Hi,
Please check the data retention for Device Events and number of Device events present in AirWave under System > Syslog and Traps. This should give an idea if the Device Events were never purged or not.
All of upgrades, fresh install includes recompilation of code. So, if there are stuck processes services which are down etc , the recompilation code will restart them and you may see some cleanups happening (if they are previously stuck).
Vacuuming is done if there are any bloated tables, if they are not bloated much, you wont see much difference with vacuuming them. you should be able to see the big tables under Database statistics under System > Performance page. Also, Check the AMON statistics and RabbitMQ graphs to see any anomalies over the period and if the in/out processing rate has huge difference.
------------------------------
Regards
Gowri Amujuri
------------------------------
Original Message:
Sent: Aug 03, 2023 01:27 PM
From: cbjohns
Subject: Airwave Disk Utilization Increase/Spikes after upgrade/rebuild to 8.3.0.1
Hi Gowri,
Thanks for checking. Looks like nightly maintenance based on logs is about 38 minutes. The interesting thing I've observed is that table scans, row deletions/updates/adds - all seem consistent graph wise. But the one funny observation is the "Device Events" dropped post-upgrade and haven't climbed - which made me think maybe Airwave was cleaning up/deleting rows more often - but then i would have figured I'd see that reflected in the other graphs. TAC's going to help with some vacuuming next week. Just very peculiar.
Original Message:
Sent: Aug 03, 2023 09:26 AM
From: Gowri Amujuri
Subject: Airwave Disk Utilization Increase/Spikes after upgrade/rebuild to 8.3.0.1
I do not see the issue on a lab server with 100ish devices. The Disk utilization is same as before.
How long is your nightly maintenance running post upgrade?
------------------------------
Regards
Gowri Amujuri
Original Message:
Sent: Aug 02, 2023 04:09 PM
From: cbjohns
Subject: Airwave Disk Utilization Increase/Spikes after upgrade/rebuild to 8.3.0.1
I was curious if anyone else has seen the same behavior. We successfully went through the upgrade process of migration to 8.3.0.1, and have noticed our Disk Utilization is now spiking every 20 minutes or so [believe this partially correlated to SNMP fetches]. We've consulted with TAC, but they've advised that the current software recommends 256GB RAM [at 128GB as that was recommendations back during 8.2.5 believe] per our device count of approximately 3500 devices. They've made some adjustments, but haven't noticed any difference. Just seems odd/peculiar - even with the switch from Centos 7 TO RHEL - but wanted to see what others have been. Thank you.