toasters

toasters@lists.teaparty.net

13520 discussions

RE: ontap 6.0 only 350 cif-sessions?
by Bagg, Rod 25 Sep '00

25 Sep '00

We have identified bug 30873 which can impact CIFS performance in ONTAP 6.0 when the number of active CIFS sessions on a filer exceeds 200. Symptoms of the problem can be client session disconnects, hangs or very poor CIFS performance. A 'cifs stat' would typically show the counter named 'Max pBlk Exhaust' to be a very high number - 100,000s or more. We have identified the root cause of this problem and have developed the fix which is currently undergoing QA and load testing. ONTAP release 6.0R1 is tentatively scheduled to be released by the end of this week to resolve this bug. You can view the online bug report on the NOW site at: http://now.netapp.com/NOW/cgi-bin/bugfind.on?bugs=30873&rels= On this same link, you can click on "Bug Watcher" to be automatically notified when the bug fix has been released. Rod Bagg Manager, Sustaining Engineering Network Appliance 408-822-6042 (office) 408-205-9533 (cell) 408-994-2696 (pager) 408-822-4410 (fax) Get answers NOW! - NetApp On the Web http://now.netapp.com > -----Original Message----- > From: Datta, Diptish > Sent: Monday, September 25, 2000 12:19 PM > To: Bagg, Rod > Subject: FW: ontap 6.0 only 350 cif-sessions? > > > From: andrew-m.webster(a)db.com [mailto:andrew-m.webster@db.com] > <mailto:[mailto:andrew-m.webster@db.com]> > Sent: Friday, September 22, 2000 8:17 AM > To: ulrich.neumann(a)brose.de > Cc: toasters(a)mathworks.com > Subject: Re: ontap 6.0 only 350 cif-sessions? > > > > > Greetings from Japan, > > We had much the same issue. If you are seeing the same thing, then the "Max > pBlk Exhaust" from the "cifs stat" goes into the millions. It's normally at 0. > We had to rollback to OnTap 5, which wasn't that easy. > We basically had a full day outage, which for an Investment Bank was a very > bad thing. :-( > Regards > > -- > Andrew Webster > PC, NetApp Filer, TSM > Sanno Move Project > Deutsche Real Estate Consulting Ltd. - Tokyo > Internal (but outside Tokyo): 88 408-7542 > Tel. +81-3-5156-7542 > Fax. +81-3--5156-4811 > Mobile: 0907 184 8958 > > Andrew-M.Webster(a)db.com > > > > ---------------------------------------- Message History > ---------------------------------------- > > From: ulrich.neumann(a)brose.de on 22/09/2000 12:16 PM GMT > > > > To: toasters(a)mathworks.com > cc: > Subject: ontap 6.0 only 350 cif-sessions? > > > > we just upgrated to ontap 6.0 on out clustered f760. unfortunately wehn more > than 350 connection over cifs are established, the filer isn't responding > anymore. the http and nfs-connections word furthermore. only the cifs > connections are lost an d cannot be established. after a reboot the filer > works till around 350 connections. > > any ideas? > > greetings from germany > Brose Fahrzeugteile > Client-Server-Technologien > > > Ulrich Neumann > > >

1 0

F840
by Jeff Stampes 25 Sep '00

25 Sep '00

Okay, now that the press release is out, and we all know the technical specs of the 840, what I want to know is: Who's using one? Is the throughput as good as the specs show? C'mon, spill it. We've got one on order, and will put it through it's paces as soon as we get it. We use it specifically for multi-client compilation of our software. I through around 20 Solaris systems (mmostly Ultra80's and Ultra60's), a dozen or so HPC3000 systems and a dozen or so NT boxes at it to compile all three platforms we support, and we currently are stressing out our 760. I'll let y'all know what i find. -Jeff

2 2

ontap 6.0 only 350 cif-sessions?
by Ulrich Neumann 25 Sep '00

25 Sep '00

we just upgrated to ontap 6.0 on out clustered f760. unfortunately wehn more than 350 connection over cifs are established, the filer isn't responding anymore. the http and nfs-connections word furthermore. only the cifs connections are lost an d cannot be established. after a reboot the filer works till around 350 connections. any ideas? greetings from germany Brose Fahrzeugteile Client-Server-Technologien Ulrich Neumann

