IM was very unresponsive. It seemed to stay in a 'hung' state. Hour-glassing effect noticed. Alarm messages flashing fast updates in the alarm console. Note that it may also take a long time to open up but that can be due to the number of nodes (hubs and robots) its trying to refresh/display.
Checking the IM messages via View->Messages, or checking the nimbus_manager.log the IM showed a get_info probe callback communication error/timeout.
Jun 30 09:49:45:097  Infrastructure Manager: Warning - Request (get_info) has timed out (timeout interval: 4 sec)
Jun 30 09:49:45:098  Infrastructure Manager: get_info - nas - communication error
Jun 30 09:04:02:089  Infrastructure Manager: Warning - Request (get_info) has timed out (timeout interval: 10 sec)
Jun 30 09:04:02:089  Infrastructure Manager: get_info - /NMS/NMS-Server/NMS-Robot/nas - communication error
To resolve this I did the following:
1. Clicked the "Hold Alarms" button in the alarm sub console just so i could work more easily with the IM without delay/hour-glassing/hourglass effect.
Note that you must allow it to catch up and normalize before you can start working with the IM again. This may take up to 10 minutes depending on alarm traffic/volume and the number of robots/nodes in your environment.
2. Open the nas GUI and select the Status window
3. Rt-click and choose Advanced->Reorganize nas alarm database. Wait until its done - you can check the nas.log.
(You might see messages similar to "nas: NAS is reorganizing database, the request ('set_visible') is denied." This is normal. It just indicates that nas is still reorganizing database and it cannot work on the request mentioned in the message.)
4. Via Raw configure I set the nas bulk_read_size to 500.
5. Once completed I re-enabled the live alarm refresh by deselecting the "Hold Alarms" button.
and the IM was now responsive with no further significant delays.
Note that you can also disable "Attempt Internet archive integration" to speed up IM response.
Note: On installations where this web connection can not be established (e.g. due to firewall configurations), a timeout (and an error message) will appear each time the Infrastructure Manager application is started. Turn off this option to avoid this timeout.
Note that any changes made while the IM is not responding may take some time to take effect.
Bottom line is right now the best way to make IM respond is to keep the Hold Alarms button depressed until you're finished with the rest of your work.
Note that the IM "probe Request" timeout value can also be adjusted if you notice IM Messages in the alarm sub-console Messages window related to timing out.
***Also keep in mind that you should not have any alarms hitting the IM over and over again needlessly. If you check the alarm count for an alarm and its very high e.g., 11,000 - its a good candidate for either a change in threshold so you don't get flooded with unnecessary alarms or its time for a pre-processing rule to exclude the alarm if it's just 'noise' and the alarm cannot currently be avoided/eradicated at its source.
Please Update This Required Field