upgrade from 0.8.6f to 0.8.6h. some graphs not showing

[flipflap]

hi recently upgraded from version f to version h. upgrade successful as per cacti first time log but the problem is some of the equipment are not graphing. what did i miss? what do i need to check? thanks.

also attached is a result when i try to add a new device. all i get is a blank page. thanks.

OS is win2k | cacti 0.8.6h

[fmangeant]

Hi

have you applied theses patches : http://www.cacti.net/download_patches.php ?

Pre-patched files are available here : http://www.cacti.net/downloads/patches/0.8.6h/

[flipflap]

just to clarify. i just have to overwrite the files with the pre-patched fiels right? thanks for the reply!

[fmangeant]

just to clarify. i just have to overwrite the files with the pre-patched fiels right? thanks for the reply!

Yes, you just have to overwrite the files. Check the NTFS permissions after...

[flipflap]

thanks! hope it works. crossing any finger that i could cross

[flipflap]

still not graphing

[fmangeant]

Sorry, I'm not a Windows specialist...

Maybe this post will help you : http://forums.cacti.net/viewtopic.php?t=11747

[flipflap]

i'm getting this from one of the equipments that are not graphing. what does it mean. i could poll it correctly before the upgrade

08/14/2006 11:19:11 PM - CMDPHP: Poller[0] Host[51] DS[822] WARNING: Result from SNMP not valid. Partial Result:
08/14/2006 11:19:10 PM - CMDPHP: Poller[0] Host[51] DS[822] WARNING: Result from SNMP not valid. Partial Result:
08/14/2006 11:19:09 PM - CMDPHP: Poller[0] Host[51] DS[818] WARNING: Result from SNMP not valid. Partial Result:
08/14/2006 11:19:09 PM - CMDPHP: Poller[0] Host[51] DS[818] WARNING: Result from SNMP not valid. Partial Result:

[flipflap]

Operating System: win2k
Webserver: 5.0
Cacti: 0.8.6h (patched with plugin support for cacti h version)
Cactid: 0.8.6e
MySQL: 4.1
PHP: 5.0.4
RRDTool (CYGWIN version): NA using rrdtool binary 1.0.49
Net-SNMP: 5.2.1-1
Cygwin (cygwin1.dll version): NA

Thoroughly read my windows install guide and make sure you haven't missed anything. CHECK

Read my FAQ which address common problems. CHECK

Have you applied the latest patches for Cacti? Read this on how to apply them. CHECK

If you're not getting any graphs or rrd files created, it's likely because Cacti is not collecting valid data. To find out what is really going wrong, do the following and include it as either an attachment or QUOTE in your post. CACTI collects valid data for some equipments and invalid data for some equipments as per cacti log

Turn the polling level in cacti to debug (Settings | Polling level) for one cycle, wait for Cacti to run, and paste the output of the actual log file (<webroot>/cacti/log/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. please see attached

Another reason why graphs are not being displayed is because of some error specific to rrdtool. Go to Graph Mangement. Click on a graph. In the top right corner, enable Graph Debugging. Does rrdtool report an error? no error for the graphs that are not showing

If you think there might be a permission problem. ran file monitor no denied entry

[BSOD2600]

1) Do other parts of the cacti website function fine?
2) Does cacti consider Host[129] and Host[51] to be up?
3) When viewing the log file from within the cacti website, you can click on the text to find out what DS[933] or DS[822] are. What query are they associated with?
4) Since you were previously having problems with Host[51] and the attached log file is Host[129], it doesn't seem to be a reoccurring problem. Possibly a device rebooting?
5) Any reason you're not using the latest PHP?

[flipflap]

1) Do other parts of the cacti website function fine?
yes
2) Does cacti consider Host[129] and Host[51] to be up?
yes
3) When viewing the log file from within the cacti website, you can click on the text to find out what DS[933] or DS[822] are. What query are they associated with?
822 --> unicast statistics for interface
933 --> traffic statistics for interface
4) Since you were previously having problems with Host[51] and the attached log file is Host[129], it doesn't seem to be a reoccurring problem. Possibly a device rebooting?
nope they are not rebooting
5) Any reason you're not using the latest PHP?
no reason. just haven't thought of upgrading it

[flipflap]

what does "PCOMMAND: Poller[0] Host[110] WARNING: Recache Event Detected for Host" mean?

why do i get a lot of "CMDPHP: Poller[0] Host[x] DS[x] WARNING: result from snmp not valid. partial result"?

thanks

[BSOD2600]

what does "PCOMMAND: Poller[0] Host[110] WARNING: Recache Event Detected for Host" mean?

Means cacti detected the index count or uptime changed (as in the device rebooted) for a device so cacti is refreshing the data queries.

why do i get a lot of "CMDPHP: Poller[0] Host[x] DS[x] WARNING: result from snmp not valid. partial result"?

Means an OID is no longer valid. If you change the logging level to debug, you can see exactly what OID it was trying to poll. Typically a recache fixes this. If not, click the green circle next to the associated data query for the device. There is also a php script that comes with cacti that can do recaching automatically for you.


I'd try updating PHP to see if the hundreds of their bugs fixes your problem. Just to reiterate, the only problem you have is you're unable to add new devices?

[flipflap]

i was able to add a device today that i was previously not able to add. i don't know why it decided to work as it should now. i'll try adding some more devices and see if i still have the problem. the other problem i have is some of the devices stoppped graphing --> please take a look at the attached in this thread the one with the red fill with description "cacti equipment graph no graph after upgrade" some of the equipment does. if it doesn't work for a particular equipment all the associated graphs for that host would look like that or experience that.
[quote]
Means cacti detected the index count or uptime changed (as in the device rebooted) for a device so cacti is refreshing the data queries. [/quote]

so while it is refreshing it will not graph for that equipment? i've been getting lots of that from the log
[quote]
Typically a recache fixes this. If not, click the green circle next to the associated data query for the device. There is also a php script that comes with cacti that can do recaching automatically for you. [/quote]

so clicing the green circle will force a recache? how will i force a recache? what is the php script that will recache it? maybe that's the problem with my setup, cacti can't recache it so it can't graph it.

thanks for the reply

[BSOD2600]

the other problem i have is some of the devices stoppped graphing --> please take a look at the attached in this thread the one with the red fill with description "cacti equipment graph no graph after upgrade" some of the equipment does.

For that cpu usage graph, look though the log file for when cacti tries to poll that device via SNMP. Does it give a message saying "result from snmp not valid" ?

so while it is refreshing it will not graph for that equipment? i've been getting lots of that from the log

No, it recaches the device / data query first, then polls it with the updated OIDs.

so clicing the green circle will force a recache? how will i force a recache? what is the php script that will recache it? maybe that's the problem with my setup, cacti can't recache it so it can't graph it.

Yes, the the green circle will force a manual recache. Look at the poller_reindex_hosts.php script if you want to do this manually on a large scale (although really not needed unless you're constantly enabling/disabling interfaces on routers, for example). If you show the Verbose Query and all the data shows up properly, cacti should be able to successfully collect and graph it.