toasters May 2007

toasters@lists.teaparty.net

64 participants
41 discussions

RE: SnapDrive install with existing iSCSI LUNS
by Julio Calderon 24 May '07

24 May '07

If you want to test your procedure before you actually do it!! I would create a dummy lun and mount the same way dump some data in it, make sure you format the same type of file system.... Remove and try sd.... That wil raise your comfort level... My personal best practice is to try it, back it up, test restore and then do your live work.... Hope that helps Have fun Julio calderon -----Original Message----- From: Toasters [mailto:toasters@aegonusa.com] Sent: Thu May 24 12:59:08 2007 To: Nils Vogels; toasters(a)mathworks.com Subject: RE: SnapDrive install with existing iSCSI LUNS Just unmap them from the filer , install SD and then remap them with it. -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Nils Vogels Sent: Thursday, May 24, 2007 1:53 PM To: toasters(a)mathworks.com Subject: Re: SnapDrive install with existing iSCSI LUNS Hi Sean, On 5/24/07, Sean McWherter <sean(a)marinedepot.com> wrote: > > > What is the proper way to install snapdrive on a machine that already > has LUNs mapped with the iSCSI initiator? Should I disconnect the LUN > first? I did read one post on the NOW forums where a guy got hosed by > disconnecting the LUN and then installing SnapDrive after someone told > him that's what to do. He was never able to mount them back. I'm > trying to dot all my I's and cross all my T's before I dive in here. I'd say disconnecting the LUN, removing all traces in the iSCSI initiator, then installing SnapDrive, and then using SnapDrive to 'connect' to the LUN is the right way. If I were you, I'd test this using an ONTAP simulator, though ;) Grtz, Nils -- Simple guidelines to happiness: Work like you don't need the money, Love like your heart has never been broken and Dance like no one can see you.

1 0

SnapDrive install with existing iSCSI LUNS
by Sean McWherter 24 May '07

24 May '07

I need guidance in setting up SnapDrive on our Windows 2003 Enterprise x64 servers. We have two of these servers clustered hosting our SQL 2000 servers. My question is this: What is the proper way to install snapdrive on a machine that already has LUNs mapped with the iSCSI initiator? Should I disconnect the LUN first? I did read one post on the NOW forums where a guy got hosed by disconnecting the LUN and then installing SnapDrive after someone told him that's what to do. He was never able to mount them back. I'm trying to dot all my I's and cross all my T's before I dive in here.

3 2

rysnc and filers
by tmac 24 May '07

24 May '07

Okay, I am trying to copy data from a DOT-7 to a DOT-GX. I start off with: mount filer7:/vol/src /src mount filerg:/ /gx I then do (from RHEL4U2): cp -a /src/dir1 /gx I then do: rsyn -avH --delete /src/dir1/ /gx/dir1/ There is nothing changing on the source, yet it transfers files anyway. Anyone have a clue as to why? The destination get some files marked with current time, yet others are marked with an older time. -- --tmac RedHat Certified Engineer #804006984323821 Principal Consultant, RABA Technologies 240-373-3926 (office) 301-688-4705 (Lab) 214-279-3926 (eFAX)

3 2

Re: rysnc and filers
by Jerry 24 May '07

24 May '07

Or rsync --whole-file the first time, or ndmpcopy. ----- Original Message ---- From: Nils Vogels <bacardicoke(a)gmail.com> To: tmac <tmacmd(a)gmail.com> Cc: toasters(a)mathworks.com Sent: Wednesday, May 23, 2007 3:59:21 PM Subject: Re: rysnc and filers Hi tmac, On 5/23/07, tmac <tmacmd(a)gmail.com> wrote: > Okay, I am trying to copy data from a DOT-7 to a DOT-GX. > ... > > I then do (from RHEL4U2): > > cp -a /src/dir1 /gx > > I then do: > > rsyn -avH --delete /src/dir1/ /gx/dir1/ > > There is nothing changing on the source, yet it transfers files anyway. > > Anyone have a clue as to why? cp doesnt take all metadata. Try tar/cpio! # cd /source # tar cf - * | ( cd /target ; tar xf - ) -- Simple guidelines to happiness: Work like you don't need the money, Love like your heart has never been broken and Dance like no one can see you. ____________________________________________________________________________________Need a vacation? Get great deals to amazing places on Yahoo! Travel. http://travel.yahoo.com/

