[rhelv6-list] Fun with new RHEL

[Chris Adams]

Once upon a time, Chris Adams <cmadams at hiwaay.net> said:
> - My base kickstart doesn't work.  I create the filesystems in a %pre
>   and then %include a chunk that defines them (doing it this way to get
>   desired options for things like mdadm).
> 
>   Using some debugging, I see my LVs and filesystems are created, but
>   then anaconda puts up a message about "examining storage devices", and
>   the device nodes in /dev/mapper are all gone.  It only gets /boot (not
>   on LV) mounted, so then it complains that "/" isn't big enough
>   (because it doesn't exist).
> 
>   This is at least a couple of bugs:
>   - anaconda kills the existing LV device nodes
>   - anaconda doesn't notice that it didn't even mount a root filesystem

I found that if I let anaconda make the filesystems (took off the
--noformat from the "logvol" lines in the kickstart), it reactivates the
LVs as it goes and is happy.  I guess I'll bugzilla this.

Another odd thing: I have for years used a local Squid proxy for updates
(speeds up updating a bunch of similar systems and allows me to manage
the bandwidth consumed in one place).  To cache the RPMs, I've set
"useNoSSLForPackages=1" in /etc/sysconfig/rhn/up2date.  Now when I do
that, I get:

# yum update
Loaded plugins: rhnplugin
rhel-x86_64-server-6                                     | 1.8 kB     00:00     
Error: failed to retrieve repodata/6faecb305efb123bd886342dd108b407fc2b14ace71b46e66a675209e97da51a-primary.xml.gz from rhel-x86_64-server-6
error was [Errno 14] Peer cert cannot be verified or peer cert invalid

-- 
Chris Adams <cmadams at hiwaay.net>
Systems and Network Administrator - HiWAAY Internet Services
I don't speak for anybody but myself - that's enough trouble.