Problems with 0.8.6G with SNMP

Caliban · Post by **Caliban** » Mon Sep 12, 2005 10:23 pm

Having issues graphing SNMP - Mounted Partitions since I've upgraded to 0.8.6G Alot of them are not working, though some are but there doesn't seem to be any pattern on which do or don't. Both Linux and Windows boxes. I've had this happen before and all I needed to do was requery the Mounted Partitions from the Devices but thats not working now and I have alot more not working then working. I've tried using the php poller and has made no difference. I've tried clearing the poller cache with no luck and I didn't replace any of the new scripts with the old ones during the upgrade.

Here's a sampe of my Log file:

09/12/2005 11:20:48 PM - SYSTEM STATS: Time:46.1192 Method:cactid Processes:2 Threads:2 Hosts:152 HostsPerProcess:76 DataSources:2092 RRDsProcessed:876
09/12/2005 11:20:44 PM - CACTID: Poller[0] Host[233] DS[3118] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:44 PM - CACTID: Poller[0] Host[233] DS[3117] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:44 PM - CACTID: Poller[0] Host[233] DS[3116] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:44 PM - CACTID: Poller[0] Host[233] DS[3115] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:36 PM - CACTID: Poller[0] Host[222] DS[2950] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:36 PM - CACTID: Poller[0] Host[222] DS[2949] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:36 PM - CACTID: Poller[0] Host[222] DS[2948] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:36 PM - CACTID: Poller[0] Host[222] DS[2947] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[221] DS[2851] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[221] DS[2850] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[221] DS[2849] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[221] DS[2848] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[220] DS[2842] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[220] DS[2841] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[220] DS[2840] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[220] DS[2839] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[219] DS[2833] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[219] DS[2832] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[219] DS[2831] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[219] DS[2830] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[218] DS[2824] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[218] DS[2823] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[218] DS[2822] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[218] DS[2821] WARNING: Result from SERVER not valid. Partial Result: ...
09/12/2005 11:20:35 PM - CACTID: Poller[0] Host[217] DS[2816] WARNING: Result from SERVER not valid. Partial Result: ...

Unfortunately I didn't backup my SQL database before upgrading so I can't go back to 0.8.6F so I need some help here as this is a production system.

BSOD2600 · Post by **BSOD2600** » Mon Sep 12, 2005 11:31 pm

Turn the polling level in cacti to debug (Settings | Polling level) for one cycle and paste the output of the log file (./cacti/cacti.log). If it contains WEBLOG entries, then don't bother posting since they just clutter up the log. Turn off WEBEVENTS logging and try again.

DAP · Post by **DAP** » Mon Sep 12, 2005 11:58 pm

Hi, I am having the exact same problem. Here is my cacti.log file for your review.

Thankyou.

Post by **gandalf** » Tue Sep 13, 2005 5:38 am

What version of cactid are you running?
Reinhard

sohailb · Post by **sohailb** » Tue Sep 13, 2005 12:46 pm

Same for me and I am running 0.8.6e cactid and 0.8.6g cacti.

Setarcos · Post by **Setarcos** » Tue Sep 13, 2005 12:57 pm

I had this same issue with the tar.gz version of the distribition for 0.8.6g. It ended up being the file permissions on the scripts in <path_cacti>/scripts/

sohailb · Post by **sohailb** » Tue Sep 13, 2005 1:43 pm

Even after making sure all the permissions were exactly the same as my last install, it still doesnt work.

I removed the hosts and readded them and on the data query it shows 0 items.

I set the permissions to 755 for *.pl, *.sh and *.php in <cacti>/scripts

Caliban · Post by **Caliban** » Tue Sep 13, 2005 4:16 pm

I set the permission of all scripts to 777 to be sure no change, cleared the poller cache still not working.. Funny thing is some that are working aren't working well.. lots of drop offs.. the used spaced showing but not total. And its only affecting the SNMP mounted partition checks not the interface or the CPU. I'm running Fedora Core 4 full updated with the RPM of Cacti 0.8.6G with Cactid 0.8.6E but even with the php poller it doesn't work. I've included the poller log at the setting you asked for hopefully you can figure out whats wrong. Had to zip it to get it to fit.

Setarcos · Post by **Setarcos** » Tue Sep 13, 2005 5:01 pm

Caliban, these all appear to be Script Server scripts that are generating errors. Have you tried manually running these ala http://forums.cacti.net/viewtopic.php?t=7738 ?

