CentOS7 linux metrics errors logged after update from 1.17.1 to 1.17.2.

I updated pmm-client a week ago an today noticed 180MB of pmm-linux-metrics log filed with:

time=“2019-11-03T22:35:51+02:00” level=error msg=“Error reading textfile collector directory /usr/local/percona/pmm-client/textfile_collector: open /usr/local/percona/pmm-client/textfile_collector: no such file or directory” source=“text
file.go:81”

I downgraded to 1.17.1 and the messages are gone.

tools-release-x86_64
pmm-client-1.17.1-1.el7.x86_64
pmm-client-1.17.2-1.el7.x86_64

/usr/local/percona/pmm-client/textfile_collector should actually be textfile-collector, however when I renamed textfile-collector to textfile_collector, some other error appeared in the log.

Anyone else seeing these errors?

If you could by any chance let us know the other errors you mention (when you renamed the file) that could be useful.
I checked in our bugs database and could not find anything related to this.It may be that someone can help you with this but given the symptoms (i.e. no errors when you downgrade) I wonder if you would be willing to record this in our Jira system? That way the engineering team will take a look and you can follow their assessment/advice.
Information on raising Jira tickets can be found here: https://www.percona.com/forums/questions-discussions/percona-monitoring-and-management/50690-pmm-troubleshooting-and-how-to-report-a-bug
The URL is https://jira.percona.com

I just upgraded again and then mv /usr/local/percona/pmm-client/textfile-collector /usr/local/percona/pmm-client/textfile_collector - so far no errors in the log. Too bad I didn’t make a note of the error after renaming it earlier. All in all there must be a typo somewhere in the code. :slight_smile: I’ll bugrep it later.

That would be great if you could…
And thank you indeed for updating this post! Appreciated!

https://jira.percona.com/browse/PMM-4923

It went under packaging, but that might not be right. :slight_smile:

I’m seeing the same issue. After renaming the directory, I also see the following error logged every 1 sec:
[INDENT]