Re[4]: [zfs-discuss] Replacing a drive in a raidz2 group

2007-01-15 Thread Robert Milkowski
Hello Jason, Sunday, January 14, 2007, 1:26:37 AM, you wrote: JJWW Hi Robert, JJWW Will build 54 offline the drive? IIRC there hasn't been ZFS+FMA integration yet. -- Best regards, Robertmailto:[EMAIL PROTECTED]

[zfs-discuss] Eliminating double path with ZFS's volume manager

2007-01-15 Thread Philip Mötteli
Hi, Since they have installed a second path to our Hitachi SAN, my Mac OS X Server 4.8 mounts every SAN disk twice. I asked everywhere, if there's a way, to correct that. And the only answer so far was, that I need a volume manager, that can be configured to consider two volumes as being

Re: [zfs-discuss] Question: ZFS + Block level SHA256 ~= almost free CAS Squishing?

2007-01-15 Thread Darren J Moffat
Pawel Jakub Dawidek wrote: On Mon, Jan 08, 2007 at 11:00:36AM -0600, [EMAIL PROTECTED] wrote: I have been looking at zfs source trying to get up to speed on the internals. One thing that interests me about the fs is what appears to be a low hanging fruit for block squishing CAS (Content

Re: [zfs-discuss] Eliminating double path with ZFS's volume manager

2007-01-15 Thread Robert Milkowski
Hello Philip, Monday, January 15, 2007, 10:44:49 AM, you wrote: PM Hi, PM Since they have installed a second path to our Hitachi SAN, my PM Mac OS X Server 4.8 mounts every SAN disk twice. PM I asked everywhere, if there's a way, to correct that. And the PM only answer so far was, that I need

[zfs-discuss] iSCSI on a single interface?

2007-01-15 Thread Dick Davies
Hi, are there currently any plans to make an iSCSI target created by setting shareiscsi=on on a zvol bindable to a single interface (setting tpgt or acls)? I can cobble something together with ipfilter, but that doesn't give me enough granularity to say something like: 'host a can see target 1,

[zfs-discuss] Mounting a ZFS clone

2007-01-15 Thread Albert Chin
I have no hands-on experience with ZFS but have a question. If the file server running ZFS exports the ZFS file system via NFS to clients, based on previous messages on this list, it is not possible for an NFS client to mount this NFS-exported ZFS file system on multiple directories on the NFS

Re: [zfs-discuss] Re: ZFS direct IO

2007-01-15 Thread Roch - PAE
Jonathan Edwards writes: On Jan 5, 2007, at 11:10, Anton B. Rang wrote: DIRECT IO is a set of performance optimisations to circumvent shortcomings of a given filesystem. Direct I/O as generally understood (i.e. not UFS-specific) is an optimization which allows data to

Re: [zfs-discuss] Re: Eliminating double path with ZFS's volume manager

2007-01-15 Thread Dominic Kay
Go poke around in the multipath Xsan storage pool properties. Specifies how Xsan uses multiple Fibre Channel paths between clients and storage. This is the equiv of Veritas DMP or [whatever we now call] Solaris MPxIO /d 2007/1/15, Philip Mötteli [EMAIL PROTECTED]: Hi, Monday, January 15,

Re: [zfs-discuss] question about self healing

2007-01-15 Thread Kyle McDonald
Richard Elling wrote: roland wrote: i have come across an interesting article at : http://www.anandtech.com/IT/showdoc.aspx?i=2859p=5 Can anyone comment on the claims or conclusions of the article itself? It seems to me that they are not always clear about what they are talking about.

Re: [zfs-discuss] Eliminating double path with ZFS's volume manager

2007-01-15 Thread Torrey McMahon
Robert Milkowski wrote: 2. I belive it's definitely possible to just correct your config under Mac OS without any need to use other fs or volume manager, however going to zfs could be a good idea anyway That implies that MacOS has some sort of native SCSI multipathing like Solaris Mpxio.

Re: [zfs-discuss] iSCSI on a single interface?