You might also want to look here: http://forums.cacti.net/viewtopic.php?t=9286

Post by **TheWitness** » Tue Sep 13, 2005 8:34 pm

OUr documentation is out of date. The upgrade instructions indicate that you must copy all of your old scripts from the old scripts directory to the new one.

However, when you do that, you will break the default script server scripts ss_host_disk.php and ss_host_cpu.php which both changed in the 0.8.6g release to support SNMP v3. If you simply copy back the ones from the RPM/ZIP/TAR file, you should be Ok.

Also, please keep in mind that in earlier version of Cacti, the script did not pay attention to the hosts timeout. In some cases, on busy servers, the poll of disk takes about 8-10 seconds.

TheWitness

sohailb · Post by **sohailb** » Tue Sep 13, 2005 8:56 pm

That works.

Thanks!

Caliban · Post by **Caliban** » Tue Sep 13, 2005 9:26 pm

When I copied my old scripts into my new scripts folder I only copied those that didn't already exist I didn't replace any of the new ones. As for the Host timeouts which do you mean? I tried this one on the Device section
SNMP Timeout setting it as high as 50000 with no change. I also went into the settings and tried Script and Script Server Timeout Value under the poller went as high as 8000 with that and raised Ping Timeout Value as well with no luck.

Post by **TheWitness** » Wed Sep 14, 2005 5:05 am

Post a snippet from your poller cache. Then also open a command prompt and run that snippet and see if you get data.

TheWitness

Caliban · Post by **Caliban** » Wed Sep 14, 2005 1:49 pm

Here's one that I know isn't working:
Data Source: NewsAlertDB1 - Hard Drive Space
RRD: /var/www/html/cacti/rra/newsalertdb1_hdd_total_1484.rrd
Action: 2, Script: /var/www/html/cacti/scripts/ss_host_disk.php ss_host_disk 172.18.4.202 114 1:161:500:public:::MD5::[None] get used 2

I ran it two ways from the command prompt.. once raw:
# /var/www/html/cacti/scripts/ss_host_disk.php ss_host_disk 172.18.4.202 114 1:161:500:public:::MD5::[None] get used 2
/var/www/html/cacti/scripts/ss_host_disk.php: line 1: ?php: No such file or directory
/var/www/html/cacti/scripts/ss_host_disk.php: line 2: =: command not found
/var/www/html/cacti/scripts/ss_host_disk.php: line 4: /bin: is a directory
/var/www/html/cacti/scripts/ss_host_disk.php: line 5: syntax error near unexpected token `E_ERROR'
/var/www/html/cacti/scripts/ss_host_disk.php: line 5: `error_reporting(E_ERROR);'

and once with PHP:
# php /var/www/html/cacti/scripts/ss_host_disk.php ss_host_disk 172.18.4.202 114 1:161:500:public:::MD5::[None] get used 2
PHP Notice: Undefined offset: 1 in /var/www/html/cacti/scripts/ss_host_disk.php on line 20
PHP Notice: Undefined offset: 2 in /var/www/html/cacti/scripts/ss_host_disk.php on line 21
PHP Notice: Undefined offset: 3 in /var/www/html/cacti/scripts/ss_host_disk.php on line 37

I checked the file creation date of my scripts.. the "ss_" ones are Sept 8 along with the others I didn't copy over.. the old scripts are Sept 12 which are the ones that didn't exist in the new install so I'm pretty sure I didn't copy over by mistake.

Caliban · Post by **Caliban** » Wed Sep 14, 2005 1:57 pm

Ok I just figured you should be doing this with the script_server running here's a few I tried:
/var/www/html/cacti/scripts/ss_host_disk.php ss_host_disk 172.18.4.202 114 1:161:500:public:::MD5::[None] get total 4
0
/var/www/html/cacti/scripts/ss_host_disk.php ss_host_disk 172.18.4.202 114 1:161:500:public:::MD5::[None] get used 2
5048049664
/var/www/html/cacti/scripts/ss_host_disk.php ss_host_disk 172.18.124.243 197 1:161:50000:public:::MD5::[None] get total 1
0

Got 1 out of 3

Problems with 0.8.6G with SNMP

Problems with 0.8.6G with SNMP

Who is online