard: queue bulk_size: (hub queue setting that can influence queue throughput)
Note1: This probe is written in Perl, but packaged with Perl source and compiled version. By default the probe will be started with teh compiled version (see doc what to change to run teh Perl source)
Note2: There is also a LUA queuecheck script available that reports/alarms only on the queued thresholds. This LUA script is still a valid solution if you want to have a **** solution with fewer customization possibilities and less qos metrics.
File: QueueCheck_2.9.9.2.zip is the probe to import into your archive
File: QueueCheck_utilities_2.9.9.2.zip contains the doc file, dashboard zip file and example files
New in 2.9.2:
- when a hub is started recently, the probe will wait 1 interval (for this hub) to avoid false alarms for probes that are starting
- new QoS: qos_queuecheck_diff, this metric will contain the number of alarms/qos processed by each queue in the previous probe interval. (activated by config: qos_on_diff)
- extra ci_type for the new qos metric in file: doc_ci_type_add.txt
- logging is reworked so that with loglevel=0 you can clearly check what the probe did
- new config parameters: hub_exclude and nas_exclude (in regex format)
- create (optional) alarm when an ATTACH queue has no consumers (activated by config: alarm_level_attach_no_consumer) at queue level
- create (optional) alarm when a GET queue is not connected (activated by config: alarm_level_get_not_connected) at queue level
- default config file can be used as an example, it will not send emails and all alarms are set to level: 1 (INFO) (only if nas/hub are inactive an alarm will be send)
- new version of mail script: nas_mail_queued.bat
New in 2.9.3:
- new parameters: get_queue_exclude and attach_queue_exclude to add exclude filters
- alarm_on_diff and alarm_level_diff_queue to define alarming on # processed_alarm/qos in last probe interval
- probe will now recognize a (remote) hub restart to calculate the difference in qos/alarm processed in the last probe interval
- add a second dashboard example based on qos_queuecheck_diff metrics
New in 2.9.4:
- add parameters: request_timeout and request_retry used in the UIM callbacks
- add parameter: log_subscriber_info to list of used subscribers for each hub
- parameters: get_queue_exclude, attach_queue_exclude and the queue section name parameter are now matched against the UIM address (name is now also regex)
- add parameter check_regex that can validate your regex parameters
New in 2.9.5:
- add queue section parameters: ignore_alarm_time_start and ignore_alarm_time_end. This gives you the possibility to suppress alarms during a time period, example during backup of the CA_UIM database
- add in log file the occurrence # of each problem
- add parameter display_clear_in_log; this will add a log entry for each clear that is executed
- when starting the probe a sql query will obtain the open alarms for probe queuecheck, from the nas_alarms table, and will acknowledge these. (instead of trying to execute a clear for every possible alarm)
- when monitoring the queue type TEMP for the number of queued qos/alarms, the alarm will contain the id and subject of the temp queue to help you finding the problematic queue.
- added 3 starting scenarios in the doc and utilities zip file.
New in 2.9.8:
- add qos: qos_queuecheck_rate (message rate between polls) activated by: qos_on_rate
- add qos: qos_queuecheck_count (total messages processed since start queue) activated by: qos_on_count
- add qos: qos_queuecheck_bulksize activated by: qos_on_bulksize
- add qos: qos_queuecheck_hubstatus: activated by: qos_on_hubstatus
- add qos: qos_queuecheck_hubuptime: activated by qos_on_hubuptime
- add sample dashboards with a _start_queuecheck dashboard
- add qos: qos_queuecheck_hubversion: activated by qos_on_hubversion
New in 2.9.9:
- add qos: qos_queuecheck_connect activated by: qos_on_attach_connect and qos_on_get_connect
- add new Overview dashboard with all important qos metrics on 1 page
- changed probe package to start with compiled (Windows) version & a example cfg file that is ready to be used. (you only need to configure, if not done previously, the file: nimsoft_generic.dat with the uim, sql access details)
New in 2.9.9.1:
- alarm suppression key was not always filled in correctly
- changed dashboard color green so that digits are more readable
- uim 23.4 changed the format of the hub version
- excluded nas queues from the "rate" dashboard because nas rate is not available
New in 2.9.9.2
- Several small changes to have more status info on the dashboards.
- new cfg parameters (attach_inactive_if_hub_inactive & qos_on_nas_connect) to create a more detailed dashboard view
- added lastpoll on the main dashboard views so you can see/check when a problem started
Problems, remarks, suggestions, ideas and enhancements are very welcome
#queuecheck #tool #probe #qos #monitor #threshold #alarm #mail #dashboard #hub #nas #queue #check