Oracle_asm_diskgroup error

[Infonet]

We’re using a oracle_asm_diskgroup to check the Diskgroup on Redhat Server.
The error that I am occuring is this one:

[agent] Execute data source
[piggyback] Execute data source
No piggyback files for ‘PDREDHAT01’. Skip processing.
No piggyback files for ‘10.250.0.252’. Skip processing.
WARNING: Exception while parsing agent section ‘oracle_asm_diskgroup’: ValueError(“invalid literal for int() with base 10: ‘ASMCFG/’”,)
File “/omd/sites/stevanatogroup/lib/python/cmk_base/data_sources/host_sections.py”, line 299, in _update_with_parse_function
return parse_function(section_content)
File “/omd/sites/stevanatogroup/share/check_mk/checks/oracle_asm_diskgroup”, line 157, in parse_oracle_asm_diskgroup
“fg_disks”: int(fg_disks),

systemd_units does not support discovery. Skipping it.
ps_lnx does not support discovery. Skipping it.
ps.perf does not support discovery. Skipping it.
Completed.

If on the console we run the command this is the output:

Any idea how we can fix this issue? We have other 2 server monitor like this and they’re working properly (same config files).

Samuele

[tosch (Tobias Schönau)]

Which version are you running the agent and you monitoring server?

[Infonet]

Hi tosch,
we are using version 1.6.0p6 (agent) and 1.6.0p6 (server).
Samuele

[tosch (Tobias Schönau)]

In this version the ASM-checks should be included within the mk_oracle plugin already.
Please delete the mk_oracle_asm file and configure your plugin (mk_oracle.cfg) for ASM monitoring.

Monitoring Oracle databases efficiently

[Infonet]

Hi again,
what I’m trying to understand is how from CLI is working and from Webconsole is not?
On the other 2 servers we have exactly the same conf.

[Rendanic (Thorsten Bruhns)]

Hi,
the output looks like an old linux plugin, due to missing ‘|’ field separator in agent output.
I added ASM SQLs to mk_oracle some years ago, due to limitions in asmcmd command which was used in mk_oracle_asm.

Kind Regards
Thorsten

[tosch (Tobias Schönau)]

It is working from CLI because it just outputs. Every check has a counter part on the monitoring server which is parsing the output and so needs to understand the output the script on the client has produced. This output is different if you use the mk_oracle_asm script or the functions now included inside the mk_oracle script.

Maybe you have some different outputs on the other two servers which doesn’t produce an incompatable output and seem to work.