[Linux-cluster] partitioning of filesystems in cluster nodes

[RR]

Right, it is indeed what I want to do. But now let me understand the basics
of GFS. GFS actually runs on the SAN but the GFS drivers/software that I
install on each of my cluster node just allows each of these nodes to see
these volumes? Something analogous to say iscsi-initiators on a node to via
the LUNs on an iSCSI SAN? If that's true, then is it possible for me to say
have my /opt/local installed on the GFS managed filesystem on the SAN such
that whatever application is installed once in this directory can be
accessed by all  nodes mounting that filesystem? So kind of get install
once, use everywhere kind of a deal?

Thanks so much
RR

-----Original Message-----
From: linux-cluster-bounces at redhat.com
[mailto:linux-cluster-bounces at redhat.com] On Behalf Of Greg Forte
Sent: Thursday, June 22, 2006 4:20 PM
To: linux-cluster at redhat.com
Subject: Re: [Linux-cluster] partitioning of filesystems in cluster nodes

No, I'm pretty sure what that's saying is that you can't build a GFS
filesystem _on top of_ a software RAID volume.  (Although you probably
could, using GNDB ...) but unless I misunderstood you, you're talking about
non-GFS filesystems on the local hard drives, with GFS on a SAN. 
This is a "standard" configuration, and certainly there's no reason you
can't use software raid on the disks that will host non-GFS volumes.

-g

RR wrote:
> Hi Greg,
> 
> Thanks for the input. What you say makes sense, but I was getting 
> these ideas from places like these: 
> http://kbase.redhat.com/faq/FAQ_78_3105.shtm
> 
> Have I totally misunderstood this? Maybe this is saying the same thing 
> as you are but I understand it wrong? I'm definitely not exporting any 
> of my local volumes or filesystems to anywhere, all I want to do is 
> have this particular directory tree which is based in /var/spool to 
> actually be on the SAN and have all nodes read/write to it at their 
> discretion. Nothing too complicated. But this FAQ seems to say that if 
> I'm running GFS on my nodes, then those nodes cannot be running 
> software RAID. Somewhere else I'd read I can't also have LVM volumes 
> on these disks. What I originally had planned was to have software 
> raid using md and LVM volumes on each machine to allow for RAID and
resizing on each node and then have GFS.
> 
> I'm all confused now! Please clarify someone?
> 
> Thanks, and sorry if this is stupid! :( RR
> 
> 
> 
> -----Original Message-----
> From: linux-cluster-bounces at redhat.com 
> [mailto:linux-cluster-bounces at redhat.com] On Behalf Of Greg Forte
> Sent: Thursday, June 22, 2006 3:47 PM
> To: linux clustering
> Subject: Re: [Linux-cluster] partitioning of filesystems in cluster 
> nodes
> 
> Where have you gotten these ideas?  LVM and software raid are not 
> mutually exclusive with GFS.  Your local disks can easily have LVM 
> volumes and use software (or hardware) RAID, no problem.  Unless 
> you're trying to export local disks as GFS volumes via GNDB (which I 
> can't speak to, have never done that), there shouldn't be any 
> restrictions on how you can configure the local disks.  If you want 
> certain portions of the local filesystem(s) to be located on the SAN
instead, it's as simple as a symlink (as Bob indicated).
> 
> -g
> 
> RR wrote:
>> Thanks Bob, that was my thought as well but just wasn't sure if that 
>> was the best way to implement partitioning. I suppose I'm still a bit 
>> not comfy with GFS considering I've never seen it in action but 
>> should do in a few days time, as I'm starting to set up the cluster 
>> suite and GFS
> now.
>> I'm assuming I still cannot use Logical Volumes on the cluster nodes 
>> while the nodes are being used within a cluster in conjunction with 
>> GFS? How do people provide for RAID1 type redundancy against system 
>> boot-disk failure for each of their cluster nodes if I can't setup 
>> software raid for these nodes? Do I run a cron job to sync up the two 
>> drives within the system so in case of a drive failure, I can at 
>> least boot the system using the secondary drive?
>>
>> Thanks,
>> RR
>>
>> -----Original Message-----
>> From: linux-cluster-bounces at redhat.com 
>> [mailto:linux-cluster-bounces at redhat.com] On Behalf Of Robert 
>> Peterson
>> Sent: Thursday, June 22, 2006 4:46 AM
>> To: linux clustering
>> Subject: Re: [Linux-cluster] partitioning of filesystems in cluster 
>> nodes
>>
>> Hi RR,
>>
>> For your the local root partitions on the individual nodes, it's 
>> probably best to use ext3.
>> On the SAN, use GFS and Red Hat Cluster Suite.  Then perhaps you can 
>> create a symlink from your local node's mount point to the SAN, e.g.
>> from /mnt/gfs_san/var/spool to its local /var/spool.
>>
>> Regards,
>>
>> Bob Peterson
>> Red Hat Cluster Suite
>>
>>
>> --
>> Linux-cluster mailing list
>> Linux-cluster at redhat.com
>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>
> 
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
> 
> 

--
Linux-cluster mailing list
Linux-cluster at redhat.com
https://www.redhat.com/mailman/listinfo/linux-cluster