toasters

toasters@lists.teaparty.net

13531 discussions

Filer hung up tight
by Maglinger, Paul 23 Aug '01

23 Aug '01

We have a 740 filer running 6.0.1R3. Last night I got a page that "nothing was working". I headed into work and found that: 1) I could not telnet into the filer. 2) I could not get in through console either. 3) The display on the front of the filer said 850 ops but the activity led wasn't flashing at all. I figured I had no choice, so I shut it down hard and brought it back up. Everything came up fine. The message file posted a time stamp at 11:00pm but nothing at midnight (this was 1:30am). The message file also said that the fiiler was down for 6057 secs. The crash folder had one file in it called "bounds" and just contained "-1 -1". Backups were being performed at the time using Veritas Backup Exec. The backup log said everything was going fine until it found a "corrupted file", then it started skipping files until the filer was rebooted and it started in backing up again. We're trying to figure out what went wrong but didn't have much to go on. Anyone have any thoughts where to look now (or if this happens again)? "No matter where you go, there you are." Paul Maglinger System Administrator Shoe Carnival, Inc.

3 2

RE: RE2 whoops: Tape SAN Help
by Collie, Michael 23 Aug '01

23 Aug '01

Hello Mike, Please note that NetApp does not currently support the Overland Data library in a FC SAN configuration. The configurations described in the attached URL http://www.netapp.com/osn/info/config.html#fcsan outlines how filers should be configured in our supported tape SAN environment. We require that the tape library login in as a NL Port. Thanks Michael. -----Original Message----- From: Mike Ball [mailto:MBall@datalink.com] Sent: Thursday, August 23, 2001 2:11 PM To: 'Ambrose_Earle(a)shamrockfoods.com'; toasters(a)mathworks.com Subject: RE2 whoops: Tape SAN Help Ambrose, I just finished ready the docs referenced earlier on the subject. It looks like the Netapps do support full fabric? I could have swore that I was told they logged in as arbitrated loop. Anyway, if they are full fabric you will not need Quickloop. I would still keep the zoning I suggested in case other devices are plugged into the switch. Thank you, Mike -----Original Message----- From: Ambrose_Earle(a)shamrockfoods.com [mailto:Ambrose_Earle@shamrockfoods.com] Sent: Tuesday, August 21, 2001 6:40 PM To: toasters(a)mathworks.com Subject: Tape SAN Help We are having enormous problems configuring our new tape libraries with our filers. We keep getting different answers from people and I hope that someone on this list has something similar in production and they would be willing to share their information. Here's what we hope to have running: 2 F840c's 1 Brocade 2400 2 internal fibre channel bridges inside an Overland Data library a few other servers on the switch NetVault with dynamic drive sharing option So, should everything be fabric, do we need to zone, any other ideas? Thanks for your time,

1 0

RE2 whoops: Tape SAN Help
by Mike Ball 23 Aug '01

23 Aug '01

Ambrose, I just finished ready the docs referenced earlier on the subject. It looks like the Netapps do support full fabric? I could have swore that I was told they logged in as arbitrated loop. Anyway, if they are full fabric you will not need Quickloop. I would still keep the zoning I suggested in case other devices are plugged into the switch. Thank you, Mike -----Original Message----- From: Ambrose_Earle(a)shamrockfoods.com [mailto:Ambrose_Earle@shamrockfoods.com] Sent: Tuesday, August 21, 2001 6:40 PM To: toasters(a)mathworks.com Subject: Tape SAN Help We are having enormous problems configuring our new tape libraries with our filers. We keep getting different answers from people and I hope that someone on this list has something similar in production and they would be willing to share their information. Here's what we hope to have running: 2 F840c's 1 Brocade 2400 2 internal fibre channel bridges inside an Overland Data library a few other servers on the switch NetVault with dynamic drive sharing option So, should everything be fabric, do we need to zone, any other ideas? Thanks for your time,

1 0

RE: Tape SAN Help
by Mike Ball 23 Aug '01

23 Aug '01

Ambrose, I have setup Tape San's before but not with Netapp. I may be able to help a bit here: 1) You will need to license the Brocade switch for Quickloop mode. You will need this because the Netapp logins into the fabric as a arbitrated loop. Set the ports the Netapps login into as Arbitrated loop mode. 2) How is the Overland Data logging into the fabric? Usually they will login to the switch as N_Port but check with the vendor. I believe Overland Data use the ATTO bridge. If they are N_Port they will login as full fabric. 3) I would zone each Netapps Fibre loop as a separate zone. For example, Zone 1 would consist of Netapp 1 FC port and both the Overland Data's FC ports. The second zone would consist of Netapp 2 and both the Overland Data's FC ports. Hope this helps, Mike -----Original Message----- From: Ambrose_Earle(a)shamrockfoods.com [mailto:Ambrose_Earle@shamrockfoods.com] Sent: Tuesday, August 21, 2001 6:40 PM To: toasters(a)mathworks.com Subject: Tape SAN Help We are having enormous problems configuring our new tape libraries with our filers. We keep getting different answers from people and I hope that someone on this list has something similar in production and they would be willing to share their information. Here's what we hope to have running: 2 F840c's 1 Brocade 2400 2 internal fibre channel bridges inside an Overland Data library a few other servers on the switch NetVault with dynamic drive sharing option So, should everything be fabric, do we need to zone, any other ideas? Thanks for your time,

