[CentOS] /run and /var/run as tmpfs - how many RPM's broken?

Wed Aug 26 08:52:39 UTC 2015
Fabian Arrotin <arrfab at centos.org>

Hash: SHA1

On 26/08/15 10:28, Gary Stainburn wrote:
> This is pretty much a retorical question, although I would like to
> know the extent of the problem and how quickly it's gonna be
> sorted.
> My problem has already been described in a previous post, i.e.
> clamd fails after a reboot.
> The cause apparently is that /run and therefore /var/run has at
> some point been moved onto tmpfs which means that it doesn't
> survive a reboot.
> The effect of this is that any folders (such as /var/run/clamd.*
> that gets created when RPMs are installed don't survive, and the
> first time the server is rebooted things get broken.
> Googling the problem has shown at least a couple of other packages
> now experiencing the same problem. including from memory pcp and
> MongoDB.
> (BTW, I'm still receiving emails from that too good to be true
> young lady)

Solution for packages not ready for el7 is to use tmpfiles.d :

For the other issue (spam), as said in a previous email it's not
coming through the centos mail server[s] but an abuse report has be
sent to the cloud provider hosting the VMs used to send those spam
messages directly to the list users.


- -- 
Fabian Arrotin
The CentOS Project | http://www.centos.org
gpg key: 56BEC54E | twitter: @arrfab
Version: GnuPG v2.0.22 (GNU/Linux)