How Big Should /var Be for ABRT to Run?

[dd_wizard]

I tried to report a Rhythmbox bug yesterday, but ran out of space on my 2GB /var partition while ABRT downloaded the debug info. I run /var/tmp as a tmpfs, but /var/chache/yum is in the normal /var LVM volume. How big should /var be? There was around 1.4 GB of free space on /var when the download started. Or is there a way to tell ABRT to skip downloading the debug info?

Thanks,
dd_wizard

[drunkahol]

There's no easy way to size /var as it's completely dependant on what apps you run.

/var is the default area for the www directory. Also mysql databases normally sit here. Squid defaults to using /var if you run it. As you pointed out, so does yum. It'll also depend on your log file rotation scheme (if you're doing any).

Unfortunately your question is akin to "how long is a piece of string"

Cheers

Duncan

[dd_wizard]

drunkahol replied:

Unfortunately your question is akin to "how long is a piece of string"

Good point!

A better question would be... How much free space is needed in /var to download the debug info needed to report an app crash with ABRT? As I mentioned, 1.4 GB allowed around 100 out of 103 debug info packages to download before I ran out of space. Is it necessary to download the debug info to use ABRT? That's a big chunk of bandwidth/disk space to tie up every time an app crashes.

dd_wizard

[SlowJet]

https://bugzilla.redhat.com/show_bug.cgi?id=529573

Are you not referring to yum downloads of rpms, not the actual abrt data cache?
The link abovet indicates it can be controlled, but to get a large update/install with the default packages I would estimate 2.6GB would do it and allow for some growth and less fragmented f/s.

SJ

[dd_wizard]

Thanks SlowJet, that's what I'm looking for.

dd_wizard