From: Neil Brown Subject: Re: async vs. sync Date: Thu, 25 Nov 2004 09:09:29 +1100 Message-ID: <16805.1689.739687.36138@cse.unsw.edu.au> References: <482A3FA0050D21419C269D13989C611307CF4B56@lavender-fe.eng.netapp.com> <41A3AFC4.6080404@int-evry.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "Lever, Charles" , nfs@lists.sourceforge.net Return-path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list2.sourceforge.net with esmtp (Exim 4.30) id 1CX5KY-0004Zv-86 for nfs@lists.sourceforge.net; Wed, 24 Nov 2004 14:09:42 -0800 Received: from note.orchestra.cse.unsw.edu.au ([129.94.242.24] ident=root) by sc8-sf-mx1.sourceforge.net with esmtp (Exim 4.41) id 1CX5KW-00048s-Vs for nfs@lists.sourceforge.net; Wed, 24 Nov 2004 14:09:42 -0800 To: jehan procaccia In-Reply-To: message from jehan procaccia on Tuesday November 23 Sender: nfs-admin@lists.sourceforge.net Errors-To: nfs-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: Discussion of NFS under Linux development, interoperability, and testing. List-Post: List-Help: List-Subscribe: , List-Archive: On Tuesday November 23, jehan.procaccia@int-evry.fr wrote: > > Perhaps it's not the place to talk about ext3 here, but if someone on > the list did already put their journal on a separate device, please > confirm me those points: > From what I read on man mkefs for ext3 FS I can create a journal on a > separate FS : > mke2fs -O journal_dev external-journal > creates the journal FS, on which device ? -> internal scsi drive of my > server or better placed on the dell/EMC SP ? Any device that is reliable and will respond very quickly to sequential writes is fine. A dedicated mirrored pair of fast SCSI drives is ideal if you can afford it. If the "dell/EMC SP" can be configured to provide a mirrored pair of other-wise lightly used drives, then a partition on that would be fine. If you put the journal on a single drive, then a failure of that drive will cost you some data loss, and a full 'fsck' of the main storage. I put my journals on internal drives. One cost of this is if the server dies (e.g. bad power supply) then you need to move the internal drives to some other computer before you can get your data back. This can be trivial, or frustratingly difficult(*). > > mke2fs -J device=/dev/external-journal /dev/emcpower > Format the FS and use the external journal just create above, but what is the recommended size of the > external journal ? when journal is internal it is said the size of the journal must be at least 1024 filesystem blocks > (in my case blocks a 4K size) so journal is at least 4 Mb, but > should it be bigger ? I think 1gigabyte is a good size - probably much bigger than needed, but not too big. I wouldn't go below 128Meg myself. I once had a 6gig journal (because there was plenty of space, so "why not"). I found out why not: On restart ext3 has to scan the whole journal. For a 6gig journal that takes a while. Not as long as fsck, but still too long. > > Finally, can I "externalize" an already internal journal from > production FS (convert journal from inside to outside without > reformating the FS ) ? Yes, but not while the filesystem is mounted. Use tune2fs to remove the internal journal: tune2fs -O ^has_journal create an external journal mke2fs -O journal_dev /dev/external-journal and attach the journal to the fs tune2fs -J device=/dev/external-journal then re-mount. NeilBrown (*) I recently needed to move a large filesystem that was on an external drive box from a failing machine to another machine. Both had hot-swap scsi drives for the internal system discs though they were from different manufacturers. The drive caddies were different but unscrewing a drive from one and putting it in the other didn't seem like too much trouble .... until we tried it. The caddy from one manufacturer was designed so there was no tolerance. A drive 0.5mm larger than the one they provided simply didn't fit. The drive I wanted to put in was 0.5mm larger :-( Fortunately, as it was a controlled change over, we could detach the journal on the system discs of one machine and re-attach a new journal on the other. If the first computer had blown up instead of started failing occasionally, it would have been *much* more awkward. ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs