No Space on RAM Disk

Ran into an issue lately with a VMware cluster that would not play nice when scanned with the update manager in Vcenter. The error that it was returning was something along the lines of “Not enough space on RAM disk”.

Turns out someone had configured the logging incorrectly and the scratch logs were being put into /tmp instead of persistent storage in order to resolve this I did the below:

Enable SSH on your host (More info Here)

cd /tmp | ls -la

This will tell you the size and details of the files that are sitting in your host’s temp folder. If you see a folder called Scratch this means that your logs were/are configured incorrectly.

The command below will let you check the space that is free on the host for each of the ramdisk mount points if you are getting the error about /tmp being full then it should say 0% free for the one in /tmp.

esxcli system visorfs ramdisk list

You can use the steps below to correct this issue.

In the VSphere console navigate to the configuration tab and click on advanced setting.

In the left hand menu select the heading SYSLOG and in the settings on the right change the to one of your datastores, leave the brackets in place they are important.

Click ok and now the directory for the logs is updated, you can now return to your ssh session on your host and repeat our first command to see if the scratch folder still exists, if so then you can run the below command to remove it.

rm -rf /tmp/scratch

You can now try running the update scan from VCenter on your host and it should go off without a hitch.