[Bug 1281074] Re: updatedb.conf: CephFS should be included in PRUNEFS

[Livio Soares]

Just wanted to cast my vote for adding "ceph" to the list of pruned
file-system types. We have a fairly large Ceph file-system that is
mounted throughout a cluster. Having the entire cluster crawl the file-
system in synchronized form (due to cron) is really disruptive. Not only
does it make the entire cluster halt due to network storm (causing all
sorts of problems on the ceph mounts), it also typically causes problems
on the Ceph servers (page allocation errors in kernel/kworker threads,
etc).

Ceph should clearly be an opt-in for updatedb, definitely not the
default behavior.

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to mlocate in Ubuntu.
https://bugs.launchpad.net/bugs/1281074

Title:
  updatedb.conf: CephFS should be included in PRUNEFS

Status in “mlocate” package in Ubuntu:
  Confirmed

Bug description:
  CephFS filesystems can become huge. It's networking filesystem for a
  client and clients should not index data on CephFS filesystems.

  I suggest that "ceph" and "fuse.ceph" is added to "PRUNEFS" in
  /etc/updatedb.conf to prevent all the clients to start indexing this
  network filesystem.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/mlocate/+bug/1281074/+subscriptions