Apinger stops feeding rrdtool
-
I've got a functioning multi-wan setup using 2.2.4 that exhibits some weird behavior from apinger over time. To counteract that, I wrote a simple cron script that stops and starts it every 3 hours. Unfortunately, that seems to cause this problem:
Oct 4 19:11:03 apinger: rrdtool respawning too fast, waiting 300s.
Oct 4 19:11:03 apinger: Error while feeding rrdtool: Broken pipe
Oct 4 19:06:03 apinger: rrdtool respawning too fast, waiting 300s.
Oct 4 19:06:03 apinger: Error while feeding rrdtool: Broken pipeNo information is fed into the RRD graphs about quality, and I'd really like to track that for my 3 connections. Is there something else I should restart when apinger is restarted?
FWIW, I have to restart apinger because it eventually drifts to stating impossibly low ping times that don't match reality. Restarting it every few hours prevents that in the Gateways display, but disrupts feeding data to rrdtool.
-
No clues out there?
When I restart apinger from the services GUI, it doesn't cause the problem with being unable to feed rrdtool. The cron command I have running now is simplified all the way down to this:
/usr/local/sbin/pfSsh.php playback svc restart apinger
..but it still causes the problem. Anyone know what the difference is between restarting apinger the way I do in cron versus what is done via the GUI?
-
Anyone got the correct method to restart apinger from cron ?! Got the same problem, really annoying !
Thanks
Rudi -
Not me…seems like this should be a simple issue to fix, both apinger going awol in the first place as well as being able to restart it like the GUI does. Still waiting for a dev to intervene, I think.
-
Anything to do with apinger misbehaving is way beyond "simple", there's good reason some of the edge case issues along these lines haven't been fixed. apinger's been completely removed and replaced in 2.3, which fixes all those issues.
Look at what services.inc does vs. what you're doing to restart and you'll probably find the answer.
-
Well, the TASK is simple… Ping this IP, from this interface, using this route, and output the return time and loss here. That's simple enough I can do the majority, if not all, of it in a DOS script. Hopefully 2.3 has de-complicated whatever apinger did in <2.3 such that the complexity of the process matches the (lack of) complexity of the task. ;)
Thanks for the pointer to services.inc. I'll have to take a look at that.