|
> Good Morning Friends .
> Im a new member on these list. Hope i can learn a lot from u all.
Hello, welcome!
> About this subject.... This year im going to make a
> graduation project with the subject of High Availability and LVS.
> Im planning to make a project with these design :
> 2 Computers running as the Virtual Server with HA (Heartbeat
> and DRDB) .
> 3 - 4 Real Servers - Using IDE Disks, since I cant afford
> for external storage and the university dont have it.
I like IDE disks now. They are fast and cheap.
> So ... last weeks i was reading a lot of things about HA and
> LVS integration...
> My point is the same that Ron is asking ....
> How can i make the replication on these 3 Real Servers since
> its going to use IDE Disks. Does Rsync makes the job on this
> case ? Is there any other method available ?
Rsync is probably the tool you want for your job. It will only push/pull
the bits that have changed; thus its inherantly more efficient. I also like
that there is no single point of failure and is pretty simple & reliable
software. Other methods you might want to consider are DRDB, NFS, and
expect scripts.
Maybe you will find this thread interesting :
http://marc.theaimsgroup.com/?l=linux-virtual-server&m=102932527302836&w=2
> On this scenario should i have a "Master Server" in one of
> this 3 Servers ? So i make all the alterations on this one,
> then replication process starts from this Master to the other
> 2 servers ? Is that right or i dont need a "master" one?
> I was searching for some good documentation on using rsync on
> this scenario, but didnt find anything. Any docs u guys can
> indicate me?
Sounds right. If I were you I'd run rsync through SSH (-e ssh) and setup
automatic key acceptance. I think there are some good posts in the archives
(http://marc.theaimsgroup.com/?l=linux-virtual-server) about this.
Cheers,
P
|
