On my quest to find the cause of some packet loss I’ve found the CPU on both my primary and secondary to be constantly active.
I don’t think the CPU usage is the cause of the packet loss but I’m curious as to what’s causing it as I’ve not seen it like this before.
On my primary I’m seeing this with top (constant 14-16% CPU)
19 root 20 0 0 0 0 S 15.6 0.0 6045:59 events/0
and on my secondary (constant 14-16% usage) (after a reboot which did not fix it)
22 root 20 0 0 0 0 S 14.9 0.0 3:28.80 events/3
There’s nothing strange in any of the logs, dmesg etc so I’m not sure what’s going on.
I’m running a stock 64-bit FPBX 2.11 install 5.211.65-14
In a 2.6 kernel, events/cpu_number is a mechanism (it replaces the old keventd) for handling low-level requests that need to run asynchronously. All sorts of notifications run through there … it’s very ad hoc. So if you’re seeing this “pegging the meter,” all it’s really telling you is that there’s a loop somewhere.
In a quick google search this may be related to your NIC.
I’m trying to workout top’s TIME+ to get an idea as to when this started. It looks like it may have started after the last FPBX upgrade script when I went from 5.211.65-12 to 14
I’ve tried resetting the network cards and unplugging both (all on the secondary, I don’t want to touch the primary just yet) but events/3 still sits at a constant 15% CPU
events/3 is on the secondary
events/0 is on the primary
Update:
It’s something to do with DAHDI which I know got updated on the last FPBX script. If I do an amportal stop events/3 on the backup drops to 0%. If I wanrouter start events/3 goes back to 15% CPU. Running amportal start kicks everything back into life but events/3 is still 15%
Running wanrouter start is OK until the part where it starts DAHDI. At this point events/3 jumps to constant 15% CPU
Cheers