toasters February 2006

toasters@lists.teaparty.net

129 participants
137 discussions

Re: wiping disks on an old F230
by Davin Milun 18 Feb '06

18 Feb '06

zero'ing spares on a NetApp filer really does write zeros to all blocks. Davin. At 22:14, on Feb 17, 2006, "Blake Golliher" wrote: > Or maybe, it just unallocated those blocks, and says "that data there? > just ignore it". Which to me is more likely. > > > -Blake > > On 2/17/06, Adrian Ulrich <toaster(a)blinkenlights.ch> wrote: > > > > > I don't belive spares_zero acutally overwrites data, > > > > It does. > > > > Proof: > > Adding a 'zeroed-out' disk to an aggr/trad_volume takes just a few > > seconds. > > > > Why? -> Because the filer *knows* that the disk is filled with \0's so > > the parity-disk does not have to care about the new disk.

1 0

RE: wiping disks on an old F230
by Mats.Oberg＠tietoenator.com 18 Feb '06

18 Feb '06

Well, if I remember correctly "rc_toggle_basic" was the thing to use before "priv set" existed.... /Mats -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Sphar, Mike Sent: den 15 februari 2006 23:42 To: toasters(a)mathworks.com Subject: wiping disks on an old F230 We've got an old F230 that I was finally able to decommission today, and I figure even though I've deleted all the data it would probably be a good idea to scrub the disks a bit for security purposes. Problem is I just can't remember (and can't seem to find searching around NOW) what the best way to do that back in OnTap 5.3.7R3D18 was. I know I could do disk sanitize now, but not sure what if anything was available in older versions. Any suggestions? (I fully concede that I might be having a complete brain failure about the painfully obvious solution I should already know. But I even searched around NOW a bit and couldn't find an answer.) -- Michael W. Sphar - IS&T - Lead Systems Administrator SMBU Engineering Support Services, BMC Software

5 5

RE: wiping disks on an old F230
by Glenn Walker 18 Feb '06

18 Feb '06

Unallocating the blocks would be just an update to the spacemap metadata and would be very quick. The zero process takes a considerable amount of time and constitutes lots of disk reads\writes (LEDs are very busy). From what I can tell, it's an outright 'zeroing' process where every block is overwritten... albeit not to the level that disk sanitize does (8 overwrites I believe??) -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Blake Golliher Sent: Saturday, February 18, 2006 1:15 AM To: Adrian Ulrich Cc: toasters(a)mathworks.com Subject: Re: wiping disks on an old F230 Or maybe, it just unallocated those blocks, and says "that data there? just ignore it". Which to me is more likely. -Blake On 2/17/06, Adrian Ulrich <toaster(a)blinkenlights.ch> wrote: > > > I don't belive spares_zero acutally overwrites data, > > It does. > > Proof: > Adding a 'zeroed-out' disk to an aggr/trad_volume takes just a few > seconds. > > Why? -> Because the filer *knows* that the disk is filled with \0's so > the parity-disk does not have to care about the new disk. > >

1 0

RE: How to mount a snapmirror lun w/o snapdrive
by Glenn Walker 18 Feb '06

18 Feb '06

I've not used this functionality without snapdrive, but there is always a cmd way :-) SnapDrive just automates things. In this case, it checks the metadata of the volume on the SM destination and senses that it is indeed a SnapMirror'ed volume. Then it breaks the SM relationship to make it writeable and subsequently restores the LUN in question that you want to connect to. Be warned - the SFSR functionality can take some significant time... however, SnapDrive 4.0\ONTAP 7.1 significantly reduces this operation by performing the block compare\filesystem writes behind the scenes while the LUN is 'restored' via LUN clone\LUN split (instead of the old SingleFileSnapRestore). In your case, however... you'd have to trust that the LUN you wish to connect to is consistent - snapdrive does do by using the last consistent snapshot that it took (typically the '@snapmir@' one, if I recall). The difference between lun.rws and lun.lun is this: lun.lun is indeed your LUN. Lun.rws is the 'temp space' allocated\associated with a read\write LUN based off of a snapshot... there is no such thing (*ahem* yet) as a read-only LUN in ONTAP. The .rws is just a placeholder for any changes that you may make to the snapshot-version of the LUN you mount to. Be warned: creating a .rws file (done automatically via SnapDrive when you connect to the LUN in a snapshot) will LOCK the snapshot you created it from!! This is done to ensure that the mounted\active .rws LUN can point to the data blocks needed to represent its contents (if you deleted the snapshot, then you'd point to null data - not good). Any subsequent snapshot taken _after_ mounting the .rws file will also lock the previous snapshot. Be careful here! Glenn ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Steven Mandrake Sent: Friday, February 17, 2006 4:14 PM To: toasters(a)mathworks.com Subject: How to mount a snapmirror lun w/o snapdrive Hi, I have a lun that resides in a qtree and I am using volume level snapmirroring to another filer. My primary filer has snapdrive for connecting and creating luns. My secondary filer does not have snapdrive. I want to mount my snapmirrored luns to another host. I understand there is a command line way of doing this. (lun -f). Has anyone done it? Also what is the difference between a lun.rws filer and a lun.lun file? Thanks, SM

