1 Min. Load Average (BSD only - No Linux support in net-snmp?)
5 Min. Load Average (BSD only - No Linux support in net-snmp?)
15 Min. Load Average (BSD only - No Linux support in net-snmp?)
ucd/net snmp Device (Disk) I/O templates - Updated - v3.1
Moderators: Developers, Moderators
-
- Posts: 8
- Joined: Wed Jul 09, 2008 5:20 pm
- Location: En cage à la S.P.A
I managed to install this scripts in my Cacti system (v0.8.7e) and I'm trying to graph diskIO for a linux host running net-snmp-5.4.2.1.
The SNMP part seems ok. I can get the snmptable for the diskIO MIB, and here's an example query, run from the cacti system, for one device (index 257) - I don't get IO load averages as expected (they're bsd-only afaik):
problem is, in Cacti I get graphs for the diskION[Read|Written] values, but not for diskIO[Reads|Writes] (actually what I'm looking for). I don't get empty graphs (ie, no values), I actually don't get any image rendered for those graphs. Sample screenshot of the two kinds of graph (bytes R/W and number of R/W ops):
All my diskIO data queries, and the 'Reads/Writes' in particular', look exactly like the ones screenshotted at page 3 of this thread.
Looking at data sources for this specific disk on this specific server, I have two of them, each one with its rrd file, one for Bytes R/W and the other for number of reads and writes.
The RRD file for the diskIO[Reads|Writes] data sources does exist and gets updated by the poller (tried both cacti's and spine, I'm usually using the latter).
'rrdtool info' run on the files shows that both my diskIO RRDs (for this disk, one is the bytes r/w datasource and the other is the number of r/w datasource) contain all the diskIO "columns" (diskION[R/W], diskIO[R/W], diskIOLA[1/5/15]).
I'm starting to suspect the problem is a really stupid one hidden somewhere, but I really don't know where to look. Any help would be appreciated. Thanks.
The SNMP part seems ok. I can get the snmptable for the diskIO MIB, and here's an example query, run from the cacti system, for one device (index 257) - I don't get IO load averages as expected (they're bsd-only afaik):
Code: Select all
$ snmpwalk -v2c -cXXXXX -mALL 172.16.0.5 diskIO | grep \\.257
UCD-DISKIO-MIB::diskIOIndex.257 = INTEGER: 257
UCD-DISKIO-MIB::diskIODevice.257 = STRING: sda
UCD-DISKIO-MIB::diskIONRead.257 = Counter32: 655539200
UCD-DISKIO-MIB::diskIONWritten.257 = Counter32: 2424565760
UCD-DISKIO-MIB::diskIOReads.257 = Counter32: 3243491
UCD-DISKIO-MIB::diskIOWrites.257 = Counter32: 14268833
UCD-DISKIO-MIB::diskIONReadX.257 = Counter64: 120914623488
UCD-DISKIO-MIB::diskIONWrittenX.257 = Counter64: 513525673984
All my diskIO data queries, and the 'Reads/Writes' in particular', look exactly like the ones screenshotted at page 3 of this thread.
Looking at data sources for this specific disk on this specific server, I have two of them, each one with its rrd file, one for Bytes R/W and the other for number of reads and writes.
The RRD file for the diskIO[Reads|Writes] data sources does exist and gets updated by the poller (tried both cacti's and spine, I'm usually using the latter).
'rrdtool info' run on the files shows that both my diskIO RRDs (for this disk, one is the bytes r/w datasource and the other is the number of r/w datasource) contain all the diskIO "columns" (diskION[R/W], diskIO[R/W], diskIOLA[1/5/15]).
I'm starting to suspect the problem is a really stupid one hidden somewhere, but I really don't know where to look. Any help would be appreciated. Thanks.
mr.Scamp wrote:What data do i have to input in the fields on the picture below to get it working?
In what field do i have to type the disk`s name such as hda or ad0?
I had the same issue. I finally figure it out. I had the graph templates associated to the host. I removed the graph association and associated the data query to host.
Then when I picked create graphs a new section was added.
Hope this helps people.
ko
HELP ME!!
