toasters February 2004

toasters@lists.teaparty.net

57 participants
48 discussions

F760 question
by Scott Baker 04 Mar '04

04 Mar '04

I was hoping to find an answer in the archives, but no luck. We've picked up a NetApp F760 and have had no problems with the system. Recently we've tried adding more disks to the system, but seem to be hitting a limit. The config looks like 14x72GB disks (FC9 BCS) 14x144GB disks (DS14) This takes us to the maximum of the F760, 3TB, but we can only seem to add 14x72GB disks (1 72GB for spare, 2 RAID groups) 8x144GB disks (1 144GB for spare, 1 RAID group) These disks are all in one volume, /vol/vol0, but we can't seem to add any of the additional 5 144GB disks to the volume. When we try to add the disks the system tells us it can't do this. From what I've read we should simply need to indicate how many disks we want to add to the volume (ie 5) and it should create a new RAID group if needed. In this case our RAIDGROUP size is set to 8, so it should create the new RAIDGROUP. But this isn't happening. We're running DoT 6.4R1. Have we hit the maximum filesystem size for the F760? I'm not sure what the limit is. We're well within the 56 disks/loop limit. Any help would be appreciated. If we have hit a filesystem size, then we'll probable have to create a new volume (which we'd prefer not to do). However, if we do have to do this, any suggestions on how to make the new volume appear to simply look like part of the other exported filesystem? Cheers, SMB --------------------------------- Do you Yahoo!? Get better spam protection with Yahoo! Mail

2 2

Re: Cluster failover question
by Davin Milun 28 Feb '04

28 Feb '04

At 15:20, on Feb 25, 2004, jeff.mery(a)ni.com wrote: > We've also done rolling upgrades of DOT as mentioned below on our old F760 > cluster. Upgrade one head, fail over, reboot that head, giveback, upgrade > the other head, fail over, reboot the second head, and giveback. The > first head will complain that it's not at the same DOT level as its > partner, but we've never had a problem with data integrity. This also > only happens for a short period between the first giveback and the second > failover. I don't know if this is supported or not so as always, test if > you can. > > One note, we've never done this for major version upgrades (i.e. 5.x to > 6.x) this way. IIRC we've only done minor version upgrades this way (i.e. > 6.4.1 to 6.4.2 or 6.4.1P2 to 6.4.2). > > Jeff Does this work? Is it safe? We've always been told fairly explicitly by NetApp that this will not work. And have never tried it, for risk of ending up in an unstable or inconsistent state. I've always wished that I could do this - because with many other devices, this is a standard method of using a clustered-pair of devices to avoid a downtime, while doing an OS upgrade. But NetApp has never supported it. Davin. -- Davin Milun E-mail: milun(a)cse.Buffalo.EDU

2 1

RE: F760 question
by Reinoud Reynders 28 Feb '04

28 Feb '04

The max volume size of a F760 is 2 TB. So you must create at least 2 volumes. Best regards, Reinoud Reynders University Hospital Leuven Belgium -----Oorspronkelijk bericht----- Van: Scott Baker [mailto:buyandsell2k5@yahoo.com] Verzonden: za 28/02/2004 4:17 Aan: toasters(a)mathworks.com CC: Onderwerp: F760 question I was hoping to find an answer in the archives, but no luck. We've picked up a NetApp F760 and have had no problems with the system. Recently we've tried adding more disks to the system, but seem to be hitting a limit. The config looks like 14x72GB disks (FC9 BCS) 14x144GB disks (DS14) This takes us to the maximum of the F760, 3TB, but we can only seem to add 14x72GB disks (1 72GB for spare, 2 RAID groups) 8x144GB disks (1 144GB for spare, 1 RAID group) These disks are all in one volume, /vol/vol0, but we can't seem to add any of the additional 5 144GB disks to the volume. When we try to add the disks the system tells us it can't do this. From what I've read we should simply need to indicate how many disks we want to add to the volume (ie 5) and it should create a new RAID group if needed. In this case our RAIDGROUP size is set to 8, so it should create the new RAIDGROUP. But this isn't happening. We're running DoT 6.4R1. Have we hit the maximum filesystem size for the F760? I'm not sure what the limit is. We're well within the 56 disks/loop limit. Any help would be appreciated. If we have hit a filesystem size, then we'll probable have to create a new volume (which we'd prefer not to do). However, if we do have to do this, any suggestions on how to make the new volume appear to simply look like part of the other exported filesystem? Cheers, SMB _____ Do you Yahoo!? Get better spam protection with Yahoo! Mail <http://us.rd.yahoo.com/mailtag_us/*http://antispam.yahoo.com/tools>

