toasters

toasters@lists.teaparty.net

3 participants
13514 discussions

RE: wafl scan reallocate
by Glenn Walker 08 Mar '06

08 Mar '06

I agree with Herbert's comments that it works exceptionally well in 7.x! Be warned, if you reallocate the source, the SM updates are going to be large and take a long time (potentially, depending on how much data is shifted around). As for measure_layout: it has always been my experience that this is something good to look at, but not the rosetta stone for fragmentation... it is an average of all the blocks on the volume and how fragmented they are. What I've always looked at to determine fragmentation is statit output (from perfstat) as it drills down a bit more to what the disks are actually doing. That by itself is also sometimes misleading, however... I'm sure NetApp support will review both to come up with a more accurate prediction. Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Herbert Langenbach Sent: Tuesday, February 21, 2006 4:26 AM To: philip.boyle(a)eircom.net; toasters(a)mathworks.com Subject: RE: wafl scan reallocate Philip, a wafl scan reallocate is just a defragmentation. Did NetApp ask you to run a wafl scan measure_layout? What was the result? BTW, I do not remember any improvement with reallocate in ONTAP 6.x It moves blocks around and comes back with pretty much the same fragmentation as it was before. It really works in 7.x However, to rule out fragmentation you need the measure_layout value. Herbert -----Original Message----- From: Philip Boyle [mailto:philip.boyle@eircom.net] Sent: Tuesday, February 21, 2006 9:06 AM To: toasters(a)mathworks.com Subject: wafl scan reallocate Hi guys, I am seeing increased cpu on a filer. It appears to jump when the filer kicks off a number of snapmirrors. I've updated the threshold so that the mirroring runs a little more frequent and this appears to have reduced the high cpu issue. Note, I'm only running approximately 4 snap mirror threads on a clustered environment so I'm not hitting any mirroring thread limitations. This increase in cpu is unexpected as the filer is running around 50% less snapmirror threads and 40% less volumes from 6 months ago. so if anything I should see a drop in cpu activity. Support are advising me to run a 'wafl scan reallocate' as one volume is close to 80% full. I'm not convinced and reluctant to kick off wafl commands. For your information, I'm running 6.5.3P4 on a 940C cluster. Has anyone else had similar issues or have ran 'wafl scan reallocate'. Any recommendations for debugging? I've passed on the relevant 'sysstat' data for a 24hr period to Netapp Support. Regards, Philip --

3 2

IO stats on NetApp
by Stewart, John 08 Mar '06

08 Mar '06

Is there any way to show actual disk IO operations per second. sysstat shows cifs and nfs ops, but we're wondering how that translates into actual IOps underneath the hood (or is it all hidden away)? thanks johnS

3 2

RE: IO stats on NetApp
by Callaway, Michael 08 Mar '06

08 Mar '06

Don't iops show on the front panel? Thanks, Michael NTTA/IT http://www.ntta.org 214.461.2033 -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Stewart, John Sent: Tuesday, March 07, 2006 4:35 PM To: 'toasters(a)mathworks.com' Subject: IO stats on NetApp Is there any way to show actual disk IO operations per second. sysstat shows cifs and nfs ops, but we're wondering how that translates into actual IOps underneath the hood (or is it all hidden away)? thanks johnS

2 1

Re: IO stats on NetApp
by uabmbn＠uab.ericsson.se 08 Mar '06

08 Mar '06

>Is there any way to show actual disk IO operations per second. sysstat >shows cifs and nfs ops, but we're wondering how that translates into actual >IOps underneath the hood (or is it all hidden away)? Well, I guess the problem is that there's not one clear definition of "disk IO operation". At what level in the protocol stacks...? A.s.o. sysstat -x talks about kB/s for the physical devices which is good enough IMO. Anyway. -x has been there for "ages", can't remember when it appeared in ONTAP but it wasn't recently. You do write IOps as disk IO operations. This may not necessarily be the same as FSops, i.e. operations in the WAFL FS per time unit. That's not shown by sysstat -x AFAIK. Then again it is probably an issue of definition of the term "FSop". I think it would be hard for ppl to understand what it is they're seeing, i.e. what the internal unit "op" really means. Cheers, /M -- Michael Bergman Email: Michael.Bergman(a)ericsson.com Sr Systems Analyst EAB/OET/R, R&D IT Stockholm Phone: +46 8 7275709 R&D IT Management Armborstv 14, S-125 25 Alvsjo, Sweden -- "Language: a dialect spoken by some ppl with a large fleet and strong army."

