losing data values for disk space (only in discoverd items)

[traxman]

Hi guys,

I have zabbix 5.0.40 running for more than 3y and it works quite well. Few months ago I noticed that some servers that are in vmware vCenter are randomly losing data. By that I mean when I go to “last data” graphs I don’t see data in graphs more than few hours or 2-3 days, and there should be 60d data gathering. Housekeeping and template “Template VM VMware Guest” is set to 60d and trend to 365d.
“Probably” most of the virtual servers (around 640 VMs in vmware cluster) have 60d history data but from time to time I run into this situation that VM has only few hours or days history data.
Can someone give me some hint what could be the problem or where to look?

In Zabbix_server log sometimes I see records that say:
vmware.vm.vfs.fs.size[{$URL},{HOST.HOST},C:\,free]" became unsupported
vmware.vm.vfs.fs.size[{$URL},{HOST.HOST},C:\,free]" became supported
And when I check ID from that record in zabbix gui I see that this record belongs to that problematic VM. And one more thing. This happens only to “Mounted filesystem discovery” items. There are few more non-discovery items in all VMs, and these items have normal history data (60d)

[traxman]

I managed to figure out why is this happening.
Few months ago, I changed (because of some other reasons) the value for template “Template VM VMware Guest”:
“Mounted filesystem discovery”  Keep lost resources period = 0 (zero).
And this is why the values are deleted sometimes daily.

But there is still a problem why the items “vmware.vm.vfs.fs.size” are becoming unsupported.
I figure out that too, because of the vmware tools. Sometimes in linux/windows VMs vmware tools stop working or something and file system items become unsupported and deleted with value “0”. And this will happen again, but will not be deleted because the value is returned to 30d.
The mystery that remains is “vmware tools” and their buggy process. I don’t know why is this happening in vCenter cluster in random order and after few hours item again becomes supported and value gathering process continue to work normally. Except when the vmware tools stop completely and vmware says that vmware tools are not running. Than the items stay unsupported forever or until machine is restarted or vmtools started manually.
And idea why?