1 0

RV: CPU workload increase after upgrade
by Carlos Alvarez 27 Feb '04

27 Feb '04

-----Mensaje original----- De: Carlos Alvarez [mailto:carlosa@ddmsa.com] Enviado el: Viernes, 27 de Febrero de 2004 10:35 a.m. Para: 'Jack Lyons' Asunto: RE: CPU workload increase after upgrade I have 18GB ZCS in F740 so, does this feature count? if it so could I disable? Thanks -----Mensaje original----- De: Jack Lyons [mailto:jack.lyons@martinagency.com] Enviado el: Viernes, 27 de Febrero de 2004 10:13 a.m. Para: 'Carlos A. Alvarez' Asunto: RE: CPU workload increase after upgrade I think by default with 6.x and above block checksums are enabled by default where as before 6.x they were not. That might account for the increase. Jack -----Original Message----- From: Carlos A. Alvarez [mailto:carlosa@ddmsa.com] Sent: Friday, February 27, 2004 7:59 AM To: toasters(a)mathworks.com Subject: CPU workload increase after upgrade Hi All, I recently upgrade a F740 from 5.3.4 to 6.4.1 without problems except that the CPU workload increase 10 % with the same traffic. Is this performace correct? Are new releases more CPU intensive than olders? Any comment would be apreciated. Thanks This email and its contents may be confidential. If it is and you are not the intended recipient, please do not disclose or use the information within this email or its attachments. If you have received this email in error, please delete it immediately. Thank you.

3 3

CPU workload increase after upgrade
by Carlos A. Alvarez 27 Feb '04

27 Feb '04

Hi All, I recently upgrade a F740 from 5.3.4 to 6.4.1 without problems except that the CPU workload increase 10 % with the same traffic. Is this performace correct? Are new releases more CPU intensive than olders? Any comment would be apreciated. Thanks

2 1

RE: rsh on NetCache 5.6
by Heal, Kenneth 26 Feb '04

26 Feb '04

Hi There is an RFE (Request for Enhancement) to allow you to use the /etc/hosts.equiv for this in a similar way to what is possible on the filer; it is not currently scheduled. The RFE number is 73672. It might be worth contacting Sales to push to get this implemented. Best regards Kenneth Heal -----Original Message----- From: Kerry Herschel [mailto:margaret.k.herschel@jpl.nasa.gov] Sent: Thursday, February 26, 2004 8:15 PM To: Armstrong, Larry Cc: toasters(a)mathworks.com Subject: Re: rsh on NetCache 5.6 Thanks Larry. It looks like I have to use the root password: -l root:password when I rsh. I don't have to with DOT and I'm not sure why I have to with NetcCache 5.6. I don't want to have a password in clear text go out on our network. But it looks like you don't have to use the password. Is that correct? Thanks, Kerry At 2/26/2004 10:57 AM, Larry Armstrong wrote: >Howdy Kerry! > >I'm not sure if it's obvious in the docs, but it should be supported. > >$ rsh C3100 -lroot version >NetApp Release 5.6: Wed Jan 14 16:20:46 PST 2004 > >---larry >:) >_ >* Kerry Herschel (margaret.k.herschel(a)jpl.nasa.gov) done spit this rhetoric: > > Hi, > > > > I haven't found any documentation for rsh on Netcache 5.6. Does anyone > know > > if it is supported? > > > > Thanks, > > > > Kerry > > > > ================================================================ > > Margaret.K.Herschel(a)jpl.nasa.gov > > JPL Information Services > > Jet Propulsion Laboratory > > Phone 818.354.1111 > > > > DISCLAIMER: The personal and professional opinions presented herein > > are my own and do not, in any way, represent the opinion or policy of > > JPL. > > > > > > > >-- >Larry Armstrong ><surfer(a)netapp.com> > >"This space intentionally left blank" >------------------------------------------------------- ================================================================ Margaret.K.Herschel(a)jpl.nasa.gov JPL Information Services Jet Propulsion Laboratory Phone 818.354.1111 DISCLAIMER: The personal and professional opinions presented herein are my own and do not, in any way, represent the opinion or policy of JPL.