2007-01-15 Thread Rick McNeal
On Jan 15, 2007, at 8:34 AM, Dick Davies wrote: Hi, are there currently any plans to make an iSCSI target created by setting shareiscsi=on on a zvol bindable to a single interface (setting tpgt or acls)? I can cobble something together with ipfilter, but that doesn't give me enough

Re: [zfs-discuss] Eliminating double path with ZFS's volume manager

2007-01-15 Thread Jason J. W. Williams
Hi Torrey, I think it does if you buy Xsan. Its still a separate product isn't it? Thought its more like QFS + MPXIO. Best Regards, Jason On 1/15/07, Torrey McMahon [EMAIL PROTECTED] wrote: Robert Milkowski wrote: 2. I belive it's definitely possible to just correct your config under Mac

Re: [zfs-discuss] Re: ZFS direct IO

2007-01-15 Thread Jason J. W. Williams
Hi Roch, You mentioned improved ZFS performance in the latest Nevada build (60 right now?)...I was curious if one would notice much of a performance improvement between 54 and 60? Also, does anyone think the zfs_arc_max tunable-support will be made available as a patch to S10U3, or would that

Re: [zfs-discuss] Eliminating double path with ZFS's volume manager

2007-01-15 Thread Torrey McMahon
Got me. However, transport multipathing - Like Mpxio, DLM, VxDMP, etc. - is usually separated from the filesystem layers. Jason J. W. Williams wrote: Hi Torrey, I think it does if you buy Xsan. Its still a separate product isn't it? Thought its more like QFS + MPXIO. Best Regards, Jason On

Re: [zfs-discuss] Mounting a ZFS clone

2007-01-15 Thread Robert Milkowski
Hello Albert, Monday, January 15, 2007, 5:55:23 PM, you wrote: AC I have no hands-on experience with ZFS but have a question. If the AC file server running ZFS exports the ZFS file system via NFS to AC clients, based on previous messages on this list, it is not possible AC for an NFS client to

Re: [zfs-discuss] Eliminating double path with ZFS's volume manager

2007-01-15 Thread Jason J. W. Williams
Hi Torrey, Looks like its got a half-way decent multipath design: http://docs.info.apple.com/article.html?path=Xsan/1.1/en/c3xs12.html Whether or not it works is another story I suppose. ;-) Best Regards, Jason On 1/15/07, Torrey McMahon [EMAIL PROTECTED] wrote: Got me. However, transport

Re: [zfs-discuss] Re: RE: On the SATA framework

2007-01-15 Thread Frank Cusack
On January 15, 2007 11:58:10 AM -0800 Andrew Pattison [EMAIL PROTECTED] wrote: The SATA frame work has laready been integrated and is available on Solaris 10 Update 3 and Nevada. update 2 as well, yes? I thought U2 is when the first SATA support was announced, for the Marvell controller.

Re: [zfs-discuss] question about self healing

2007-01-15 Thread Richard Elling
Kyle McDonald wrote: Richard Elling wrote: roland wrote: i have come across an interesting article at : http://www.anandtech.com/IT/showdoc.aspx?i=2859p=5 Can anyone comment on the claims or conclusions of the article itself? It seems to me that they are not always clear about what they are

Re: [zfs-discuss] Question: ZFS + Block level SHA256 ~= almost free CAS Squishing?

2007-01-15 Thread Mike Gerdts
On 1/10/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Dick Davies [EMAIL PROTECTED] wrote on 01/10/2007 05:26:45 AM: On 08/01/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I think that in addition to lzjb compression, squishing blocks that contain the same data would buy a lot of space

[zfs-discuss] Re: Re: Eliminating double path with ZFS's volumemanager

2007-01-15 Thread Philip Mötteli
Go poke around in the multipath Xsan storage pool properties. Specifies how Xsan uses multiple Fibre Channel paths between clients and storage. This is the equiv of Veritas DMP or [whatever we now call] Solaris MPxIO You mean, I should find some configuration file? Well, I can't find one.

[zfs-discuss] Re: Eliminating double path with ZFS's volume manager

2007-01-15 Thread Philip Mötteli
Robert Milkowski wrote: 2. I belive it's definitely possible to just correct your config under Mac OS without any need to use other fs or volume manager, however going to zfs could be a good idea anyway That implies that MacOS has some sort of native SCSI multipathing like

[zfs-discuss] Re: Eliminating double path with ZFS's volume manager

2007-01-15 Thread Philip Mötteli
Looks like its got a half-way decent multipath design: http://docs.info.apple.com/article.html?path=Xsan/1.1/ en/c3xs12.html Great, but that is with Xsan. If I don't exchange our Hitachi with an Xsan, I don't have this 'cvadmin'. This message posted from opensolaris.org