EWC - V9.21.15 - AP3620 reboots unexpected (if users generate too much traffic )

[M_Nees]

My customer complains that a AP3620 reboot only based on the user load. Is this possible ??

I use EWC V9.21.15.

ap.log tell me that the regarding AP has bootet - but which trace file or log message tell me the reason for the last boot ?

Please help.

Regards

[Ronald_Dvorak]

Go to GUI > Logs > AP > Traces
Select the AP and click on retrieve traces.

If the controller is in a good mood that should download the AP trace file.
Open the .tar file and you'll find ap.logLastReboot.gz and one or more files with the name ap-report.txt*.gz.

Might be that you'd find the cause in this files or you'd need to open a ticket so the GTAC could take a look in the incl. dump.

BTW, does the > Logs > AP > Logs > Severity:All .... doesn't show the reboot cause ?

[M_Nees]

You are right getting Trace Files from AP is not very reliable... i need 3 tries ...

unfortunately ap.logLastReboot.gz is corrupted but ap-report.txt tell me something:
Jan 1 00:00:52 kernel: kTrapMonitor: process cap is missing (fatal) -- reboot in [5+ sec]

As you see below is decide opening a case.

Regards,
Matthias

[Ronald_Dvorak]

I'd upgrade to v10, there are no known bugs in v10 for AP26xx and AP36xx.

[M_Nees]

Good joke ....