vSAN File Services node keeps restarting

Products

• VMware vSAN

Issue/Introduction

The time machine of container keeps failing back to original location/host after 3-4 mins as the vSan File Services VM is not stable.

In this case it fails back to the previous Host.

vSan File Services VM is not stable as seen in skyline in File Server Health

Environment

VSAN 7.0.3 - 2node ROBO + Witness

Cause

This is caused by the vSan File Services VM log disk has no space.

When reviewing the support logs collected for the vSan File Services VM we can see the logs were not collected due to space issues.

Copying fsvm cmd service log
  adding: fsvm_logs/ (stored 0%)
  adding: fsvm_logs/fsvmcmdserver.log.5 (deflated xx%)
  adding: fsvm_logs/containers/xxxxxxxx/samba/samba_logs/10.xxx.x.xx.log-xxxxxxxx-xxxxxxxxxx.gz (deflated x%)
  adding: fsvm_logs/containers/xxxxxxxx/samba/samba_logs/10.xxx.xx.xxx.log-xxxxxxxx-xxxxxxxxxx.gz
zip I/O error: No space left on device
zip error: Output file write failure (write error on zip file)

Resolution

Delete the vSan File Services VM as it is stateless and no containers will be running on it.

A new vSan File Services VM will be deployed automatically.

Additional Information

Workaround:

 

Delete old files on the vSan File Services VM log disk.



If you choose to try the workaround aim to get the usage below 2GB in order for the vSan Files Services to regain operational state.