Hi Tim, i have already configure ZFS filesystem. everything work fine as expected.
I have only one isssue, if you can help. I want to transfer file from ZFS file system to UFS file system. How can i do that. i have read we cannot use ufsdump for this. The zfs send / receive could be used or is there any equivalent. Kind regards, Sailesh Mohabeer Systems Administrator Mauritius Network Services 2nd Floor C&R Court, Labourdonnais Street, Port Louis. Tel: (230) 211 5228 Fax: (230) 211 2414 ----- Original Message ----- From: "Tim Foster" <[email protected]> To: "Preethivirajsingh Sailesh Mohabeer" <sailesh.mohabeer at mns.intnet.mu> Cc: "opensolaris-help" <opensolaris-help at opensolaris.org> Sent: Tuesday, October 23, 2007 5:13 PM Subject: Re: [osol-help] ZFS RaidZ > Hi there, > > On Tue, 2007-10-23 at 03:51 -0700, Preethivirajsingh Sailesh Mohabeer > wrote: >> I have installed solaris 10 with ZFS as filesystem. I am using the >> Server as an application server. As you all know we cannot boot root >> filesystem from ZFS, i have mirror my root and swap file system. i >> have created two mirror metadevice on two disks. Furthermore, i have >> created file system using ZFS using two disks. I have used RAIDZ with >> the two disk. >> >> What i want to What will happen to my application if one disk fail in >> the RaidZ. > > Why are you using a 2 disk RAID-Z instead of a mirror ? > > But regardless, you should be able to still run with one failed disk in > your raidz set. > > More at: > http://docs.sun.com/app/docs/doc/819-5461/6n7ht6qrs?a=view#gamtu > http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide#RAID-Z_Configuration_Requirements_and_Recommendations > > Remember, you can always experiment around with pool configurations > using files, eg. here's me creating a small pool, writing a 20mb file, > checking it's contents, damaging one of the devices the redundant pool > is made of, scrubbing to detect errors, and then checking to see if the > file has survived: > > # mkfile 64m /tmp/file1 > # mkfile 64m /tmp/file2 > # zpool create raidz mypool /tmp/file1 /tmp/file2 > # digest -a md5 file > # cd /pool > # dd if=/dev/urandom of=file bs=1024k count=20 > # mkfile /tmp/file2 > # zpool scrub pool > # zpool status -v pool > # cd /pool > # digest -a md5 file > > cheers, > tim > > -- > Tim Foster, Sun Microsystems Inc, Solaris Engineering Ops > http://blogs.sun.com/timf
