List Archive: gentoo-cluster
Note: Due to technical difficulties, the Archives are currently not up to date.
provides an alternative service for most mailing lists.c.f. bug 424647
Brian Kroth wrote:
> I currently manage about 40 Window, OSX, and Hardened Gentoo servers.
> I will soon have 12 P4 servers that were previously used as video
> encoders free as well as an Apple XRaid. With all this spare hardware
> I thought I'd research setting up a cluster of servers running Apache
> for load balancing and high availability. I'm also looking into a
> MySQL cluster, but that wouldn't require a shared filesystem. I'm
> wondering if anyone has done something like this before and in
> particular knows a good filesystem to use so that each of the servers
> can access and potentially write to the same storage array. I've
> accomplished the same thing with XServes running OSX, but they like to
> charge you a pretty penny for the XSan software that allows this which
> I thought I'd try to avoid if possible. So far I've seen only GFS,
> but haven't gotten much reading done on it yet. Any other tips or
> insights would be appreciated as well.
I currently manage 4 different shared storage systems on gentoo with
SAN-storage based on AoE (coraid 1520 machines)
We choose GFS for the filesystem.
It works, although the throughput is not spectacular.
This has partly to do with GFS and partly because the underlying storage
hardware doesn't support faster throughput.
If you have any specific questions about GFS I'd be happy to answer them.
Apart from that we're currently planning a Lustre deploy for fast-storage.
This is still very much in alpha, so no comments yet.
I've looked at implementing a mysql-cluster in great detail and let it
go because it's mostly meant for a setup with a fixed database-size.
If you experience (large) growth in your database volume clustering is
not very suitable (we're in that situation)
We're currently load-balancing our databases with LVS (ipvsadm) and a
multi-master replication setup.
This works out really well for us, serving estimated average 15Kq/s
with 50 servers on a very large dataset (> 40Gb) and a fairly random
I would strongly disadvise running multiple databases off GFS.
>From what I gathered it works well for active/passive failover sets
and/or active/active failover, but performance will be hurt compared to
a "normal" fs.
firstname.lastname@example.org mailing list