Nagios Support Forum

Hi,

We have detected a problem with one graph.

The check gives 24 datasources in perfdata output:
'tbs_users_usage_pct'=83.90%;90;98 'tbs_users_usage'=302420MB;324403;353238;0;360447 'tbs_users_alloc'=314541MB;;;0;360447 'tbs_undotbs1_usage_pct'=4.35%;90;98 'tbs_undotbs1_usage'=1425MB;29491;32112;0;32767 'tbs_undotbs1_alloc'=23328MB;;;0;32767 'tbs_temp_usage_pct'=7.56%;90;98 'tbs_temp_usage'=2478MB;29491;32112;0;32767 'tbs_temp_alloc'=19970MB;;;0;32767 'tbs_system_usage_pct'=2.66%;90;98 'tbs_system_usage'=872MB;29491;32112;0;32767 'tbs_system_alloc'=1024MB;;;0;32767 'tbs_sysaux_usage_pct'=4.45%;90;98 'tbs_sysaux_usage'=1457MB;29491;32112;0;32767 'tbs_sysaux_alloc'=2048MB;;;0;32767 'tbs_rdsadmin_usage_pct'=1.66%;90;98 'tbs_rdsadmin_usage'=67MB;3686;4014;0;4096 'tbs_rdsadmin_alloc'=1024MB;;;0;4096 'tbs_perfstat_usage_pct'=34.84%;90;98 'tbs_perfstat_usage'=3567MB;9216;10035;0;10240 'tbs_perfstat_alloc'=4024MB;;;0;10240 'tbs_indx_usage_pct'=90.91%;90;98 'tbs_indx_usage'=55854MB;13824;15052;0;15360 'tbs_indx_alloc'=61440MB;;;0;15360

But only de firts 21 datasources are graphing correctly, the last 3 are not generating any graph.

Is there any kind of limit of datasources in the same graph or rrd file?
How can we force to graph the 24 datasources?

Hi Scobos,

Could you generate a system profile and send it to me via Private Message ?

Also could you run the following as root and provide the output:

Thanks and Best Regards,
Keith

I have already sent it.

Hi scobos,

Received - the files are accessible to our support team for review.

It looks like you are trying to graph tablespace for an Oracle database. I didn't see anything that was out of proportion for the rrd files identifiable as Oracle servers which is good. We do have a couple of quick questions:

1) How long have you been graphing the tablespace for this host ?

2) When you started graphing it how many data sources were being sent (have they added new tablespace)?



Thanks and Best Regards,
Keith

Hi Keith,

About your questions:

1) I can´t remember. Maybe couple years ago.

2) This seems be the problem. We don´t know how many datasources are graphing at start, but seems that the number of datasources are increased. I believe we have RRD_STORAGE_TYPE as SINGLE in pnp4nagios configuration cause we never change it anywhere.

This is our pnp4nagios release:
cat /usr/local/nagios/etc/pnp/pnp4nagios_release
PKG_REL_DATE="05-02-2009"
PKG_VERSION="0.4.14"
PKG_NAME="pnp"

We have found the way to convert some rrd graphs from SINGLE to MULTIPLE, we are in 0.4 version instead 0.6
http://docs.pnp4nagios.org/pnp-0.6/rrd_convert

Anyway, we try it. We get the rrd_convert.pl script from other installation and try to get the command list:
./rrd_convert.pl --cfg_dir=/usr/local/nagios/etc/pnp/ --list_commands
XML Files analyzed 17545
XML Files found 0
XML Files without RRD 0
Old XML Files ignored 6410
Number of unique check_commands 66
\ List of Check Commands
|- check_xi_oracletablespace 281
...

And have tried a dry-run:
/usr/local/nagios/etc/pnp/rrd_convert.pl --cfg_dir=/usr/local/nagios/etc/pnp/ --check_command=check_xi_oracletablespace --dry-run

And I have moved the new rrd files to /usr/local/nagios/share/perfdata/HOST and launch the following, because xml file stay unchanged:
sed -i 's/SINGLE/MULTIPLE/g' ORACLE_Ocupacion_TS_MESPRD.xml

But the new rrd files are not upgraded, only is upgrading the original rrd file:
ORACLE_Ocupacion_TS_HOST.rrd

And again all datasources are SINGLE:
grep SINGLE ORACLE_Ocupacion_TS_HOST.xml
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>
<RRD_STORAGE_TYPE>SINGLE</RRD_STORAGE_TYPE>

Is there any wrong in these tasks?
Can we change from SINGLE to MULTIPLE?
Maybe it works without --dry-run?

XI doesn't support MULTIPLE type RRDs and RRDs will likely go away before development adds support for the MULTIPLE type.

You'll need to do this (this will fix all RRDs that don't have enough datasources):

https://support.nagios.com/kb/article/n ... g-149.html

Or to delete a datasource (this will fix an RRD that has too many datasources):

https://support.nagios.com/kb/article/n ... e-497.html

Or you can just delete the backend RRD file (or rename it) and the RRD will be rebuilt on the next check with the proper number of metrics, choosing this method will cause you to lose the historical performance data for the service though.

Hi,

We try the first option and seems work fine.

Many thanks

Hi,

We try the first option and seems work fine.
Many thanks

Great, did you need assistance with anything else or shall we close this out?

--Benjamin