2 1

RE: Presenting Netapp as a location disk
by Ulrich Neumann 25 Sep '00

25 Sep '00

i don't see any benefit for doing it this way. you can use a seperate gigabit-connection from nt-box to the toaster. then use the "net use" command or say to frontpage to use "\\toaster\inetpub". so you have near the performance of a san and the toaster as your storage. new -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com]On Behalf Of Mark D Simmons Sent: Montag, 25. September 2000 00:43 To: Richard Cc: toasters(a)mathworks.com Subject: Re: Presenting Netapp as a location disk On Wed, Sep 20, 2000 at 01:35:13PM +0100, Richard wrote: > I know this is not really, a question that applies just to Netapp's but at the moment > I am looking at a SAN that allows the shared remote disk array to be seen by a NT > system as if it is a local disk drive (this helps make frontpage happy). And I am trying > to find a way to do this with a Netapp as a remote filestore, surly someone has written > a NTFS filesystem driver for NT that reads from a network drive, but presents itself as > a local drive? If a SAN can offer this it must be possible? I've informally mentioned that the same thing would be very nice on Unix too, both in the distant, and recent past. Currently I manage no filers so I cannot really RFE this. As a concept it's in serious conflict with the appliance philosphy. However, I think the benefits to some users might even justify their stumping up cash separately to buy it.. -- -Mark ... an Englishman in Derby ...

1 0

Presenting Netapp as a location disk
by Richard 24 Sep '00

24 Sep '00

I know this is not really, a question that applies just to Netapp's but at the moment I am looking at a SAN that allows the shared remote disk array to be seen by a NT system as if it is a local disk drive (this helps make frontpage happy). And I am trying to find a way to do this with a Netapp as a remote filestore, surly someone has written a NTFS filesystem driver for NT that reads from a network drive, but presents itself as a local drive? If a SAN can offer this it must be possible? Richard richard(a)gits.co.uk

2 1

RE: ontap 6.0 only 350 cif-sessions?
by Ulrich Neumann 24 Sep '00

24 Sep '00

the update was driven from technics of amdahl. i don't know if they run the wat. i don't think so. a guy told me that the script would not be working and crashed - but i'm not shure. a further problem for me is, that i'm not the admin of the toaster - i'm the admin of the nt-domain(s) ... so another team did the work together with amdahl - i stand beside and look for 16 hours what they are doing ... it was not the gold-medal for amdahl ... now its sunday the sun is shining and i must cook potatoes for my family. i think on monday the would be new news ... have a nice weekend (i haven't - we copied all data back on our compaq servers ...) newmy -----Original Message----- From: Sam Rafter [mailto:rafter@netapp.com] Sent: Freitag, 22. September 2000 23:25 To: andrew-m.webster(a)db.com Cc: ulrich.neumann(a)brose.de; toasters(a)mathworks.com Subject: Re: ontap 6.0 only 350 cif-sessions? Hey! Is anybody running the Workload Analysis Tool? It should be popping up on your screen before you download 6.0, to see if your workload is going to be impacted by upgrading to 6.0. I'm curious to know if the WAT isn't working, or if people aren't running it. Thanks! andrew-m.webster(a)db.com wrote: > > Greetings from Japan, > > We had much the same issue. If you are seeing the same thing, then the "Max pBlk Exhaust" from the "cifs stat" goes into the millions. It's normally at 0. We had to rollback to OnTap 5, which wasn't that easy. > We basically had a full day outage, which for an Investment Bank was a very bad thing. :-( > > Regards > > -- > Andrew Webster > PC, NetApp Filer, TSM > Sanno Move Project > Deutsche Real Estate Consulting Ltd. - Tokyo > Internal (but outside Tokyo): 88 408-7542 > Tel. +81-3-5156-7542 > Fax. +81-3--5156-4811 > Mobile: 0907 184 8958 > > Andrew-M.Webster(a)db.com > > ---------------------------------------- Message History ---------------------------------------- > > From: ulrich.neumann(a)brose.de on 22/09/2000 12:16 PM GMT > > To: toasters(a)mathworks.com > cc: > Subject: ontap 6.0 only 350 cif-sessions? > > we just upgrated to ontap 6.0 on out clustered f760. > unfortunately wehn more than 350 connection over cifs are established, > the filer isn't responding anymore. > the http and nfs-connections word furthermore. only the cifs connections > are lost an d cannot be established. > after a reboot the filer works till around 350 connections. > > any ideas? > > greetings from germany > > Brose Fahrzeugteile > Client-Server-Technologien > > Ulrich Neumann -- Sam Rafter Escalations Jerk Network Appliance rafter(a)netapp.com

1 0

Re: Slow automount in Solaris 8
by Darrell Fuhriman 23 Sep '00

23 Sep '00

benny.wistrom(a)cn.saab.se writes: > I'm having trouble with automount between a F760 filer and Solaris 8. > When I cd to a nfs-mounted directory it takes approximately 15 seconds > before it's done. Has anyone had this problem? If so, how did you solve > it? Where are your automount maps coming from? Make sure /etc/nsswitch.conf directly reflects reality. Darrell

