*sorry - twice today I have forgot to include the mailing list in the reply* Hiya, Using the 'gluster' command in gluster 3.1.x is the preferred method to set up volumes so there should be no need to alter config files manually (I am unsure of the state of custom translators and would not know if you need them for your situation). But just to recap... Distributed Volumes: ----------------------------- # gluster volume create test-volume transport tcp server1:/exp1 server2:/exp2 server3:/exp3 server4:/exp4 This method will use the storage allocated in every brick for use as a large file system. There is no failover - so if you reboot a server, the filesystem will be unavailable until the server is back online (if you loose the server and its bricks - some files will be unrecoverable). Files are written to the storage so as to attempt to make each brick use approximately the same space. (file1 -> server1 ... file2 -> server2 ... etc) Distributed Replicated Volumes: ----------------------------------------------- # gluster volume create test-volume replica 2 transport tcp server1:/exp1 server2:/exp2 The above example shows what could be considered similar to a raid mirror. Using 'replica 2' means that files will be written to 2 bricks and if there are only 2 servers, then each will contain all the files. If one server becomes unavailable, files can still be accessed (if you loose a server, all data is still available). The key point to note here is the 'replica 2' statement. You can possibly think of it as [(replica count - 1) = bricks that can go down before the file system is unavailable] In the case of 10 servers, the files will be distributed across the bricks with each file stored on a maximum of 2 bricks which gives you the ability to safely reboot a node (one at a time). Distributed Striped Volumes: ----------------------------------------- This is great for storing very large files as they can be striped and replicated across bricks - I will not go into detail on this, but the setup is similar to 'Distributed Replicated Volumes' above. Suggestion: ----------------- For 10 servers with varying amount of space available on each, I would make partitions for bricks to reside on all of the same size. This may involve a lot of work if you have: server1:/data(120GB) server2:/data(120GB) server3:/home/g1(2TB) server4:/home/g2(2TB) But for an example, if each server contained a brick of 120GB and you used 'replica 2' - you would then have a total of 10 bricks with a single file stored on 2 bricks - giving you approximately 1/5 of distributed replicated storage used for the replication. (Not sure if my figures are correct, but in essence - you would get more storage and a degree of failover) It is also recommended that the number of bricks should be a multiple of the replica count. It is possible to use multiple bricks on a single server, for example 20 bricks across 10 nodes using replica 4 - but then it should be noted that a single server containing 2 bricks would be safe to reboot - but rebooting a second server will take down an additional 2 bricks and could make the filesystem unavailable until one of the servers are back online depending on where the bricks have been configured in the replication. I hope that helps On 28 March 2011 22:37, s.varadha rajan <rajanvaradhu at gmail.com> wrote: > Hi Team, > > Anybody can help me for my below query ? > > Regards, > Varadharajan.S > > ---------- Forwarded message ---------- > From: s.varadha rajan <rajanvaradhu at gmail.com> > Date: Wed, Mar 23, 2011 at 6:27 PM > Subject: Req help for configuring glusterfs 3.1.2 > To: gluster-devel at nongnu.org > > > Hi, > > I would like to implement glusterfs 3.1 in my concern.We have around 10 > servers > and all are holding diff applications such as > webserver(apache,Tomcat),vmware,DNS server....all the servers are having > Diff > disk capacity such as 1 TB,2 TB like that.All the systems are having Ubuntu > 10.04 > > My Requirement: > > 1.Need to connect all the servers through glusterfs 3.1.x > 2.If i configured as Replication method, i am not getting Disk space.So i > need > to configure as like striping method.but glusterfs doesn't provide, fail > over > for this. > 3.for e.x if i take > server1:/data(120GB),server2:/data(120GB),server3:/home/g1=2TB,server4:/home/g2=2TB......i > want to connect all the servers.so that i can get big storage space for > all.if i > go for striping or distributed, if one server fails, i can't access the > volume > and get an error as "Transport end point not connected" > > 4.Suppose if i go for "unify" - Translator will solve my requirement ? but i > think no support for unify config under > ?3.1.2. version. > > Please let me know the solution and idea for config for this.i am searching > in > google for the past 10 days but no proper result. > > Regards, > Varadharajan.S > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users > >
