why doesn't yum cache anything?

Sean Middleditch elanthis at awesomeplay.com
Fri Dec 31 06:44:10 UTC 2004


On Thu, 2004-12-30 at 22:36 -0800, Jamie Zawinski wrote:
> Here's a typical use case for yum with me:
> 
>   - notice that "xmtr" is not installed (but "mtr" is)
>   - yum list '*mtr*'
>   - wait through ~60 seconds of:
> 
>     Setting up Repo:  base
>     repomd.xml                100% |=================| 1.1 kB   
> 00:00     
>     Setting up Repo:  updates-released
>     repomd.xml                100% |=================|  951 B   
> 00:00     
>     Reading repository metadata in from local files
>     base      : ############################################# 2622/2622
>     updates-re: ############################################# 405/405
> 
>   - yum -y install mtr-gtk
>   - wait through that same ~60 seconds of junk before it actually
>     starts downloading the package.
> 
> So my questions are:
> 
>   - Is there some way to make yum cache all that crap?  I'd be 

All of the above *IS* the cache.  The repomd.xml lookups are the only
non-cache bits - those are the files yum checks online to see if there
are only real updates.  Notice their very small size.

The later bits (the hash marks) are yum reading the cache files from the
disk.

The problem you're perceiving (slow operation as yum starts up) isn't at
all due to lack of caching, but perhaps very inefficient handling of the
cache - a lot of data has to be parsed and such, when it could perhaps
be stored in a more ready-to-process format.




More information about the fedora-devel-list mailing list