An Oracle instance generates a lot of trace files being monitored by exporter

[AndriyDmytrenko]

Hi, We've discovered the issue recently, which is not directly related to the exporter itself, but Oracle itself has a bug about it, so this is mostly kind of suggestion. The story is following: we've discovered a lot of trace files in Oracle's diag directory - one trace per scraping event, many thousands files. We have found the support doc stating that no one should use third party monitoring tool (I don't agree with them). So I have investigated the issue and found that Oracle without ASM generates trace file for ASM-related queries. I've removed the query from v$asm_diskgroup from default-metrics.yml and there is no trace file anymore. So my suggestion is to document it somehow, or to move ASM-related stuff into separate metrics file and document how to enable it if ASM is used... Thanks.

[Yannig]

Thank you for your feedback. Do you think you can provide a entry in the README.md in the FAQ/Troubleshooting section?

It could be interesting to launch certain queries under conditions checked when launching the exporter. If the database does not support ASM, no need to launch the request for example.

[AndriyDmytrenko]

Hi, Yes, I think I will be able to write a few lines of description, but then I'll ask you to edit it since my English isn't so good yet. I was thinking about runtime detection of ASM presence, but didn't find a good way to do that. But the idea to ask ASM views once at launch time is good enough, one trace is better than thousands of them. I'll think what can I do here. Thanks.

[Yannig]

Don't worry too much about English. Being French myself and not necessarily very comfortable with English, a simple and short description should do just fine :)