1 0

rsh on NetCache 5.6
by Kerry Herschel 26 Feb '04

26 Feb '04

Hi, I haven't found any documentation for rsh on Netcache 5.6. Does anyone know if it is supported? Thanks, Kerry ================================================================ Margaret.K.Herschel(a)jpl.nasa.gov JPL Information Services Jet Propulsion Laboratory Phone 818.354.1111 DISCLAIMER: The personal and professional opinions presented herein are my own and do not, in any way, represent the opinion or policy of JPL.

2 3

RE: Oracle on CIFS?
by Jeff Asher 26 Feb '04

26 Feb '04

You will probably also want to modify the host registry so that the Oracle services are dependant on and start after the lanmanworkstation service. Without those edits the oracle services may fail to start on reboot because they can't access the remote shares. Jeff Asher ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Yarmas, Tom Sent: Thursday, February 26, 2004 10:39 AM To: Dirk Seidel; toasters(a)mathworks.com Subject: RE: Oracle on CIFS? Biggest issue with CIFS and database is stateful nature of the CIFS protocol and "robustness" (is that a word?) of the re-director on Windows. What this means is that if the re-director gets overwhelmed (can happen, but you may never see it) the CIFS connection will be lost then re-established, but the Oracle Service will crash. Usually not a big problem, but you do need to manually re-start it. Also, be aware that when using UNC pathnames or drive letter access to CIFS shares with Oracle (the former is the best way to go) the Oracle Services must be set to run as a domain user, not a local machine account. So this complicates the process of creating new databases since the service will be started, but must be stopped and modified to use a different log in - one that has domain access to the Filer shares and has the "Run as a service" priviledge. This usually means that you cannot use the Oracle database assistant to create the databases directly. You can use the database assistant to create the scripts to create the databases, but you must create the service, make the mods to the service, then create the database (using the scripts created by the database assistant). -tom -----Original Message----- From: Dirk Seidel [mailto:dirk@seidel-pb.de] Sent: Wednesday, February 25, 2004 6:42 PM To: toasters(a)mathworks.com Subject: Oracle on CIFS? Hi out there, has any one experiences with running oracle on a Windows Server on CIFS? You don't find many documents about it on the Netapps Site. All the whitepapers concerning Windows and Oracle are about SAN. What are the pros / cons and limitations of running Oracle on a CIFS Share? Thank You! Dirk

1 0

RE: Oracle on CIFS?
by Yarmas, Tom 26 Feb '04

26 Feb '04

Biggest issue with CIFS and database is stateful nature of the CIFS protocol and "robustness" (is that a word?) of the re-director on Windows. What this means is that if the re-director gets overwhelmed (can happen, but you may never see it) the CIFS connection will be lost then re-established, but the Oracle Service will crash. Usually not a big problem, but you do need to manually re-start it. Also, be aware that when using UNC pathnames or drive letter access to CIFS shares with Oracle (the former is the best way to go) the Oracle Services must be set to run as a domain user, not a local machine account. So this complicates the process of creating new databases since the service will be started, but must be stopped and modified to use a different log in - one that has domain access to the Filer shares and has the "Run as a service" priviledge. This usually means that you cannot use the Oracle database assistant to create the databases directly. You can use the database assistant to create the scripts to create the databases, but you must create the service, make the mods to the service, then create the database (using the scripts created by the database assistant). -tom -----Original Message----- From: Dirk Seidel [mailto:dirk@seidel-pb.de] Sent: Wednesday, February 25, 2004 6:42 PM To: toasters(a)mathworks.com Subject: Oracle on CIFS? Hi out there, has any one experiences with running oracle on a Windows Server on CIFS? You don't find many documents about it on the Netapps Site. All the whitepapers concerning Windows and Oracle are about SAN. What are the pros / cons and limitations of running Oracle on a CIFS Share? Thank You! Dirk

1 0

MX records.
by Jerry 26 Feb '04

26 Feb '04

doh. Bug 104202 already exists. I would tend to think it's more than a level 5 "suggestion", but as long as it's logged I suppose. Jerry __________________________________ Do you Yahoo!? Yahoo! Mail SpamGuard - Read only the mail you want. http://antispam.yahoo.com/tools

2 3

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 2004