synchronizing cobbler installations

[Jeremy Mordkoff]

Hi all,


I have a need to synchronize the content and config between multiple instances of cobbler.
Each instance is running in a VM and new instances are created by cloning the VM image. 
These instances are spread around geographically and some do not have very fast internet
connections.


For config, I have /var/lib/cobbler and /etc/cobbler in source control. I commit my
changes on the master instance and then update on all of the slave instances and then
restart cobbler. Seems to work well and my git repo is already accessible over the
internet so this is convenient.


But I am not so keen on putting my repo mirrors in source control because they are large.
Plus I have a pip wheelhouse and an NPM proxy cache.


Any suggestions on how to keep these all in sync? I am considering just using rsync, but
that will require moving my master instance (or a copy thereof) someplace accessible by
all.


I guess this isn't a cobbler question at all, but perhaps someone has some insights
that might help me.


Thanks

---
Jeremy Mordkoff
RIFT.io
77 S Bedford St, Burlington, MA 01803

[Alastair Munro]

Hi

Cobbler has replication features. See the docs: 
http://www.cobblerd.org/manuals/2.6.0/3/2/8_-_Replication.html

I also wrote this last year, which has been working well for us: 
https://github.com/cobbler/cobbler/wiki/Speeding%20up%20triggers%20to%20s...

You might want to look at katello/satellite 6, which has a capsule 
system based on The Foreman. In essence you have a capsule/proxy per 
site, which does all your repo mirroring, local dhcp/tftp/etc. This 
would require alot more effort.

Alastair Munro



On 2015-04-30 23:53, Jeremy Mordkoff wrote:
>  Hi all,
>  
>  I have a need to synchronize the content and config between multiple
>  instances of cobbler. Each instance is running in a VM and new
>  instances are created by cloning the VM image. These instances are
>  spread around geographically and some do not have very fast internet
>  connections.
>  
>  For config, I have /var/lib/cobbler and /etc/cobbler in source
>  control. I commit my changes on the master instance and then update on
>  all of the slave instances and then restart cobbler. Seems to work
>  well and my git repo is already accessible over the internet so this
>  is convenient.
>  
>  But I am not so keen on putting my repo mirrors in source control
>  because they are large. Plus I have a pip wheelhouse and an NPM proxy
>  cache.
>  
>  Any suggestions on how to keep these all in sync? I am considering
>  just using rsync, but that will require moving my master instance (or
>  a copy thereof) someplace accessible by all.
>  
>  I guess this isn't a cobbler question at all, but perhaps someone has
>  some insights that might help me.
>  
>  Thanks
>  
>  ---
>  Jeremy Mordkoff
>  
>  RIFT.io
>  
>  77 S Bedford St, Burlington, MA 01803
>  
>  
>  _______________________________________________
>  cobbler mailing list
>  cobbler(a)lists.fedorahosted.org
>  https://lists.fedorahosted.org/mailman/listinfo/cobbler