Strange CPU usage on AP

CPU graph are not avaible on web UI.  CPU load can be measured via SNMP.

And  new strange statistic...

I have  two sectors on the site (different frequencies). Both now in sync 75/25 mode. 

Yesterday i  changed GPS sync source  to internal one of them (2). No more any cpu spikes...

CPU load of both sectors, and traffic graph:

 cpu-1.png

cpu-2.png

traf-1.png

traf-2.png

Please test on lab...

We have 22 ePMP gps-sync APs  (2.4 and 5Ghz) using the the little GPS puck and every one of them exhibit this behavior. We assumed it was a graphing glitch since we haven't noticed any performance issues.   We also have one we just set up and even though it doesn't have a single SM connected to it yet it still has these spikes every few minutes except it only spikes to around 92% instead of 100%.

Hello Cambium Support is possible to do something with this case? In our monitoring all CPU peaks ( caused probably by GPS ) they have impact to time response (ping) for our customers. Thank you for any progress ....

How many SMs (maximum) on your 5GHz sectors? 

Maximum 25 SMs on sync AP,  mostly 10-15 SM per AP

Any progress?

Hello Pavel,

Issue has been opened and it is under investigation now.

Thank you.

Just adding this. This AP is running 2.5.1 and it doesn't have a single client radio on it.  It, like all our AP's, uses the GPS puc for GPS sync.

New FW v2.5.2.  No progress.

Hi all,

After investigation was found that root issue is incorrect CPU load reporting.

There is no influence on performance.

It will be fixed soon.

Thank you.

This is still occuring in ePMP release 3.4.1 when polling via SNMP.  Yes, we realize the number needs to be divided by 10 to get the usages with one decimal point.   We will get spikes where the raw return values are 1000 for a number of minutes. 

This can be on AP's with or without clients for us.

Any progress??

Same issue here. 3.4.1 with no clients. 100% spikes for minutes. ePMP 2000 and ePMP 1000.

same here, 1 client no traffic and cpu got 98% at 3.4.1 firmware please help

I'll test CPU usage for some days on internal sync source mode. I suspect that the CPU usage spikes are caused by GPS sync procedure.

error code 0xc00000e9

I was researching this issue this morning, and found this thread. We're running 3.5.1 on an ePMP2000 AP. I've attached a screenshot from our onPrem cnMaestro 2.2.1 which includes a graph of CPU. The CPU reporting issue seems to still be present in 3.5.1

is there any progress on this case? . because I experienced the same thing

Cambium has recommended newer firmware in the 4.x series in general, but as far as I know not specifically for this issue.

I see 4.5 is the newest currently. Which firmware @arifdf are you on for the graph you posted?

CPU usage is generally all over the map.

This we found is tied at least partially to how many packets have to be processed through priority queues. So if you have traffic priority enabled on the AP or an attached SM then you will see the CPU usage looking weird. Another cause for wonky CPU usage is SM's that are re-registering or lots of retransmissions.

We have also noticed that the CPU usage graph tends to be more of an indicator of a very busy AP rather than a metric used to diagnose an issue. As long as your not seeing 100% for long durations (longer than 15/30 mins) then I would not be too worried about it.