RRD graphing

nocer · Nov 27, 2008, 9:10 AM

My cpu seems to have stoped charting, is this related to this topic?

yup, and "Quality" graph too :'(

databeestje · Nov 28, 2008, 10:35 PM

the quality graphs have been changed over to a new format because we use the apinger files.

This means removing the /var/db/rrd/*-quality.rrd files. And then reboot.

The processor graph should have been fixed. This was a off by one.

The wan being show down is definitely a issue with pppoe and pptp interfaces. This will need some time on my part. Not having a pppoe or pptp setup makes it that more time consuming.

databeestje · Nov 28, 2008, 11:54 PM

does

echo " gateway '". get_interface_gateway("wan") ."' found\n";

on the command page produce output on your pppoe install?

databeestje · Nov 29, 2008, 12:15 AM

I have added a debug statement that should log this to the system logs.

Please check a new snapshot and see if you see any found dynamic gateway x.x.x.x in the system logs and paste the output here.

nocer · Nov 29, 2008, 1:36 AM

Hello databeestje, thanks for follow ups;

@databeestje:

does
echo " gateway '". get_interface_gateway("wan") ."' found\n";
on the command page produce output on your pppoe install?

yup, it correctly looks my gateway; "gateway '210.xxx.xxx.5' found" is the output. Which is my real WAN IP obtained by PPPoE from ISP.

@databeestje:

Please check a new snapshot and see if you see any found dynamic gateway x.x.x.x in the system logs and paste the output here.

Well, it looks there's no snaps since 25th…snap server seems to be taking a nap now.

cheers,

2.0-ALPHA-ALPHA
built on Tue Nov 25 14:59:09 EST 2008
FreeBSD 7.1-PRERELEASE

nocer · Nov 29, 2008, 5:26 AM

Hi,

@databeestje:

I have added a debug statement that should log this to the system logs.

Please check a new snapshot and see if you see any found dynamic gateway x.x.x.x in the system logs and paste the output here.

Surely found "PHP: found dynamic gateway 210.xxx.xxx.130" in the system log.

BTW, still no luck with CPU graph. Quality graph is back but dead red.

cheers,

2.0-ALPHA-ALPHA
built on Fri Nov 28 22:55:11 EST 2008
FreeBSD 7.1-PRERELEASE

mino · Nov 29, 2008, 8:52 AM

php: /status_rrd_graph_img.php: Failed to create graph with error code 1, the error is: ERROR: No DS called 'delay' in '/var/db/rrd/wan-quality.rrd'

And the quality graphs are only showing a error that you need to look in the logs.

databeestje · Nov 29, 2008, 9:25 AM

mino, please remove the quality rrd files and let them be recreated. The format has changed.

databeestje · Nov 29, 2008, 9:30 AM

nocer, I will have to look into it further. It is not obvious to me what causes this yet. The debug statement did help though. Now I know it is triggering the right path.

That the quality graph is red is correct since it pings 127.0.0.2 which is a dummy address which is normally used when it can't find the gateway for a dynamic interface. That one always times out :-)

I'll have another look at the cpu graph then.
Is this a single or dual core box?

nocer · Nov 29, 2008, 10:28 AM

Thank you always databeestje :)

I'm running Atom230@1.60GHz, a single core but HT capable with SMP kernel. As for the quality graph, it used to ping my gateway correctly and created right graph, and I of course deleted 127.0.0.2 from the apinger.conf and restarted the process(not a reboot) so there's nothing apinger to do with 127.0.0.2.

cheers,

databeestje · Nov 29, 2008, 12:20 PM

If the gateway IP comes through normally in the code it will automatically configure apinger to ping the right gateway.

I'll add more debugging and see if I can work out what is happening.

databeestje · Nov 29, 2008, 11:29 PM

I think I may have solved it, the debugging statements are all there but it took a while to set up a testing rig.

Please test another snapshot.

nocer · Nov 30, 2008, 4:30 AM

Hi,

Some excerpts from system.log those seem to be related for gateway thingy:
–-
Nov 30 12:13:45 check_reload_status: reloading filter
Nov 30 12:13:34 apinger: ALARM: wan(127.0.0.2) * down *
Nov 30 12:13:29 check_reload_status: updating dyndns
Nov 30 12:13:24 check_reload_status: reloading filter
Nov 30 12:13:24 apinger: Exiting on signal 15.
Nov 30 12:13:24 php: : Found dynamic gateway '210.xxx.xxx.6'
Nov 30 12:13:24 php: : Found dynamic gateway ''
Nov 30 12:13:23 php: : Creating rrd update script

Name Gateway Monitor Status Description
WAN 127.0.0.2 127.0.0.2 Offline Last succes Jan 1 09:00:00 1970 Interface WAN Gateway
WAN 210.xxx.xxx.6 210.xxx.xxx.6 Online Last succes Nov 30 13:14:54 2008 Interface WAN Gateway

I dunno what's been changed since last snap tho :P

Still no luck with CPU/Quality graphs.

cheers,

2.0-ALPHA-ALPHA
built on Sat Nov 29 08:14:04 EST 2008
FreeBSD 7.1-PRERELEASE

databeestje · Nov 30, 2008, 9:53 PM

as soon as the gateway status page is correct and not showing the duplicate WAN entry it will be automatically picked up in the quality graph. Ermal has made another fix for the pppoe problem which might have nailed it since he uses pppoe.

I will look further into the cpu graph problem.

databeestje · Nov 30, 2008, 10:05 PM

CPU grapjh should be fixed in CVS

Shraik · Dec 1, 2008, 4:40 AM

Version 2.0-ALPHA-ALPHA
built on Sun Nov 30 21:11:20 EST 2008
FreeBSD 7.1-PRERELEASE

Wan=PPPoe

Status: RRD Graphs -> Quality
don't worked

LOG:
Dec 1 11:34:51 php: /status_rrd_graph_img.php: Failed to create graph with error code 1, the error is: ERROR: No DS called 'delay' in '/var/db/rrd/wan-quality.rrd'

databeestje · Dec 1, 2008, 6:18 AM

Shraik, you need to remove the /var/db/rrd/*-quality.rrd files. The format has changed. 3 times is golden.

nocer · Dec 1, 2008, 8:02 AM

Hello,

Thank you always databeestje, everything is back to as it supposed to be since your last post, Sun Nov 30 21:11:20 EST 2008 build. The CPU and Quality graph started drawing again correctly, and the Gateway also looks normal, no more 127.0.0.2 is shown, no annoying apinger logs in the system.log any more.

cheers,

databeestje · Dec 1, 2008, 4:46 PM

That's good to hear.

Thank you for helping testing!

mino · Dec 2, 2008, 1:12 PM

Hi,

today i look into my "System Log" and it show many errors every 5 Minutes these messages are repeated in the log:


apinger: Error while feeding rrdtool: Broken pipe
apinger: rrdtool respawning too fast, waiting 300s.

My quality graph is gone and this messages are shown since the automatic PPPOE restart which took place(as reported in the other topic this feature is still buggy). I hope i can help.