1 0

RE: IO stats on NetApp
by Glenn Walker 08 Mar '06

08 Mar '06

Those are FCP\CIFS\NFS IOPS... not the same. The command you're looking for is sysstat (kb/s on disk reads\writes) or statit (diag level command for actual IOPS). Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Callaway, Michael Sent: Tuesday, March 07, 2006 6:16 PM To: Stewart, John; toasters(a)mathworks.com Subject: RE: IO stats on NetApp Don't iops show on the front panel? Thanks, Michael NTTA/IT http://www.ntta.org 214.461.2033 -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Stewart, John Sent: Tuesday, March 07, 2006 4:35 PM To: 'toasters(a)mathworks.com' Subject: IO stats on NetApp Is there any way to show actual disk IO operations per second. sysstat shows cifs and nfs ops, but we're wondering how that translates into actual IOps underneath the hood (or is it all hidden away)? thanks johnS

1 0

RE: Using old FC shelves without filer (EM)
by Glenn Walker 08 Mar '06

08 Mar '06

That makes it sound a lot like the old FC7 shelves... Even worse than FC8/9! -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Michael Bergman Sent: Tuesday, March 07, 2006 3:35 PM To: spork(a)bway.net Cc: toasters(a)mathworks.com Subject: Re: Using old FC shelves without filer (EM) >Right, I figured that part out, but below the two FC ports there is >another module with two DB-9s marked "Comm 1" and "Comm 2". Just >curious >what those do. Anyone here ever console into those? Well, yes... sort of. But not "console". We used those modules here, it's long time ago now but I'm pretty sure that those serial ports are what I think they are: the DB9's are RS232 interfaces into the EM (Environmental Module). With these old shelves, the signal path for the EM controls were not in-band SCSI but on these separate DB9 serial ports :-( They were daisy chained with separate serial cables. I doubt that you'd be able to interface with that EM protocol on another system anyway, to my knowledge no tech info on this is readily available. Besides, NetApp had big problems with quality on those old EM's replaced them with new ones -- those had in-band SCSI signaling and hence no serial ports on them. It's not unlikely that those EM's are broken in some way anyway, they must be really old by now. So, just ignore those serial ports. There'll be no way for you to make the host system read environ-info from the shelves, but AFAIK they'll signal with alarms (audio and visual) on the actual shelf as well. Hope so at least... :-\ Good luck, /M -- Michael Bergman Email: Michael.Bergman(a)ericsson.com Sr Systems Analyst EAB/OET/R, R&D IT Stockholm Phone: +46 8 7275709 R&D IT Management Armborstv 14, S-125 25 Alvsjo, Sweden -- "Qui vicit non est victor nisi victus fatetur." - Ennius

1 0

Re: Using old FC shelves without filer (EM)
by Michael Bergman 07 Mar '06

07 Mar '06

>Right, I figured that part out, but below the two FC ports there is >another module with two DB-9s marked "Comm 1" and "Comm 2". Just >curious >what those do. Anyone here ever console into those? Well, yes... sort of. But not "console". We used those modules here, it's long time ago now but I'm pretty sure that those serial ports are what I think they are: the DB9's are RS232 interfaces into the EM (Environmental Module). With these old shelves, the signal path for the EM controls were not in-band SCSI but on these separate DB9 serial ports :-( They were daisy chained with separate serial cables. I doubt that you'd be able to interface with that EM protocol on another system anyway, to my knowledge no tech info on this is readily available. Besides, NetApp had big problems with quality on those old EM's replaced them with new ones -- those had in-band SCSI signaling and hence no serial ports on them. It's not unlikely that those EM's are broken in some way anyway, they must be really old by now. So, just ignore those serial ports. There'll be no way for you to make the host system read environ-info from the shelves, but AFAIK they'll signal with alarms (audio and visual) on the actual shelf as well. Hope so at least... :-\ Good luck, /M -- Michael Bergman Email: Michael.Bergman(a)ericsson.com Sr Systems Analyst EAB/OET/R, R&D IT Stockholm Phone: +46 8 7275709 R&D IT Management Armborstv 14, S-125 25 Alvsjo, Sweden -- "Qui vicit non est victor nisi victus fatetur." - Ennius

