yum update of glibc needs space that's already there

[Thiers Botelho]

Hi Duncan,
 
[for shortness - following broken thread under 
http://www.redhat.com/archives/fedora-list/2004-April/msg00686.html ]
 

On Mon, 05 Apr 2004 08:19:41 -0500 "duncan brown" <duncanbrown email com> 
asked:
 

|is this a server or a workstation? 

That's a dual boot workstation box along with win2k prof .


|you should probably try cleaning out various cache files you may have 
sitting around, 

Well for the moment I'll treat them on a case by case basis until the 
FINAL solution (whatever that might be).  :/


|a 2GB drive (sadly) doesn't give you alot of room to play around with in 
a modern distribution.

I know . . . if no simpler solution comes by I'll do a complete refitting 
of both system's partitions with Partition Magic. But I'm still putting 
the final touches on my backup solution, so I'm not quite ready for that 
yet.


|
|[duncan atom Cache]$ pwd
|/home/duncan/.mozilla/default/1dky1wp0.slt/Cache
|[duncan atom Cache]$ du -sh .
|34M     .
|
|that'll free up some space, if you try that... i have a script that i run 
every morning and at least once a day that cleans out this dir...
|
|rm -fv ~/.mozilla/default/*/Cache/*

I just used explore2fs from windoze to visually inspect mozilla caches of 
root and of the other user. I wouldn't gain more than 4 Mb if I cleaned 
both.

|
|it sounds like you may need to do some spring cleaning, or hunting for 
packages you never use to clean up that last bit of space.

ooo is a serious candidate for disintegration, but I don't think it should 
all end this way (yet)    :((


|-d

???


|have you tried to install these upgrades by hand?  and MAKE SURE you 
install the one appropriate to your archetecture with glibc 
(i386/i686/etc), otherwise you'll hose your machine.

No, this I wouldn't like to do. At least not without trying first to 
massage all partitions as explained above, which I think will ease further 
occurrences of the problem.

Cheers

Thiers

|
|----- Original Message -----
|From: "Thiers Botelho" <thiers fosfertil-ultrafertil com br>
|Date: Mon, 5 Apr 2004 10:11:42 -0300
|To: fedora-list redhat com
|Subject: Re: yum update of glibc needs space that's already there
|
|> Hi Duncan,
|> 
|>  On Mon, 05 Apr 2004 07:51:34 -0500 "duncan brown" <duncanbrown email 
com> 
|> suggested:
|> 
|> |... you should clean out your yum cache and try this again
|> 
|> |[root atom yum]# yum clean
|> |[ snip ]
|> |Cleaning packages and old headers
|> [root atom yum]#
|> 
|> 
|> Well, I hadn't mentioned it for simplicity's sake. But my update 
procedure 
|> is (ALWAYS):
|> 
|> ==================================================
|> df -m
|> yum clean
|> df -m
|> yum update xyz* *** 2 or 3 packages at a time
|> df -m
|> yum clean
|> df -m
|> ==================================================
|> 
|> 
|> On this particular case I also ran command block below (pasted and 
adapted 
|> from /etc/cron.daily/tmpwatch )
|> 
|> ==================================================
|> /usr/sbin/tmpwatch 120 /tmp
|> /usr/sbin/tmpwatch 120 /var/tmp
|> for d in /var/{cache/man,catman}/{cat?,X11R6/cat?,local/cat?}; do
|>     if [ -d "$d" ]; then
|>         /usr/sbin/tmpwatch -f 120 $d
|>     fi
|> done
|> ==================================================
|> 
|> 
|> And noticed no difference whatsoever under 'df -m' .     :((
|> 
|> Thiers