1 0

Oracle 11i Application Suite on NetApp?
by Stewart, John 17 Feb '06

17 Feb '06

Does anyone here run the Oracle 11i Application suite (ERP) on a NetApp, via iSCSI or otherwise? We're considering a 3020 as part of our big Oracle hardware replacement project, and wanted to talk to some folks who have actually done it. The concern from the DBA folks is how a RAID4/6 system (with WAFL) can possibly compete with a RAID0/1 system... thanks! johnS

2 1

How to mount a snapmirror lun w/o snapdrive
by Steven Mandrake 17 Feb '06

17 Feb '06

Hi, I have a lun that resides in a qtree and I am using volume level snapmirroring to another filer. My primary filer has snapdrive for connecting and creating luns. My secondary filer does not have snapdrive. I want to mount my snapmirrored luns to another host. I understand there is a command line way of doing this. (lun -f). Has anyone done it? Also what is the difference between a lun.rws filer and a lun.lun file? Thanks, SM

2 1

RE: Netapp vs eva800 with Microsoft Windows Storage Server 2003
by Carter K. George 17 Feb '06

17 Feb '06

well, for serving up windows CIFS file shares, the windows storage server platform is pretty strong. when combined with hp's polyserve-based cluster services, it has client transparent failover, access to the same shares from multiple nodes at the same time, and snapshot integration with the eva8000. now if the workload is NFS, that's another matter. since an hp windows storage server cluster scales linearly as you add nodes, the performance can be tremendous with an eva8000 on the backend. we have test results on CIFS showing linear scaling up to 1,600 Megabytes per second (not Megabits) on an 8 node Windows cluster, with all nodes mounting, reading, writing, and exporting the same file systems from all nodes concurrently, with automatic load balancing of client CIFS connections across the nodes. carter ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Langborg Tom Sent: Tuesday, February 14, 2006 11:36 PM To: toasters(a)mathworks.com Subject: Netapp vs eva800 with Microsoft Windows Storage Server 2003 Hi I need some arguments for netapp vs Microsoft Windows Storage Server 2003 with eva8000. I now that the doesn't have snappvault, opensnapvault. But how is uptime with Microsoft Windows Storage Server 2003 and max volumes performens and so on? The arguments that we have is the cost otherwise, we are complete happy with our fas940 and r200. We use 90% nas and 10% san. Regards tom

2 1

R100 ggregate size limited to 8 TByte?
by Peter W. Osel 17 Feb '06

17 Feb '06

Hello fellow toasters, I plan to migrate my R100 (84 136 Gbyte disks) (yes I should upgrade to newer systems ;-), to flexible volumes, ideally I would create a single aggregate (12+12+12+12+12+11+11 disks leaving 2 HotSpares). Someone mentioned that this would not be possible, and that the R100 has a limit of 8 Tbyte for the aggregate size, thus I had to create two aggregates? Is this true? could not find any hint on the NOW site yet ... Cheers --pwo

4 3

Malware incident
by VDSAdmin 17 Feb '06

17 Feb '06

Panda Antivirus has found the following malware in the message: Server : VDSEXCH Sent by : toasters(a)mathworks.com Address : toasters(a)mathworks.com To : Donald MacQueen Subject : Returned mail: Data format error Date : 17/02/2006 11:08:55 Sent by You File : attachment.exe inside: attachment.zip Virus : W32/Mydoom.N.worm - Deleted http://www.pandasoftware.com

1 0

unknown syslog-message
by josef radinger 17 Feb '06

17 Feb '06

got today messages on my netapp 3020c: [filer: nfsd.assist.queue.exceeding.limit:warning] <javascript:popEMS('nfsd.assist.queue.exceeding.limit');>: There are 132 items on the NFS assist queue, which exceeds the cutoff limit of 130. found nothing in my handbooks, and syslog-translator does not know what this would be. at the same time we had some severe performance-problems, but we dont know yet, if these were triggered by that message. is there something i could do? yours josef radinger it-services medical university innsbruck

4 4

← Newer
1
2
3
4
5
6
7
8
...
14
Older →

Jump to page:

2025

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 February 2006