Page MenuHomePhabricator

Delete /srv/vdb/varnish* files on all Varnish instances
Closed, ResolvedPublic

Description

Varnish had a bug that prevented it from starting (Could not mmap SILO...).

The beta cluster upload varnish cache to bails out with 503 yesterday ( T75922: upload.beta.wmflabs.org is throwing 503s ). The fix is quite easy:

/etc/init.d/varnish stop
rm /srv/vdb/varnish.*
/etc/init.d/varnish start

Brandon Black investigated yesterday failure on T75943: Investigate Varnish 3.0.5plus~x-wm7 throwing on beta cluster "Could not mmap SILO" and confirmed that the varnish version should not be impacted. On the other hand, previously existing file would still cause the bug.

This task request the deletion of /srv/vdb/varnish.* files on all beta cluster varnish instance so now they are all in a clean state.

Event Timeline

hashar raised the priority of this task from to Needs Triage.
hashar updated the task description. (Show Details)
hashar changed Security from none to None.
hashar subscribed.
hashar claimed this task.

I have deleted the files using: service varnish stop && rm -v *; service varnish start

On instances:

  • deployment-cache-mobile03
  • deployment-cache-text02
  • deployment-cache-upload02
  • deployment-sca-cache01
  • deployment-parsoidcache02