<br><br><div class="gmail_quote">On Mon, Apr 20, 2009 at 12:56 PM, ESGLinux <span dir="ltr"><<a href="mailto:esggrupos@gmail.com">esggrupos@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Hello, <br><br><br><br><div class="gmail_quote">2009/4/20 Gordan Bobic <span dir="ltr"><<a href="mailto:gordan@bobich.net" target="_blank">gordan@bobich.net</a>></span><div class="im"><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div>On Mon, 20 Apr 2009 18:18:22 +0200, ESGLinux <<a href="mailto:esggrupos@gmail.com" target="_blank">esggrupos@gmail.com</a>> wrote:<br> > Hello,<br> ><br> > first, thanks for your answer,<br> ><br> > I suspected it but why can i do it with NFS. ?<br> <br> </div>Not sure I understand your question. NFS is a network file system like CIFS<br> specifically designed to be mounted from multiple clients simultaneously.<br> ext3 is only designed with a single accessor in mind.</blockquote></div><div><br>I´ll try to explain myself<br><br>I have a partition /dev/sda<br><br>/dev/sda on /iscsivol type ext3 (rw)<br><br>but this partition is a target iscsi on another server. I format the partition with ext3 but its not a local disk, is a target iscsi.<br> <br>with this configuration I have the filesystem corrupted.<br><br>second scenario<br><br>I have <br>192.168.1.198:/nfsexport/ 6983168 2839168 3783552 43% /mnt<br><br>but the parttion 192.168.1.198:/nfsexport/ is again ext3 the diference is that I use nfs as network protocol instead of iscsi.</div> </div></blockquote><div><br>(overly simplified explanation)<br>This is because with iSCSI you are mounting the filesystem directly as a block device. Through NFS, the filesystem isn't accessed directly. The NFS server handles all the actual I/O, so only one system is actually touching the filesystem. <br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div class="gmail_quote"><div class="im"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div><br> > the nodes never are going to be active at the same time so I can mount<br> the<br> > shares via NFS. With NFS when I create a file in a share automatically i<br> > got it in the shared mounted by all the clients.<br> <br> </div>I still don't understand your question - that is what NFS is designed for.</blockquote></div><div><br>Yes I agree with you, but I thought with iscsi i can do the same as with NFS. </div></div></blockquote><div> <br>No, iSCSI and NFS are completely different. One is SCSI encapsulated in Ethernet- the volumes are going to look local to the system. The other is a conventional file sharing protocol.<br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div class="gmail_quote"><div class="im"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div><br> > In this case I don´t need to write to the share concurrently<br> ><br> > can be this configuration a problem?<br> <br> </div>No, it's fundamentally impossible. In order to have a FS that can be<br> mounted simultaneously from multiple nodes, it has to be aware of multiple<br> nodes accessing it, which means that it needs coherent caching. Local file<br> systems like ext3 don't have this. When one node writes to the ext3 file<br> system, the other node will have cached the inodes as they were originally,<br> and it won't bother hitting the disk to re-read the contents, it'll just<br> return what it already has cached. And almost certainly corrupt the file<br> system in the process.<br> <br> You cannot have a shared ext3 volume with writing enabled. Period.</blockquote></div><div><br>ok understand it, <br><br>but (always there is a but ...) <br><br>I only want to share a directory in which one node writes at one and when it fails the other node has the diretory mounted with the data and can write to it. <br> <br>Before I have known about cluster my decission would been to mount the shares with NFS. Now I want to be more sofisticated and want to use cluster tools, so I thought to mount it with iSCSI instead of NFS, but always with the ext3 as the underlying filesystem. <br> <br>Perphaps this is my mistake. <br><br>any suggestion that makes me see the light ;.)<br></div></div></blockquote><div><br>If you want to have multiple systems access the same volume via iSCSI, you need to use a filesystem that is multiple-node aware such as GFS.<br> </div></div>