We have multiple hubs configured in our environment with multiple robots for each hub and cdm probe deployed for them.
USM shows data for the robots under the primary hub, but if we select a server on a different hub we see "No Metrics Available". These metrics are visible in Infrastructure Manager, though.
The configuration is like this:- All servers have a robot installed and their hub is either the primary hub or another hub
- "Secondary" hubs have installed :baseline_engine 2.76 3 Baseline Enginecdm 6.20 170 CPU, Disk and Memory performance probehub 7.93 8390 Message Concentrator and Redistributorjava_jre 1.80 15 JRE for Windows and Linux platformsntevl 4.24 150 Windows NT Event Log watcherppm 3.47 1707051613 Probe Provisioning Managerprediction_engine 1.34 1 Prediction Enginerobot_update 7.93 9350 Updates Nimsoft Robots to current versionvs2008_redist_x64 1.00 01 Visual Studio 2008 SP1 redistributables for x64vs2008_redist_x86 1.00 01 Visual Studio 2008 SP1 redistributables for x86vs2010sp1_vcredist_x64 1.01 1 Visual Studio 2010 SP1 redistributables for x64
- The robots have:cdm 6.20 170 CPU, Disk and Memory performance probentevl 4.24 150 Windows NT Event Log watcherrobot_update 7.93 9350 Updates Nimsoft Robots to current versionvs2008_redist_x64 1.00 01 Visual Studio 2008 SP1 redistributables for x64vs2008_redist_x86 1.00 01 Visual Studio 2008 SP1 redistributables for x86vs2010sp1_vcredist_x64 1.01 1 Visual Studio 2010 SP1 redistributables for x64
Am I missing something? Is there something more to be installed/configured?
Is your Primary HUB box able to connect to 480XX ports in robots under non-Primary hubs ?
Yes, telnet is working, Infrastructure Manager is working.
Ports 48000-48015 are allowed on firewall. Cdm has 48008 and telnet on that port is working.
Even more, one of the non-primary hubs is in the same subnet as the primary hub, with no firewall between them. Cdm was installed on it before it became a hub and there were metrics available in USM. This changed when it became a hub and there are no more metrics available.
It looks QOS message (from robots) not arriving at Primary HUB.
As Luc said, check HUB queue.
If you take one of those secondary hub configuration and under the status tab:
- do you see an attach queue for subject QOS?
- is there a queue that has an amount queued?
In Infrastructure Manager, in hub probe configuration
- on Status->Subscribers/Queues tab I have:
-- baseline_engine; Id baseline_engine
-- prediction_engine; Id prediction_engine
-- t_4 ; Id discovery_server
You need to create on each secondary hub at least 2 ATTACH queues:
- 1 for subject: alarm
- 1 for subject: QOS_MESSAGE,QOS_DEFINOTION,QOS_BASELINE
On your main hub you will need 2 extra GET queues for each secondary hub to get those 2 attach queues.
Yes, that was it.