BigIP v9 Template

[ben_c]

Hm, i've done the same and am continuing to monitor, have gone about 12 hours without the issue so far.

[bsdboy]

Good news! Is that with the CPU scripts enabled too?

[ben_c]

No, too worried to enable that yet :p

I am a little concerned that script wasn't really returning much of the correct data for me, always seem to be '0'.

[ben_c]

It heard us talking about it, it's crashed out again. IMO the max conns just lengthens the time between the problem happening but doesn't resolve it.

Seem to get a lot of mysql.sock sitting running when I do a netstat -an, doesnt look right

unix 2 [ ACC ] STREAM LISTENING 40623 /var/lib/mysql/mysql.sock
unix 2 [ ] DGRAM 4857 @udevd
unix 2 [ ] DGRAM 720921
unix 3 [ ] STREAM CONNECTED 718964 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 718963
unix 3 [ ] STREAM CONNECTED 564133
unix 3 [ ] STREAM CONNECTED 564132
unix 3 [ ] STREAM CONNECTED 515223
unix 3 [ ] STREAM CONNECTED 515222
unix 3 [ ] STREAM CONNECTED 514902 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514901
unix 3 [ ] STREAM CONNECTED 514841 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514840
unix 3 [ ] STREAM CONNECTED 514832 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514831
unix 3 [ ] STREAM CONNECTED 514829 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514828
unix 3 [ ] STREAM CONNECTED 514818 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514817
unix 3 [ ] STREAM CONNECTED 514810 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514809
unix 3 [ ] STREAM CONNECTED 514800 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514799
unix 3 [ ] STREAM CONNECTED 514789 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514788
unix 3 [ ] STREAM CONNECTED 514786 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514785
unix 3 [ ] STREAM CONNECTED 514783 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514782
unix 3 [ ] STREAM CONNECTED 514772 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514771
unix 3 [ ] STREAM CONNECTED 514769 /var/lib/mysql/mysql.sock
unix 3 [ ] STREAM CONNECTED 514768

[ben_c]

Good news! Is that with the CPU scripts enabled too?

How have you gone with the TMM CPU script bsdboy?

I've solved my poller problems with some mysql tweaking it seems, but am still getting mixed results from the cpu template (either it adds a few seconds to the poller runtimes, or just returns a 0 result.

[mantisgb]

Has anyone managed to capture SNMP info/graph CPU0 usage? I need to monitor CPU0 as in an ASM environment it is CPU0 that is doing all the work.

[bsdboy]

Good news! Is that with the CPU scripts enabled too?

How have you gone with the TMM CPU script bsdboy?

I've solved my poller problems with some mysql tweaking it seems, but am still getting mixed results from the cpu template (either it adds a few seconds to the poller runtimes, or just returns a 0 result.

I have not used the CPU scripts. I put these traffic scripts in long before they existed and already know that my CPU's are tapped out

I think citizen_elah came up with those, he's on this board along with the devcentral board. Might not be a bad idea to send him a PM.

[ben_c]

Fair enough

[Heiko]

Hello,
I have a couple of questions for using cacti in a failover LTM setup.

I have our active unit configured for cacti, ist there a way to copy all the settings/trees to a different tree. Both machines have exactly the same setup, only a different Ip Address.

Can we get some statistics at the end of the month that tell me VServer xy had a traffic of 2.3TB?

Is there a way to merge the statistics of both LTMs?

thx a lot


Heiko

EDIT:

I would like to start creating a template to graph the pools and there members, will this be possible ?

[aboling]

Check this thread slinger, it may explain why it's not working and what needs updating : http://forums.cacti.net/viewtopic.php?t=23932

I'm about to install about 16 new BigIP's so need to get this template fixed up too.

ben_c is 100% correct, this is the cause of my problem. Neither the template posted at the start of this thread, or the one linked to off of devcentral.f5.com (which just further confuses this thread IMO, it should have its own) have been rewritten to conform to the changes in 0.8.7.

Furthermore, the devcentral template includes a Perl script which is written to only use v2c even if v1 or v3 is set...I think I'm going to switch back to the template mentioned at the start of this thread and work on converting it to the new format. I'll post a revised version here if I have any success with it. Thanks again, ben_c!

