I’m seeing lots of this in /var/log/vmware/applmgmt/StatsMonitor.log (on vCSA 6.5.0.12000 Build Number 7119157)
2017-12-19T10:53:45.451+01:00 error StatsMonitor[7FD04F4AA700] [Originator@6876 sub=LinuxStorageStatsProvider(200148195552)] statvfs(/storage/invsvc) failed with errno 2
Seems there is a check on a directory every 10 minutes, but the directory does not exist. I think it used to be there in an earlier version, but not anymore.
So is this maybe a bug in the StatsMonitor that it still tries to check it? Should I just create an empty folder?
I dug a bit deeper, and found the config file for StatsMonitor (/etc/vmware/statsmonitor/statsMonitor.xml) and it contains config for /storage/invsvc
<!-- File systems to monitor for utilization. -->
...
<filesystem id="invsvc">
<name>invsvc</name>
<path>/storage/invsvc</path>
</filesystem>
Maybe this should have been removed from the vCSA65?
According to the 6.5U1 release notes it seems some alerts, like for invsvc, should have been removed..
snip from resolved Issue:
Alarm definitions for deprecated services are still visible in the user interface
Alarm definitions for certain deprecated services have not been updated. As a result, you can still see alarm definitions for deprecated services in the vSphere user interface.
This issue is resolved in this release.
Alarms have been deprecated for the following services: ts, vws, vmware-syslog, invsvc
Alarms have been updated for the following services: content-library, vpxd-svcs
So should I create this directory or should I update the configuration?