Graphing Stopped Suddenly

[joedunn]

Hi Guys

Here is my issue all of sudden all graphs across my farm stopped at 6:05pm last night. All the rrd's stopped updating too.

The poller is running, and if i run one of the commands to update a graph it completes but the png file it poops out is still at 6:05pm EST.

I've stopped the discovery plugin, and I've disabled the HTTP stats graphs cause they were the only new things I've added in the last 6 months.

When I run the poller by hand it seems to stop around here

CACTID: Host[219] DS[7688] WARNING: Result from SNMP not valid. Partial Result: ...

Not locking up, if I hit enter it just goes to the command prompt.

Any ideas what is happening, I'm loosing my mind a bit from the troubleshooting.

I'm using RHAS 4.1 with cactid and the plugin arch. Poller is running every 5 minutes

Any other information anyone requires I will happily update.

[gandalf]

This warning is quite common and should not be the root cause for your problem. Did you already check out the second link of my signature?
Reinhard

[joedunn]

Hi Reinhard

thank you for your reply, yes i did actually check that document. What is actually happening is all the RRD files stopped updating as of 18:05 yesterday.

As a test I chmod'ed them 777 and re-ran the poller with no results. I can see data being inserted into mysql and the log says it has parsed all 133 hosts successfully.

Just the graphs are not being updated, I feel like I've hit a brick wall with my troubleshooting

[gandalf]

YOur cacti is fully patched?
Reinhard

[joedunn]

YOur cacti is fully patched?
Reinhard

Version 0.8.6j
Plugin Architecture 1.1
Discovery 0.7
Thold 0.3.2

I'be disabled discovery this morning when I came in when I saw everything stopped I thought it was using too much resources. It was scanning 10.0.0.0/8

[gandalf]

YOur cacti is fully patched?
Reinhard

Version 0.8.6j
Plugin Architecture 1.1
Discovery 0.7
Thold 0.3.2

I'be disabled discovery this morning when I came in when I saw everything stopped I thought it was using too much resources. It was scanning 10.0.0.0/8

Not that good on an idea, indeed. It will fetch each address, one by one. And it takes some time for each of them. But when asking for patches, I was not referring to plugins. Find the patches at main cacti site
Reinhard

[joedunn]

YOur cacti is fully

I'be disabled discovery this morning when I came in when I saw everything stopped I thought it was using too much resources. It was scanning 10.0.0.0/8

Not that good on an idea, indeed. It will fetch each address, one by one. And it takes some time for each of them. But when asking for patches, I was not referring to plugins. Find the patches at main cacti site
Reinhard

Sorry, I just patched now, all went in save these two

root@perfmon01 cacti]# patch -p1 -N < ping_php_version4_snmpgetnext.patch
patching file lib/ping.php
Hunk #1 FAILED at 281.
Hunk #2 FAILED at 295.
Hunk #3 FAILED at 306.

[root@perfmon01 cacti]# patch -p1 -N < snmpwalk_fix.patch
patching file lib/snmp.php
Hunk #1 FAILED at 219.
1 out of 1 hunk FAILED -- saving rejects to file lib/snmp.php.rej

[joedunn]

So

As a test I deleted all data sources

then I added just one host, out of our 133 hosts

I'm seeing PHP errors

PHP Notice: Undefined index: method in /etc/cacti/lib/data_query.php on line 190
PHP Notice: Undefined index: method in /etc/cacti/lib/data_query.php on line 207