/run is too small for some clouds

Bug #1705765 reported by David-wahlstrom
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
CirrOS
Fix Released
Undecided
Unassigned

Bug Description

When cloud-init kicks off, it creates a daemon that uses /run to store data. It's very easy for the cloud provider to provide too much information in the config-drive/metadata service which causes cloud-init to crash due to /run not having enough space to store all the information. Currently, the /run mount looks like this:

tmpfs /run tmpfs rw,nosuid,relatime,mode=755 0 0

In order for cirros to work on DreamHost's OpenStack product, this needs to be bumped a bit. We are patching this to result in the following:

tmpfs /run tmpfs rw,nosuid,relatime,size=400k,mode=755 0 0

It would be fantastic if we could get this change committed upstream.

Revision history for this message
Robin H. Johnson (robbat2) wrote :

This was already fixed in commit 0041bbd2d888d6e3b47c88f79db45e19fe251932.
https://git.launchpad.net/cirros/commit/?id=0041bbd2d888d6e3b47c88f79db45e19fe251932

Changed in cirros:
status: New → Fix Committed
Revision history for this message
Dr. Jens Harbott (j-harbott) wrote :

This should be fixed in 0.4.0, please re-open if you still see an issue there.

Changed in cirros:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.