Fresh Centos Install Problem

[timdorr]

I've installed Centos 4.1 i386 and cPanel today, and logging into WHM, I get this at the top of every page that's not some sort of function page (i.e. Limit Bandwidth Usage, or Tweak Settings):

[a fatal error or timeout occurred while processing this directive]

In the cPanel error_log, there is this:

(internal death) Tue Sep 13 19:50:31 2005 [7609] error: error setting uid (0) [root]
Cpanel::SafetyBits::setuids('root') called at whostmgr/bin/whostmgr.pl line 6569
main::themecommand() called at whostmgr/bin/whostmgr.pl line 204
error setting uid (0) [root]



My assumption is that I'm installing Centos wrong, but it could be something else. Any thoughts as to what is up?

As for installing Centos, it is right to select the Server option and uncheck everything except the Development tools, Web Server, and Editors (for my own use) options? What settings should I be using for a fresh Centos install?

 

[randomuser]

I've done 2 back to back installs today on RedHat AS3 and am getting the same thing. cPanel was not communicating with BIND either. I just got done reinstalling again and see those same errors at the top of the left and right frames:

[a fatal error or timeout occurred while processing this directive]

Attempting to add a zone is useless. Something appears to be horribly broken and I'd add more information if I knew more at this time.

edit: another issue that showed up was chkservd not starting due to missing the Net-AIM perl module. The fix is easy:

cd /tmp
wget http://search.cpan.org/CPAN/authors/id/A/AR/ARYEH/Net-AIM-1.22.tgz
tar zxvf Net-AIM-1.22.tgz
cd Net-AIM
perl Makefile.pl
make
make install

service chkservd start




Not having any zones is pretty silly.

edit: WHM 10.6.0 cPanel 10.6.0-R201
RedHat Enterprise 3 i686 - WHM X v3.1.0

 

[timdorr]

BIND is easy to fix. Edit /etc/sysconfig/named and add a # in front of the ROOTDIR line.

I'm guessing something should or shouldn't be installed from my configuration. Hopefully someone knows how to do the install right.

 

[randomuser]

This is the default after the setup:


# more /etc/sysconfig/named
# Currently, you can use the following options:
# ROOTDIR="/some/where" -- will run named in a chroot environment.
# you must set up the chroot environment before
# doing this.
# OPTIONS="whatever" -- These additional options will be passed to named
# at startup. Don't add -t here, use ROOTDIR instead.


Adding an A entry is useless due to the lack of a zone. After "creating" the zone - the zone does not exist although it reports the DNS entry was created.

I'll get out of your thread now, cPanel is obviously not communicating with BIND, and I need to pursue another option. Good luck with the CentOS issues.

 

[timdorr]

Try /scripts/fixndc twice. I always had to do that under RHEL 3. You should get a response back when running rndc status, too.

 

[randomuser]

I'd ran /scripts/fixndc after both installs, but only 1 time.

At this point I've already edited /scripts/cpupdate.conf and changed release to stable, and reran upcp. This brought things down to WHM 10.1.0 cPanel 10.2.0-S83. After doing this I was able to add a zone just fine, and the directive errors went away.

Note to anyone else who may be in the same boat and considering doing this:

I've been told this may cause permission errors somewhere along the way, and that what may work would be to rm all the scripts in /scripts, and to rm /usr/local/cpanel and /var/cpanel. Then edit /etc/cpupdate.conf to =stable, and rerun "sh latest". I have not tried this method myself.

The best method from what I hear would just be to start over and install stable and not release (I would have to agree).

 

[timdorr]

Oh, but it's no fun running Stable :D

I'll try a minimal-only install tomorrow and try a yum install gcc to see if that does the trick. It's the only thing I can think of at this point.

 

[randomuser]

Oh, but it's no fun running Stable :D

hahaha you are definitely not speaking the words of someone who had to install the same VDS 3 times today (4 if you count the downgrade). Running stable is very fun at this point, because it means I can go do something else now

I'm sorry I put all this junk in your thread - I just did a search for the directive issue and replied to the first thread I saw without really thinking too much about your issue (I figured they may be related, but it's hard to tell really). I really think that 10.6.x might be a little broken. In fact, I think I'd heard that the same day it came out. Strange that it's worked on other VDSes just fine recently.

Here's what worked for me (as far as I can tell, nothing back from the customer yet but I don't expect any issues):

1. destroyed and recreated the VDS
2. echo "CPANEL=stable" > /etc/cpupdate.conf
3. wget http://layer1.cpanel.net/latest && screen sh latest

Good luck timdorr, wish I could help.

 

[timdorr]

Downgraded to Stable
Upgraded to Release

That seems to have done the trick, but it's not really optimal :/ Is there any way to force the installer to go to Stable? Or maybe the cPanel guys can just fix the issue :P

 

[randomuser]

Yeah just do steps 2 and 3 from what I put above. That forced the install to go to stable instead of the default "release". I did it with a fresh install of RedHat AS3 and it worked great.

 

[timdorr]

BTW, bug filed: http://bugzilla.cpanel.net/show_bug.cgi?id=3210

 

[btrieve]

hahaha you are definitely not speaking the words of someone who had to install the same VDS 3 times today (4 if you count the downgrade). Running stable is very fun at this point, because it means I can go do something else now

I'm sorry I put all this junk in your thread - I just did a search for the directive issue and replied to the first thread I saw without really thinking too much about your issue (I figured they may be related, but it's hard to tell really). I really think that 10.6.x might be a little broken. In fact, I think I'd heard that the same day it came out. Strange that it's worked on other VDSes just fine recently.

Here's what worked for me (as far as I can tell, nothing back from the customer yet but I don't expect any issues):

1. destroyed and recreated the VDS
2. echo "CPANEL=stable" > /etc/cpupdate.conf
3. wget http://layer1.cpanel.net/latest && screen sh latest

Good luck timdorr, wish I could help.

Out of curiosity what are you using for VDS software over redhat?