I cannot working.I am used cacti-0.8.7e
OS ubuntu 8.04 kernel 2.6.24-19-server
this is error info:
Notice: Undefined index: sgg_10 in /home/cacti/cacti-0.8.7e/graphs_new.php on line 71
Notice: Undefined index: local_data_id in /home/cacti/cacti-0.8.7e/graphs_new.php on line 214
Warning: Invalid argument supplied for foreach() in /home/cacti/cacti-0.8.7e/graphs_new.php on line 214
Warning: Cannot modify header information - headers already sent by (output started at /home/cacti/cacti-0.8.7e/graphs_new.php:71) in /home/cacti/cacti-0.8.7e/graphs_new.php on line 330
why?
OS ubuntu 8.04 kernel 2.6.24-19-server
this is error info:
Notice: Undefined index: sgg_10 in /home/cacti/cacti-0.8.7e/graphs_new.php on line 71
Notice: Undefined index: local_data_id in /home/cacti/cacti-0.8.7e/graphs_new.php on line 214
Warning: Invalid argument supplied for foreach() in /home/cacti/cacti-0.8.7e/graphs_new.php on line 214
Warning: Cannot modify header information - headers already sent by (output started at /home/cacti/cacti-0.8.7e/graphs_new.php:71) in /home/cacti/cacti-0.8.7e/graphs_new.php on line 330
why?
well wow.
Thanks for the template guys but no support after from the author?
I have the same problem posted by about half a dozen other people who had zilch reply to.
The problem been when I create graph I get 3 empty boxes.
Output Type ID
Index Value
Index Type
php-snmp is working according to cacti and php info.
Thanks for the template guys but no support after from the author?
I have the same problem posted by about half a dozen other people who had zilch reply to.
The problem been when I create graph I get 3 empty boxes.
Output Type ID
Index Value
Index Type
php-snmp is working according to cacti and php info.
I got excited when I first read this so unassociated the graph templates from the host, however when creating the graphs I still get the 3 empty boxes, the data query IS associated with the host. any other ideas?knightorc wrote:mr.Scamp wrote:What data do i have to input in the fields on the picture below to get it working?
In what field do i have to type the disk`s name such as hda or ad0?
I had the same issue. I finally figure it out. I had the graph templates associated to the host. I removed the graph association and associated the data query to host.
Then when I picked create graphs a new section was added.
Hope this helps people.
ko
--edit--
Ok I got them added, although I am not confident they will work as I see gandalf has some new template for 0.8.7?
For me the fix was I didnt associate the graphs with the data query itself.
--edit--
update
graphs arent been created, examining cacti log shows cacti isnt even attempting to make the graphs as the data sources count has not increased at all. If I goto graph management and turn on debug mode it says the rrd file does not exist.
DataSources:371 RRDsProcessed:306
it was 371 and 306 respectively before I added the graphs. So basically cacti is simply not processing them, no errors, nothing.
I searched for it in debug logs and there is no reference to either the data source or graphs.
-
- Posts: 18
- Joined: Wed Mar 19, 2008 2:42 pm
should host have something in snmpd.conf?
I just updated to 0.8.7e. I used instructions from page 1. and my data query does return disk names but no counters. Am I missing something? Seems like there needs to be some snmpd.conf information.
Re: should host have something in snmpd.conf?
I found out, that Disk I/O Bytes read and written were fairly wrong on servers due to the 32-bit counters overrunning. I attach an updated net-snmp_devio.xml which uses the 64-bit bytecounters that works for me[tm].
Just replace the original file with this one and recreate your queries and graphs. To check if your snmpd supports the 64-bit counters issue
If it shows you the counters you're lucky
Just replace the original file with this one and recreate your queries and graphs. To check if your snmpd supports the 64-bit counters issue
Code: Select all
snmpwalk -v 2c HOSTNAME -c COMMUNITY 1.3.6.1.4.1.2021.13.15|grep ReadX
- Attachments
-
- devio.zip
- Replace the original file with this one
- (695 Bytes) Downloaded 4867 times
Version 3.1 - Includes Data Query XML Template in .zip File
Hello,
It appears that my original export didn't include the data query required to generate the graphs, even though I had "included all dependencies" in my original export. I have updated the download, on the very first post of this topic thread, to include the data query required by the data and graph templates.
