[Linux-cluster] Re: Linux-cluster Digest, Vol 20, Issue 12

[Alan Wood]

DC,

While you did not directly say you planned to do so in your post, I would 
not recommend running GFS as the underlying filesystem from which you share 
home directories via samba.  SMB is stateful and not cluster aware, and 
when I implemented this last year I ran into repeated samba crashes and 
locking issues I could not resolve.  Last I checked Redhat only supports 
sharing stateless NFS on top of GFS.  I haven't inquired about this since 
linuxworld (August), so maybe the new cluster system handles samba better, 
but I would definitly try it under a good load in a test environmnet before 
rolling it out via production.  If someone out there is succesfully sharing 
a GFS filesystem via samba (and doing something else on it at the same 
time, like webserving) I'd love to hear about it.
Running a samba PDC should be fine though I would still recommend putting 
the netlogon share on another filesystem and mirroring it wherever you had 
BDCs.

eventually we simply went with an iSCSI-based san and heartbeat-based 
failover of samba.  If you have an LDAP backend everything can be made 
redundant and you can eliminate SPOF without special kernels patches and 
the cluster requirements.  of course, this does not help in load balancing 
(since the data isn't available to both nodes simultaneously) and heartbeat 
itself has a few quirks.

As for the support from redhat I think you'll find that the people on this 
mailing list as about as top notch as exist in the linux community and as 
long as you've done your homework you should be able to get very good help 
very quickly.  but if you need someone to blame and someone else's ass to 
be on the line, well...
-alan


> Date: Mon, 12 Dec 2005 14:27:29 -0500
> From: D Canfield <canfield at uindy.edu>
> Subject: [Linux-cluster] General GFS Advice?
> To: linux-cluster at redhat.com
> Message-ID: <439DCF21.4090802 at uindy.edu>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> I'm just looking for a bit of general advice about GFS... We're
> basically just looking to use it as a SAN-based replacement for NFS.
> We've got a handful of servers that need constant read/write access to
> our users' home directories (Samba PDC/BDC, web server, network terminal
> servers, etc.), and we thought GFS might be a good replacement from a
> performance and security standpoint, let alone removing the SPOF of our
> main NFS/file server.  Another place we're thinking of using it is
> underneath our mail servers, so that as we grow, SMTP deliveries (and
> virus scanning) can happen on one machine while IMAP/POP connections can
> be served through another.
>
> Unfortunately, even at academic prices, Red Hat wants more per single
> GFS node than I'm paying for twenty AS licenses, so I've been heading
> down this road by building from the SRPMS.  I mostly have a 2-node test
> cluster built under RHEL4, but a number of things have me a little bit
> hesitant to move forward, so I'm wondering if some folks can offer some
> advice.
>
> For starters, is my intended use even appropriate for GFS?  It does seem
> as though I'm looking to put an awful lot of overhead (with the cluster
> management suite) onto these boxes just to eliminate a SPOF.
>
> Another concern is that this list seems to have a lot more questions
> posted than answers.  Are folks running into situations where
> filesystems are hopelessly corrupted or that they've been unable to
> recover from?  That's the impression I feel like I'm getting, but I
> suppose a newbie to Linux in general could get the same impression from
> reading the fedora lists out of context.    The last thing I want to do
> is put something into production and then have unexplained fencing
> occurences or filesystem errors.
>
> Finally, Red Hat sales is laying it on pretty heavy that the reason the
> GFS pricing is so high is because it's nearly impossible to install it
> yourself.  That was particularly true before GFS landed in Fedora.  Now
> the claim is just that it's very difficult to manage without a support
> contract.  Is this just marketing, or does GFS really turn out to be a
> nightmare to maintain?
>
> Any insights people could provide would be appreciated.
>
> Thanks
> DC
>
>