[OpenWrt-Devel] [PATCH][RFC] libubus: added max retry mechanism

[Alexandru Ardelean]

I'll see what I can find.

At the time I created this patch I had it reproducing more often, but not
often enough to trace a cause.
The patch seems to help, but [as you've said] the error seems like
something else.
Seems to be reproducing again, after a while, and in a certain
configuration.

I'll check it out.
This high CPU usage thing got to annoy me enough that I gave it a bit more
attention.

Thanks
Alex

On Tue, Jul 29, 2014 at 4:30 PM, Felix Fietkau <nbd at openwrt.org> wrote:

> On 2014-07-29 15:23, Alexandru Ardelean wrote:
> > Quick info follow-up:  strace-ing on the PIDs (with high CPU usage)
> > shows a lot of EAGAIN errors from recvfrom() and a lot of recvfrom() +
> > poll() + clock_gettime_monotonic() calls.
> >
> > Thinking about it, would it make more sense to add a short wait instead
> > [to reduce the throttle] and let it "infinitely" loop ?  ["infinitely"
> > loop == loop until it does not error with EAGAIN]
> I think it's much more likely that this is a bug, caused by errors on
> the ubus socket not being caught or handled properly.
> If that is the case, then both approaches are wrong, and we should just
> handle the error correctly.
> Can you figure out what triggers this behavior?
>
> - Felix
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.infradead.org/pipermail/openwrt-devel/attachments/20140729/a28466f8/attachment.htm>
-------------- next part --------------
_______________________________________________
openwrt-devel mailing list
openwrt-devel at lists.openwrt.org
https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel