Yes,
Very good response. I spend a lot of time on the database looking for my "lost" metrics.
At the same time, don't forget to check that your probe sends the metrics (Dr Nimbus). Small basic check to be sure!
================== If your database queries show you anomalies ==================
Then, if you don't use the "Discovery" functions of DX IM, it is possible that your "server" or "equipment" for which you were trying to perform a port check, may not be "completely" declared at the level of these "Correlation names" (complex mechanism managed by the Discovery Server probe).
- The net_connect probe, declares the "device" and the "configuration items" of your <Port check> profile, into the "niscache" folder of the agent where the probe is installed. "Discovery Server" will regularly scan this niscache folder to update the database for Device/Correlation Items/Metrics relationships.
- This mechanism may take a long time (don't hésitate to set the log level of the Discovery Server probe to 2, and the log size to "102400" (100MB). After an hour, if your metrics are still not apparent on your target equipment, search in the log file for the name, alias and address IP of your target. Maybe you will find some conflict or error information.
Sometimes, it is necessary to manually declare all the IP addresses of a device, as well as these possible DNS aliases. Ask about "cm_data_import" (Or via UIM API if you are familiar with API and Swagger UI).
A useful tip: rather than redoing your SQL queries each time to diagnose possible conflicts, metrics problems (...) do not hesitate to develop "Dashboards", and integrate your queries with variables and drop-down menus ect ect. If you wish it, do not hesitate to ask me, I have a lot of Dashboard of diagnosis hand made (^-^)/°
Best regards,
Jonathan
------------------------------
Lead Monitoring Solution Architect
AXA Investment Managers Paris
------------------------------
Original Message:
Sent: 11-20-2020 05:39 AM
From: Luc Christiaens
Subject: port check profile is configured, but ping/smtp metrics are not displayed/found in OC
OC in 20.3.1 is not very reliable in displaying all info (in my tests)
I would use sql queries to see if qos metrics are generated
or use a tool like: https://community.broadcom.com/enterprisesoftware/communities/community-home/digestviewer/viewthread?MessageKey=bdac9084-5e91-4265-8942-2c85880bce8e that uses also sql queries to give all info, but offers via switches a lot of selection possibilities.
Original Message:
Sent: 11-20-2020 04:00 AM
From: Stefan Pivoda
Subject: port check profile is configured, but ping/smtp metrics are not displayed/found in OC
although i configured a Port Check profile via OC 20.3.1 and selected ping and SMTP inside it for a device that we would like to monitor remotely, i cant see any metrics for it at all if i open the device from the inventory menu in OC.
if i open the net_connect probe in Admin Console, i can see that both the "Monitor ICMP connectivity" and the SMTP protocol is activated and publish data is selected.
Should i see these metrics (ICMP, SMTP) OOTB somewhere in OC ? if yes, where please? if no, what's the suggested way to display them?
thank you,
stefan