My new setup is Centos 5.5 x64 running Cacti 0.8.7g and it is confirmed working.
I hope this helps.
~Rodre
It appears that my original export didn't include the data query required to generate the graphs, even though I had "included all dependencies" in my original export. I have updated the download, on the very first post of this topic thread, to include the data query required by the data and graph templates.
My new setup is Centos 5.5 x64 running Cacti 0.8.7g and it is confirmed working.
I hope this helps.
~Rodre
Re: should host have something in snmpd.conf?
Under Linux, and possibly other OSes that use the net-snmp implementation you will have to allow Cacti to walk the OIDs required for device I/O. You can do this by making the following change in your /etc/snmp/snmpd.conf file:five0.4tluv wrote:I just updated to 0.8.7e. I used instructions from page 1. and my data query does return disk names but no counters. Am I missing something? Seems like there needs to be some snmpd.conf information.
# Make at least snmpwalk -v 1 localhost -c public system fast again.
# name incl/excl subtree mask(optional)
#view systemview included .1.3.6.1.2.1.1
#view systemview included .1.3.6.1.2.1.25.1.1
view systemview included .1
~Rodre
Re: should host have something in snmpd.conf?
I was considering including this file in the template bundle, but I was not sure how it would affect 32 bit systems. I have updated the original post to include this file.webstoney wrote:I found out, that Disk I/O Bytes read and written were fairly wrong on servers due to the 32-bit counters overrunning. I attach an updated net-snmp_devio.xml which uses the 64-bit bytecounters that works for me[tm].
Just replace the original file with this one and recreate your queries and graphs. To check if your snmpd supports the 64-bit counters issue
If it shows you the counters you're luckyCode: Select all
snmpwalk -v 2c HOSTNAME -c COMMUNITY 1.3.6.1.4.1.2021.13.15|grep ReadX
I am not sure how you guys were able to associate the Graph Templates to the hosts as opposed to the Data Query. I wasn't able to do that in my previous version of my Cacti Installation, and I am not able to do it on my current one (0.8.7g). Perhaps it was a bug in the version you were using? Or maybe because it was missing the data query?knightorc wrote:mr.Scamp wrote:What data do i have to input in the fields on the picture below to get it working?
In what field do i have to type the disk`s name such as hda or ad0?
I had the same issue. I finally figure it out. I had the graph templates associated to the host. I removed the graph association and associated the data query to host.
Then when I picked create graphs a new section was added.
Hope this helps people.
ko
~Rodre
Re: ucd/net snmp Device (Disk) I/O templates - Updated - v3.
I should have posted that I have 2 of the 3 graphs working now.
I am trying to remember exactly what it was I changed to fix but I cannot unfortenatly.
I am trying to remember exactly what it was I changed to fix but I cannot unfortenatly.
Re: ucd/net snmp Device (Disk) I/O templates - Updated - v3.
Hi,
I have just added the files of diskio as write in the first post but after the first polling I see these errors in cacti log:
And in the graphs TAB there aren't any graphs of diskio but only the string name.
Any suggestions?
Best regards,
Emilien
I have just added the files of diskio as write in the first post but after the first polling I see these errors in cacti log:
Code: Select all
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:55:02 PM - SPINE: Poller[0] FATAL: Spine Encountered a Segmentation Fault (Spine thread)
11/26/2010 06:55:01 PM - POLLER: Poller[0] WARNING: There are '3' detected as overrunning a polling process, please investigate
11/26/2010 06:55:00 PM - SYSTEM STATS: Time:298.8377 Method:spine Processes:3 Threads:10 Hosts:7 HostsPerProcess:3 DataSources:195 RRDsProcessed:0
11/26/2010 06:55:00 PM - POLLER: Poller[0] Maximum runtime of 298 seconds exceeded. Exiting.
11/26/2010 06:50:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:50:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating
11/26/2010 06:50:02 PM - PHPSVR: Poller[0] ERROR: Input Expected, Script Server Terminating ....
....
Any suggestions?
Best regards,
Emilien
Who is online
Users browsing this forum: No registered users and 5 guests