[ben_c]

Check this thread slinger, it may explain why it's not working and what needs updating : http://forums.cacti.net/viewtopic.php?t=23932

I'm about to install about 16 new BigIP's so need to get this template fixed up too.

ben_c is 100% correct, this is the cause of my problem. Neither the template posted at the start of this thread, or the one linked to off of devcentral.f5.com (which just further confuses this thread IMO, it should have its own) have been rewritten to conform to the changes in 0.8.7.

Furthermore, the devcentral template includes a Perl script which is written to only use v2c even if v1 or v3 is set...I think I'm going to switch back to the template mentioned at the start of this thread and work on converting it to the new format. I'll post a revised version here if I have any success with it. Thanks again, ben_c!

Hi aboling,

Have you had any luck getting the TMM CPU graph/scripts working?

[edodd]

Here's what I'm seeing with the latest f5_bigip_vs.xml, when I try to run a verbose query for F5 BigIP - VS Statistics :

Code: Select all

+ Running data query [15].
+ Found type = '3' [snmp query].
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ XML file parsed ok.
+ Executing SNMP walk for list of indexes @ '.1.3.6.1.4.1.3375.2.2.10.2.3.1.1'
+ No SNMP data returned
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/f5_bigip_vs.xml'

My f5_bigip_vs.xml looks like

Code: Select all

        <name>Get F5 Big-IP Load Balancer Virtual Server Statistics</name>
        <oid_index>.1.3.6.1.4.1.3375.2.2.10.2.3.1.1</oid_index>
        <oid_index_parse>OID/REGEXP:^.{35}(.*)</oid_index_parse>
        <index_order>vsIndex</index_order>
        <index_order_type>alphabetic</index_order_type>

I've changed the original 32 to 33, 34, and 35, but my Poller Cache always shows the double dots (..):

Code: Select all

  	 SNMP Version: 2, Community: foo, OID: .1.3.6.1.4.1.3375.2.2.10.2.3.1.11..8.69.34.93.45.97.110.99.63
	RRD: /usr/share/cacti/site/rra/foo_vsconnections_4168.rrd 

Is there something else I should poke or restart or reset? I'm just not seeing any state change after modifying the regex -- shouldn't it break if I change the character number past a certain point? At the least, I'd expect it to change the output of whatever was in the Poller Cache.


Debian 4.0, Cacti 0.8.6i-3.5.

[edodd]

The answer to my own question is that no, contrary to all posted documentation, Cacti won't re-read the files in snmp-queries unless the device is deleted and re-added. At least, that's what worked for me.

[antifaction]

Does any one have a have to monitor the CPU's in there things yet?

I have been trying to monitor CPU0 but most of the values in the MIB tree don't show results that match the performance screen on the F5.

Anyone figure this out?

[bq]

I have all my graphs working except for memory and CPU. I'm not geting any errors, and the graph is blank ... anyone have any idea how to solve this? This is based off the template available via devcentral ... below is the debug output for memory and CPU.

/usr/local/rrdtool/bin/rrdtool graph - \
--imgformat=PNG \
--start=-86400 \
--end=-300 \
--title="bwi40b-scripps-bigip-1-p - TMM memory" \
--rigid \
--base=1024 \
--height=120 \
--width=500 \
--alt-autoscale-max \
--lower-limit=0 \
--vertical-label="Bytes" \
--slope-mode \
--font TITLE:12: \
--font AXIS:8: \
--font LEGEND:10: \
--font UNIT:8: \
DEF:a="/usr/local/cacti/rra/bwi40b-scripps-bigip-1-p_tmmmemtotal_90.rrd":TmmMemTotal:AVERAGE \
DEF:b="/usr/local/cacti/rra/bwi40b-scripps-bigip-1-p_tmmmemtotal_90.rrd":TmmMemUsed:AVERAGE \
AREA:a#0000FFFF:"Avilable memory" \
GPRINTLAST:"Current\:%8.2lf %s\n" \
AREA:b#FFAB00FF:"Memory in use" \
GPRINTLAST:" Current\:%8.2lf %s" \
GPRINTAVERAGE:"Average\:%8.2lf %s" \
GPRINTMAX:"Maximum\:%8.2lf %s\n" \
COMMENT:"The graph shows the total memory avilable to TMM compared to the memory in use." \
VRULE:1235970000#000000FF:""

RRDTool Says:

OK