[Gluster-users] R: SPOF question

[Count Zero]

Hi Roberto,

Try to reverse mirror-0 and mirror-1 such that mirror-1 is the first in the list.
Does this make a difference?


On May 26, 2010, at 2:21 AM, Roberto Lucignani wrote:

> Hi Bala,
> thank you very much for your reply. Yes I tried both methods, one building a
> vol file and one retrieving the configuration from the server and both work
> fine
> 
> However my problem do not regards the mount process but it regards the spof
> represented by node01. If it is unavailable I can't access my volume.
> 
> Why this ? and why it happens only with the first node ?
> 
> Regards
> M4dG
> 
> -----Messaggio originale-----
> Da: Bala.JA [mailto:bala at gluster.com] 
> Inviato: domenica 23 maggio 2010 20.58
> A: roberto.lucignani at caleidos.it
> Cc: gluster-users at gluster.org
> Oggetto: Re: [Gluster-users] SPOF question
> 
> 
> Hi Roberto,
> 
> Gluster Storage Platform provides client volume spec file through server for
> 
> created volumes.
> 
> You can mount using
> 
> mount -t glusterfs <server>:<volume>-<transport> <your-mount-point>
> 
> for example,
> mount -t glusterfs node01:gluster01-tcp /mnt/gluster01
> 
> Its not required to write your own spec file for mounting it.
> 
> Thanks,
> 
> Regards,
> Bala
> 
> 
> 
> Roberto Lucignani wrote:
>> Hi all,
>> 
>> I installed two Gluster Storage Platorm 3.0.4 on two servers node01 e
>> node02.
>> 
>> I created a volume called gluster01 than I mounted it on a Debian box in
>> this way:
>> 
>> 
>> 
>> mount -t glusterfs /etc/glusterfs/gluster01-tcp.vol /mnt/gluster01/
>> 
>> 
>> 
>> the gluster01-tcp.vol is the following:
>> 
>> 
>> 
>> 
>> 
>> volume 192.168.0.200-1
>> 
>>    type protocol/client
>> 
>>    option transport-type tcp
>> 
>>    option remote-host 192.168.0.200
>> 
>>    option transport.socket.nodelay on
>> 
>>    option transport.remote-port 10012
>> 
>>    option remote-subvolume brick1
>> 
>> end-volume
>> 
>> 
>> 
>> volume 192.168.0.200-2
>> 
>>    type protocol/client
>> 
>>    option transport-type tcp
>> 
>>    option remote-host 192.168.0.200
>> 
>>    option transport.socket.nodelay on
>> 
>>    option transport.remote-port 10012
>> 
>>    option remote-subvolume brick2
>> 
>> end-volume
>> 
>> 
>> 
>> volume 192.168.0.201-1
>> 
>>    type protocol/client
>> 
>>    option transport-type tcp
>> 
>>    option remote-host 192.168.0.201
>> 
>>    option transport.socket.nodelay on
>> 
>>    option transport.remote-port 10012
>> 
>>    option remote-subvolume brick1
>> 
>> end-volume
>> 
>> 
>> 
>> volume 192.168.0.201-2
>> 
>>    type protocol/client
>> 
>>    option transport-type tcp
>> 
>>    option remote-host 192.168.0.201
>> 
>>    option transport.socket.nodelay on
>> 
>>    option transport.remote-port 10012
>> 
>>    option remote-subvolume brick2
>> 
>> end-volume
>> 
>> 
>> 
>> volume mirror-0
>> 
>>    type cluster/replicate
>> 
>>    subvolumes 192.168.0.201-1 192.168.0.200-1
>> 
>> end-volume
>> 
>> 
>> 
>> volume mirror-1
>> 
>>    type cluster/replicate
>> 
>>    subvolumes 192.168.0.201-2 192.168.0.200-2
>> 
>> end-volume
>> 
>> 
>> 
>> volume distribute
>> 
>>    type cluster/distribute
>> 
>>    subvolumes mirror-0 mirror-1
>> 
>> end-volume
>> 
>> 
>> 
>> volume readahead
>> 
>>    type performance/read-ahead
>> 
>>    option page-count 4
>> 
>>    subvolumes distribute
>> 
>> end-volume
>> 
>> 
>> 
>> volume iocache
>> 
>>    type performance/io-cache
>> 
>>    option cache-size `echo $(( $(grep 'MemTotal' /proc/meminfo | sed
>> 's/[^0-9]//g') / 5120 ))`MB
>> 
>>    option cache-timeout 1
>> 
>>    subvolumes readahead
>> 
>> end-volume
>> 
>> 
>> 
>> volume quickread
>> 
>>    type performance/quick-read
>> 
>>    option cache-timeout 1
>> 
>>    option max-file-size 64kB
>> 
>>    subvolumes iocache
>> 
>> end-volume
>> 
>> 
>> 
>> volume writebehind
>> 
>>    type performance/write-behind
>> 
>>    option cache-size 4MB
>> 
>>    subvolumes quickread
>> 
>> end-volume
>> 
>> 
>> 
>> volume statprefetch
>> 
>>    type performance/stat-prefetch
>> 
>>    subvolumes writebehind
>> 
>> end-volume
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> all works fine and smooth, I can write and read on that volume without any
>> problem.
>> 
>> The problem is when the node01 is unavailable, I can't access the volume
> via
>> mount on the Debian box. This doesn't happen if it is the node02 to be
>> unavailable.
>> 
>> I expected the same behavior in the two cases, in this way the node01
>> represents an SPOF, am I wrong ? am I missing something the configuration
> ?
>> 
>> 
>> 
>> Tnx in advance
>> 
>> Rpberto
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------
>> 
>> _______________________________________________
>> Gluster-users mailing list
>> Gluster-users at gluster.org
>> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
> 
> 
> 
> 
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users