1 0

RE: rsh times out after 5 min
by Sphar, Mike 07 Mar '06

07 Mar '06

I agree totally. I don't mean to lean on anyone to change anything, just wanted to point out there are decent free ssh utilities for windows. (I use SecureCRT personally as I've used it for a long time, but it's not free.) -- Michael W. Sphar - IS&T - Lead Systems Administrator SMBU Engineering Support Services, BMC Software -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Chris Thompson Sent: Tuesday, March 07, 2006 1:35 PM To: toasters(a)mathworks.com Subject: Re: rsh times out after 5 min Mike Sphar writes > There are free/open-source ssh utilities for windows, like putty. Yes, yes, but ... It seems quite inappropriate to lean on William to convert from rsh to ssh until we know what the cause of his 5-minute cutoff is. For all we know, it will apply just as much to ssh as to rsh. If it's an "idle"-TCP-connection detector somewhere, it probably will. -- Chris Thompson Email: cet1(a)cam.ac.uk

1 0

RE: Using old FC shelves without filer
by Glenn Walker 07 Mar '06

07 Mar '06

They sound like FC8\FC9 shelves (the NetApp designated part). If I recall correctly, those old shelves were made for NetApp by a company called xyratex (www.xyratex.com) but I'm not sure you're going to find much about them. Happy hunting. As for the comm ports, those are not comm ports... they sound like the DB9\FCAL ports that you use to transfer data (using copper cables). Old filer? You're talking about a 700 series or older, or maybe an 800 series if you can find one. Issues: in order to make it work, you need to license it - NetApp won't just give you licenses since you picked it up on resell and they'll charge a premium for supporting it. You're probably looking around $1500 on the low end for old stuff that doesn't have any new software versions coming out. From the sound of it you may be better off building a linux machine with some of those new 1TB harddrives for the same money. If money was no object, I'd suggest a FAS270... but $40k is probably out of your budget (like mine). :) Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Charles Sprickman Sent: Friday, March 03, 2006 9:07 PM To: toasters(a)mathworks.com Subject: Using old FC shelves without filer Hello all, I recently had three NetApp shelves dumped in my lap (ouch!). They were attached to a host with a (now borked) Mylex ExtremeRAID 3000. I'm not having much luck finding any hardware RAID cards that compare to the Mylex card and don't cost more than a few grand... Anyhow, I may just get a simple FC card and do software RAID under FreeBSD using the new GEOM RAID tools. But before I go too far I'm trying to find some documentation on these shelves. They are FC/copper shelves. There are labels on them that read: Part Code - XL401R-FJ-02D Part Code - XL401R-FJC-02D Part Code - XL401R-FJF-02D Revision - A "Made in Ireland" On the back each shelf has a rear panel flanked by two power supplies. The rear panel has three "cards": FC In/Out, Comm1/2 w/ temp/mute buttons, shelf id select with port a/b expansion switches. Each shelf has 7 36GB drives. Some digging reveals the company that made these for NetApp was bought by Adaptec a few years ago. Where can I find some docs on the shelves themselves, specifically the two "COMM" ports? I tried hooking a terminal up to each, but no response after trying a variety of speeds. Lastly, what kind of money would I be looking at to buy a used Filer? This would be primarily for hosting mail and web stuff. We're talking about up to 10,000 mailboxes (Maildir) and a few hundred low-volume websites. If I go used, I would ideally like to buy a spare filer, or a bunch of spare parts for a filer if possible... New is way out of the budget. Thanks, Charles

3 2

Re: Fwd: rsh times out after 5 min
by Tim McCarthy 07 Mar '06

07 Mar '06

Nope. Sorry. Prior to 7.0, NetApp issued a pass-through license fee from RSA fo ruse of the SSH implementation. In 7.0, that module was completely rewritten in C code (which makes it significantly faster than the JAVA code in pre-7 releases). It was developed from the OpenSSH stuff which is free and requires no pass through license fee. If you *really* want SSH, you should look at upgrading to 7.x if you can. (i.e. out of luck with a 700 series or F85) Sheldon Mustard wrote: > On 3/7/06, Tim McCarthy <tmacmd(a)gmail.com> wrote: >> Have you tried ssh? I use it exclusively. It is easy to setup to use >> without passwords.... > > Does ssh work in 6.5 without an additional license? I have been able > to get ssl working but but not ssh ... any ideas? When I tell it to > enable ssh2 it says "SecureAdmin is not installed", do I need to > install some binaries from the now site. > > SJM > > -- > Sheldon Mustard > smustard(a)gmail.com > "There will be no order, only chaos." - Pi (1998) > -- > >

1 0

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