Cannot unmount an iSCSi Datastore because the file system is busy. Vsandevicemonitorlogd seems to be locked

[anthonyaudi]

Hello,

I'm trying to unmount an iSCSi datastore. It's a second datastore I have along with a vsan datastore. I want to remove it but I cannot. I configured the scratch logs to go to that datastore but as far as I can tell the vsandevicemonitorlogd.log is in use and I can't seem to delete it. I tried to unmount the datastore / delete it / I cannot do anything.

Can anyone tell me how I could get rid of the datastore cleanly. I can probably go in the iSCSi console and just delete it but then vmware would see it as a dead datastore.

Thanks

[TheBobkin]

Hello Anthony,

Firstly, this isn't a vSAN specific issue - you cannot remove a data store that is actively being used for logging.

Configure the logging for all hosts that have it set to this datastore to another location (this generally takes a reboot to take affect) and then try to detach it.

Bob

[anthonyaudi]

Hi Bob,

I accidentally put this in vSAN instead of vCenter. I was even looking for my thread in vCenter and couldn't find it

As for the logging, the only thing I initially configured was the scratch logs [] /stracth/log but I have since removed that, I tried to reboot one of the hosts also and still cannot remove the file.

Is there a service that I can possibly stop so I can remove this open file? Also, I see that it is vsandevicemonitord.log that is running, it this a critical file to my vSAN environment? If I stop it will my vSAN stop working? I know it's not recommended but can I put the scratch logs on the production vSAN datastore? <-- I believe that is not recommended correct?

Ideally, I wanted to put the scratch logs on my NFS store but every time I try I get a path not found error.

Thanks!

[TheBobkin]

Hello Anthony,

"As for the logging, the only thing I initially configured was the scratch logs [] /stracth/log but I have since removed that, I tried to reboot one of the hosts also and still cannot remove the file."

You need to reconfigure anything pointing to the datastore in question from ALL hosts and validate that they are indeed pointing to the new (even just temporary) location, best tip for this is to open vSphere Client > Host > Advanced Settings > search 'global' in the box.

"Also, I see that it is vsandevicemonitord.log that is running, it this a critical file to my vSAN environment? If I stop it will my vSAN stop working?"

This is a symptom, not the cause, nothing can be trying to write to files (log files or otherwise) to be able to remove a datastore.

"I know it's not recommended but can I put the scratch logs on the production vSAN datastore? <-- I believe that is not recommended correct?"

Correct, this is unsupported and not advised.

"Ideally, I wanted to put the scratch logs on my NFS store but every time I try I get a path not found error."

What are you putting as the file path and datastore name and in what syntax? Are you positive the sub-directory you are referencing exists?

VMware Knowledge Base

Bob

[tangroathi]

You have to reboot the hosts as this logdir-setting comes alive for the vsan-logging only after a reboot.