Hot answers tagged

8

Congratulations, you found an error in the RHEL documentation. The iptables rule shown here uses the tcp matcher but then attempts to specify the udp protocol. This doesn't work; the tcp matcher can only be used with protocol 6, which happens to be tcp. Thus the error you received. To correct the rule you have to first figure out which is wrong, the ...


7

Yes, what you're suggesting is possible, and precisely what you'd use these tools in combination for. You'd use Clonezilla or Cobbler to push out the initial installs, and Puppet to keep the configurations in sync once they're installed. (And FYI, the other popular configuration manager for Linux you didn't mention is called Chef.) Clonezilla doesn't have ...


3

This doesn't work like that, as Cobbler is unable to handle the mirror list functionality. The --mirror option must specify a specific URL you want to download the repo from.


3

You don't necessarily need to run your own DHCP/DNS server with a cobbler installation, but in order to boot properly via PXE, the DHCP server must send additional attributes to the clients (at least next-server and filename which contain the hostname of the TFTP server and the filename to load from it). If you can't have these attributes added to the DHCP ...


3

Answering my own question My filesystem is XFS, on top of LVM partitions. The best solution was to resize /var partition so that it has enough space to support the rsync operations. The thing is when I installed CentOS 7 on this machine I've allocated all the available disk space, so there was no free space to enlarge /var. So I decided to move /var/www/...


2

You haven't added the keys to the gpg keychain. I am running a near identical configuration to yours (CentOS 6.6/trusty) and resolved this particular issue by passing the flag --no-check-gpg. A more ideal solution would be to download and manually add the keys to your keyring.


2

Yes, managing a DHCP server is completely optional for Cobbler. Just set manage_dhcp: 0 in your /etc/cobbler/settings file.


2

The daemon's dying unexpectedly on launch (status=1). It looks like the arguments to tftpd that you have are inappropriate for your current tftp version. Running the daemon from the shell has brought this into view, you have made the server_args parameter make sense for the current daemon, and all is now well.


1

The winning solution was to move to GRUB2 from GRUB Legacy (0.97). There is a straight-line distance of about 120 miles between the DC where the server being booted is located, and the one where the TFTP/DHCP server is located. This necessarily causes some latency. Since TFTP is a non-windowing protocol, a packet has to be acknowledged before the next ...


1

Simple copy to the repository isn't enough. You have to recreate repodata. If it's RHEL/CentOS you may use: createrepo -d <path to your YUM repo> if it's system repo, you may have to add -g comps.xml, otherwise PXE installation will fail: createrepo -d -g comps.xml <path to your YUM repo>


1

Cobbler uses Cheetah as the templating engine and according to the documentation, $hostname is a built-in variable that you can use. You can see other variables with; cobbler system dumpvars --name=system


1

If you're turning on the rewrite engine anyway, just use that to match the location: RewriteRule ^/cobbler_web/ https://%{SERVER_NAME}/%{REQUEST_URI} [R,L]


1

You might have some other competing DHCP server in your net. Running a traffic capture with Wireshark can help you pinpointing this problem. BTW range dynamic-bootp 10.13.0.20 10.13.0.150; option routers 10.13.0.138; the router's IP is within the assignation range; not good.


1

I know this is a basic thing, but did you cobbler sync after making your changes? You can do this via GUI or CLI. Otherwise, they may not have taken effect.


1

I myself have seen cases where managing 1000 hosts with different configs, is much easier with puppet. Infact companies like google uses puppet for their deployment. The main design architecture of puppet is such that, it works much better than others if you configure it in the right way. For example adding your custom facts for your custom configs etc. the ...


1

Unless you set explicit dependencies, the ordering of resources may be a problem. The way you wrote it does not mean that the repo will be deployed before applying the other classes. For example this: node server1 { class { 'myrepo': } class { 'bacula': require => Class['myrepo'], } class { 'vsftpd'...


1

I would suggest to use cobbler to run the very basic ansible playbook, that doesn't need to reboot server and configures Ansible pull mode After reboot ansible should pull another playbook from its server and finishes server configuration.


1

Could you have a wrapper script in /etc/rc.local (or similar) that runs the Ansible playbook the first time the system boots, and then removes itself? Either with a rm $0 at the end of the script or by editing /etc/rc.local once complete?


1

Cobbler will download the latest 'update' distro_signatures.json using the command below: # cobbler signature update


1

Apparently, yes. Take a look at this Trac ticket on how to "allow templating variables to be used in --kopts". With regards to koan, the comments in the same ticket are informative. Relevant part reproduced here for completeness: This is implemented with @@variablename@@ for anything that runs through the templating engine (kickstart + PXE stuff). koan's ...


1

You can also consider (R)?ex, it is easy to learn, and helps in configuration management and software deployment. It needs SSH or HTTP on his nodes.


Only top voted, non community-wiki answers of a minimum length are eligible