2 1

A-SIS (Netapp DeDupe)
by Glenn Dekhayser 23 May '07

23 May '07

Doing some reading on the A-SIS feature. Will this only De-Dupe WITHIN a flexvol, or leverage 4kb block commonalities within an aggregate? I would think the savings would be dramatic across flexvols, especially in Boot-SAN environments. Glenn (the other one)

2 1

running out of inodes problem
by Magnus Swenson 22 May '07

22 May '07

Hello Toasters, Just wanted to do a quick check, what the standard practise is when running out of inodes on a volume. I have several flex volumes in one aggregate. One of the volumes only at 80% full ran out of inodes. df -i will show number of inodes used and inodes free. This is a 100G volume with 3458831 inodes. According to now.netapp.com, there are two solutions, increase inodes with the 'maxfiles' command, or add more disk space to the volume. Has anybody had experience with this and which way did you go?

7 9

lun type: windows_gpt
by Buerger, Andreas 22 May '07

22 May '07

Hi, I just created a lun via lun setup on our 6070 running 7.2.2. I mentioned a lun type called windows_gpt, does anybody know what this type is used for? Is it a new feature in 7.2 ? Regards ----------------------------------------- Andreas Buerger Wincor Nixdorf Int. GmbH WN CIO4 Computing & Storage Heinz-Nixdorf-Ring 1 33106 Paderborn Phone: +49 5251 693 4215 Fax: +49 5251 693 4200 Mail: andreas.buerger(a)wincor-nixdorf.com -----------------------------------------

3 2

ndmpcopy for Linux
by Augie Schwer 22 May '07

22 May '07

Has anyone built ndmpcopy for Linux recently? The Solaris version doesn't seem to build anymore and the ndmp.org web site does not look very active anymore. -- Augie Schwer - Augie(a)Schwer.us - http://schwer.us Key fingerprint = 9815 AE19 AFD1 1FE7 5DEE 2AC3 CB99 2784 27B0 C072

2 1

Re: Clustered heads with SATA disk shelves?
by Wilkinson, Brent 21 May '07

21 May '07

Not sure but I believe you do not want to mix fc and sata loops. Could be wrong though. -----Original Message----- From: owner-toasters(a)mathworks.com <owner-toasters(a)mathworks.com> To: Tom Samplonius <tom(a)samplonius.org> CC: toasters <toasters(a)mathworks.com> Sent: Fri May 18 19:44:32 2007 Subject: Re: Clustered heads with SATA disk shelves? clustered sata shelves work just fine. I have several terabytes attached to a 3050 cluster with out any problems. AT-FCX is required for that to work. I'd also recommend being on the latest firmware on the AT-FCX. -Blake On 5/18/07, Tom Samplonius <tom(a)samplonius.org> wrote: > > One item that I've researching that is a bit unclear, is that, can I can connect a SATA disk shelf (DS14mk2 AT) to a clustered set of FAS960 heads? > > I think there is no problem. I assume that you must use the AT-FCx modules, as they are dual-looped. I suspect that the the AT-FCx is mandatory for clustering (as opposed to the AT-FC) , but NOW seems a bit vague, at least according to what I was able to find. > > Also, I've heard rumors that connecting SATA shelves to a cluster is not really reliable. I was a bit taken a back, as NetApp as the SATA shelves have been out for a while. Is this really the case? > > > Tom >

3 2

Symantec AV not releasing pBlks
by Horton, Jamie 18 May '07

18 May '07

Anyone ever had the below issue?: I'm running 2 Win2k3 servers as my Symantec Scan Engine Servers (running Symantec Anti-Virus for Network Attached Storage 4.3). Both act as primary vscan scanners (for my R200 and FAS960 filers). Every once in awhile one or both of my filers stop servicing CIFS requests, because they run out of pBlks (control blocks). After several outages this week, I have learned that in order for everything to run clean, I have to do the following in sequence: 1. vscan off (on both filers) 2. Stop scan engine service on both win2k3 servers 3. vscan on (on both filers) 4. Start scan engine service on both win2k3 servers If I have everything on, and I stop and start vscan only on the filers (without recycling services on the win2k3 servers), I immediately start losing pBlks until they are all exhausted, then nobody can access any shares. NetApp suggested I upgrade to 7.2.2 since this version deals with pBlks better. I upgraded last night, and since I didn't recycle scan engine services on the win2k3 servers, I ran out of pBlks early this morning. Once I executed the sequence above, all was well. NOTE: we are continuing to add more users to our environment (user home drives are on the FAS960), which obviously add more to the scanning load, but that doesn't account for it happening on my R200 (software shares and application shares). Any information would be greatly appreciated. Thanks, -Jamie

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters May 2007