1 0

Re: ontap 6.0 only 350 cif-sessions?
by andrew-m.webster＠db.com 23 Sep '00

23 Sep '00

Hi Ya, Well I remember seeing the WAT tool, reading the initial page but it talk abouting setting it up on a "unix" box, so I quickly lost interest as I have very limited to access to any Unix boxes. The problem is definately not I/O or CPU related. During these problems the CPU and I/O activity is quite low. Console and NFS response are excellent, but as soon as you do any CIFS commands or CIFS file access then everything becomes very very slow. Regards -- Andrew Webster PC, NetApp Filer, TSM Sanno Move Project Deutsche Real Estate Consulting Ltd. - Tokyo Internal (but outside Tokyo): 88 408-7542 Tel. +81-3-5156-7542 Fax. +81-3--5156-4811 Mobile: 0907 184 8958 Andrew-M.Webster(a)db.com ---------------------------------------- Message History ---------------------------------------- From: rafter(a)netapp.com on 22/09/2000 09:24 PM GMT To: Andrew-M Webster/Tokyo/DBJapan/DeuBa@DBJapan cc: ulrich.neumann(a)brose.de; toasters(a)mathworks.com Subject: Re: ontap 6.0 only 350 cif-sessions? Hey! Is anybody running the Workload Analysis Tool? It should be popping up on your screen before you download 6.0, to see if your workload is going to be impacted by upgrading to 6.0. I'm curious to know if the WAT isn't working, or if people aren't running it. Thanks! andrew-m.webster(a)db.com wrote: > > Greetings from Japan, > > We had much the same issue. If you are seeing the same thing, then the "Max pBlk Exhaust" from the "cifs stat" goes into the millions. It's normally at 0. We had to rollback to OnTap 5, which wasn't that easy. > We basically had a full day outage, which for an Investment Bank was a very bad thing. :-( > > Regards > > -- > Andrew Webster > PC, NetApp Filer, TSM > Sanno Move Project > Deutsche Real Estate Consulting Ltd. - Tokyo > Internal (but outside Tokyo): 88 408-7542 > Tel. +81-3-5156-7542 > Fax. +81-3--5156-4811 > Mobile: 0907 184 8958 > > Andrew-M.Webster(a)db.com > > ---------------------------------------- Message History ---------------------------------------- > > From: ulrich.neumann(a)brose.de on 22/09/2000 12:16 PM GMT > > To: toasters(a)mathworks.com > cc: > Subject: ontap 6.0 only 350 cif-sessions? > > we just upgrated to ontap 6.0 on out clustered f760. > unfortunately wehn more than 350 connection over cifs are established, > the filer isn't responding anymore. > the http and nfs-connections word furthermore. only the cifs connections > are lost an d cannot be established. > after a reboot the filer works till around 350 connections. > > any ideas? > > greetings from germany > > Brose Fahrzeugteile > Client-Server-Technologien > > Ulrich Neumann -- Sam Rafter Escalations Jerk Network Appliance rafter(a)netapp.com

1 0

Re: ontap 6.0 only 350 cif-sessions?
by andrew-m.webster＠db.com 22 Sep '00

22 Sep '00

Greetings from Japan, We had much the same issue. If you are seeing the same thing, then the "Max pBlk Exhaust" from the "cifs stat" goes into the millions. It's normally at 0. We had to rollback to OnTap 5, which wasn't that easy. We basically had a full day outage, which for an Investment Bank was a very bad thing. :-( Regards -- Andrew Webster PC, NetApp Filer, TSM Sanno Move Project Deutsche Real Estate Consulting Ltd. - Tokyo Internal (but outside Tokyo): 88 408-7542 Tel. +81-3-5156-7542 Fax. +81-3--5156-4811 Mobile: 0907 184 8958 Andrew-M.Webster(a)db.com ---------------------------------------- Message History ---------------------------------------- From: ulrich.neumann(a)brose.de on 22/09/2000 12:16 PM GMT To: toasters(a)mathworks.com cc: Subject: ontap 6.0 only 350 cif-sessions? we just upgrated to ontap 6.0 on out clustered f760. unfortunately wehn more than 350 connection over cifs are established, the filer isn't responding anymore. the http and nfs-connections word furthermore. only the cifs connections are lost an d cannot be established. after a reboot the filer works till around 350 connections. any ideas? greetings from germany Brose Fahrzeugteile Client-Server-Technologien Ulrich Neumann

2 1

Slow automount in Solaris 8
by benny.wistrom＠cn.saab.se 22 Sep '00

22 Sep '00

Hi! I'm having trouble with automount between a F760 filer and Solaris 8. When I cd to a nfs-mounted directory it takes approximately 15 seconds before it's done. Has anyone had this problem? If so, how did you solve it? Regards Benny

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