toasters

toasters@lists.teaparty.net

13520 discussions

NDMP backup speeds
by Geoff Hardin 19 May '04

19 May '04

Toasters, What is the general consensus for NDMP backup speeds? I have four older F760Cs that are getting about 5-6 MB/s to local tape drives. The problem is that I also have another nine filers (1 F630, 1 F720, 1 F740, 6 F760C) that are backing up via 3-way NDMP and we are lucky to see 4-5 MB/s; most are in the 1-2 MB/s range. And don't even get me started on the Solaris/SunOS/Linux/HP-UX/DEC clients that backup through the drives attached to the filers (0.1 - 2 MB/s). We were looking at upgrading the tape drives and such, but we have done a few test dumps from filer to filer, filer to LTO-2, and filer to LTO-2 attached to a Sun, and the numbers didn't improve much (definitely not enough to justify buying a new tape library / tape technology). Just a little background: We are using DLT7000 tape drives, which should have a maximum through put of about 10 MB/s (2:1 compression), contained in a Quantum P3000 library. We have all our filers on two separate networks; one for general data sharing and a private one for our backups. The private network is all GbE (fibre) using the Alteon GbE cards (Gigabit Ethernet Controller II). We are doing qtree dumps, and while some of the filers have only four qtrees per volume, others have over 40. Each qtree is generally capped at 200 GB max, and for those filers that have lots of qtrees, we still try to limit each saveset to around 200 GB. So, I guess what I'm really asking is 1) "What kind of speeds are you seeing?" and 2) "Am I doing something wrong?" Many thanks, Geoff Hardin geoff.hardin(a)dalsemi.com Difficulty is the excuse history never accepts.

4 3

Undeliverable mail: Mail Delivery (failure boylebc@mail.bc.edu)
by MAILER-DAEMON＠fe3.bc.edu 19 May '04

19 May '04

Failed to deliver to '' Virus(es) found. message.scr is infected with W32/Netsky.p@MM Viruses: 1 Trojans: 0 Jokes: 0 Tests: 0

1 0

RE: NDMP backup speeds
by John Clear 18 May '04

18 May '04

On Tue, May 18, 2004 at 04:53:07PM -0500, Geoff Hardin wrote: > So, I guess what I'm really asking is 1) "What kind of speeds are you > seeing?" and 2) "Am I doing something wrong?" We get about 20MB/s to GigE SDLT drives. The drives have a built in NDMP server and are on a private GigE network with the filers. Our FAS960s can push 6+ streams at 20MB/s each, the F880 maxes out at about 5-6 streams. Our F760 used to manage 4 streams at 15+MB/s. We are switching over to LTO drives connected to DinoStor Tape Servers (http://www.dinostor.com) which do the same job as the built in NDMP server that our SDLT drives have. In our experience, and that of our NetApp SE, direct attached tape drives have worse performance then network attached drives. NetApps are optimized for sending data out of the network, not for dealing with tape drives. John

1 0

HP Omniback/Data Protector with NDMP integration option
by Derek Lai 18 May '04

18 May '04

Anyone else using HP Omniback to backup your filer? I inherited a FAS940 cluster and HP Omniback 4.10 environment. We are currently using a PC as a NTFS gateway to backup CIFS shares on NetApp. The performance is pretty poor and we are having all kinds of problems. I wanted to upgrade to the latest version which HP has renamed DataProtector 5.1. I also wanted to use NDMP to backup the CIFS shares. This is suppose to be free for us. I was told that the previous administrator had tried NDMP integration with Omniback for some time and then abandoned it. No one seems to know the reason though. Just wanted to get a feel for what experience others with similar environment have had. Thanks. Derek

1 0

RE: REQ: Survey: Maildir on Netapp Filers
by moritz.willers＠ubs.com 18 May '04

18 May '04

Has anyone tried running the sendmail *spool* dir on a filer? we tried and failed miserably, but didn't have time to investigate. - mo Visit our website at http://www.ubs.com This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. E-mail transmission cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of e-mail transmission. If verification is required please request a hard-copy version. This message is provided for informational purposes and should not be construed as a solicitation or offer to buy or sell any securities or related financial instruments.

6 5

Undeliverable mail: Hi
by MAILER-DAEMON＠fe3.bc.edu 17 May '04

17 May '04

Failed to deliver to '' Virus(es) found. game.zip/DATA.RTF .SCR is infected with W32/Netsky.p@MM Viruses: 1 Trojans: 0 Jokes: 0 Tests: 0

1 0

Undeliverable mail: Re: Test
by MAILER-DAEMON＠fe3.bc.edu 17 May '04

17 May '04

Failed to deliver to '' Virus(es) found. message_mahoneat.scr is infected with W32/Netsky.p@MM Viruses: 1 Trojans: 0 Jokes: 0 Tests: 0

1 0

formula for calculating the max number of disks on a loop without losing performance
by Derek Lai 17 May '04

17 May '04

We currently have a FAS940C cluster. Each of the FAS940 has 2 DS14 shelves of 36 GB drives and 1 DS14 shelf of 72 GB drives (a total of 6 shelves). Each of the FAS940 has only 1 FC dual-channel Host Adapter: slot 4: FC Host Adapter 4a (Dual-channel, QLogic 2312 rev. 2, 64-bit, L-port, <UP>) We are trying to isolate some performance problem we are encourtering. How do we find out if the host adapter is the bottleneck? Does anyone have a formula for calculating how many drives can be put on a single loop without losing performance. Thanks. Derek

3 2

None
by Harbhajan Julka 14 May '04

14 May '04

Hello everyone, I need help with implementing SCSI in NDMP for a NetApp filer. Anyone familiar with SCSI development will know the elements and the address assignnments of to those elements by CDB's sent to the SCSI device. I have used MODE_SENSE and READ_ELEMENT_STATUS in order to do that. But, i believe i am not unpacking the reply correctly to assign addresses to the elements. Its because when i send the POSITION_TO_ELEMENT or MOVE_MEDIUM, i use those elements addresses (Transport element, Data Transfer element and Storage element) and i get sense key indicating that the CDB was incorrect. Need help urgently. I request all the SCSI experts to reply to this. Thanks in advance, Sonia. --------------------------------- Do you Yahoo!? SBC Yahoo! - Internet access at a great low price.

1 0

Data ONTAP 6.1.3 running on an F520
by Joe C Ropar 14 May '04

14 May '04

Greetings Toaster Fans, Has anyone out there successfully run ONTAP 6.1.3 or later on an F520 running both CIFS and NFS? We are planning on migrating the data to an R150 later this year but are doing an AD migration which is not supported under 5.3.7R3 and are looking for a quick fix. We have done some preliminary testing with a spare head and 6.1.3 installed without a hitch. In addition we have created some CIFS share which are working correctly. Regards, Joseph C. Ropar Unix Systems Administrator Raytheon Systems Company cell (214) 869-7901 desk (972) 952-4306 e-mail jropar(a)raytheon.com

2 1

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