1 0

RE: Solaris/Oracle/filer tuning question
by Bond, Andrew 22 Aug '01

22 Aug '01

Chris, I normally recommend setting max_threads to at least 24. Have you got gigabit patch 106764? I think this affects reads across the board, but it's worth checking. Regards, Andrew Bond Network Appliance UK Consulting Systems Engineer Waterview House, 1 Roundwood Avenue, tel +44 (0)20-8756-6722 Stockley Park, mobile +44 (0)7801-383566 Middlesex, UB11 1EJ fax +44 (0)20-8756-6701 http://www.netapp.com Get answers NOW! - NetApp On the Web - http://now.netapp.com > -----Original Message----- > From: Chris Lamb [mailto:skeezics@measurecast.com] > Sent: 21 August 2001 18:08 > To: toasters(a)mathworks.com > Subject: Solaris/Oracle/filer tuning question > > > > Howdy, all. > > This is more a Solaris question, but since there's a filer > involved and > y'all are so helpful, I'll start with you guys. :-) > > The setup: > Sun E4500, Solaris 7 HW11/99+patches, GigE > NetApp F760, 5.3.6R2, GigE > Oracle 8.1.6.x > Two volumes for Oracle, NFSv3/UDP, 32K rsize/wsize > > The problem(s): > I/O from any single Oracle process tops out around 3-4MB/sec; > > Even with 'disk_asynch_io = false' in init.ora set to false, we > are getting sporadic and seemingly nonsensical "resource > temporarily unavailable" errors (had a case open for this one). > > The question: > Are any of the Solaris kernel NFS tunables useful in increasing > that apparent limit? Specifically: > nfs:nfs3_max_threads > nfs:nfs3_async_clusters > nfs:nfs3_nra > > The DBA and our NetApp reps/support team have gone over our > config, and > things are generally stable. After turning off asynch_io we noticed a > slight performance hit, but thought that the annoying "resource > temporarily unavailable" errors had been solved. We had another one > yesterday, first time in almost two months. :-/ > > In diagnosing the performance sluggishness, we noticed that for a > non-partitioned full table scan a single Oracle "reader" process was > limited to about 2.5-4MB/sec, while a full scan of a partitioned table > (using 6 readers) was getting 18-22MB/sec. (Obvious workaround: > increase the number of readers and writers, and partition the > table with > the bottleneck...) > > In each case, Solaris iostat was claiming that the NFS mount was "100% > busy", although response times were generally in the sub-6ms > range and the > filer's sysstat showed less than 20% cpu usage (this is a > production box, > so I can't completely isolate the load for our testing). For > _reads_ I > just can't figure out why Solaris - or Oracle itself - are throttling > performance like this... I've clocked > 50MB/sec reads in > plain old NFS > tests (bonnie, cpio, etc), and I think it's generally clear when we've > saturated the PCI bus on the filer. This Oracle bottleneck > is a mystery, > but now I'm wondering if the two problems are related. > > The Solaris Tunable Parameters Reference Manual says of the > "nfs:nfs3_max_threads" knob: > "Controls the number of kernel threads that perform asynchronous > I/O for the NFS version 3 _client_. [emphasis mine] Since NFS is > based on RPC and RPC is inherently synchronous, > separate execution > contexts are required to perform NFS operations that are > asynchronous from the calling thread." > > The default pool of threads is set to 8. There's some indication that > this may be a _per filesystem_ setting, and is not global to all NFS > client activity on the machine. Even if that's the case, > then 8 threads > per mount point might still be fairly restrictive on an 8-cpu machine, > given the fat pipe and headroom still available on the filer. > I generally > shy away from mucking about with Solaris internals ("this isn't your > father's SunOS" :-) but if that's what it takes to boost > performance I'm > all for it. > > Has anyone running in a similar environment had to tweak any Solaris > kernel tunables? Are there other Oracle/NetApp tuning hints > not discussed > in the whitepapers on the NOW site? My DBA and I would be > most grateful > for any warnings/suggestions/hints. :-) I'll be happy to provide more > detail off-list, then summarize results if anyone else is interested. > > Thanks, > > -- Chris > > -- > Chris Lamb, Unix Guy > MeasureCast, Inc. > 503-241-1469 x247 > <skeezics(a)measurecast.com> >

2 1

RE: Netapp models?
by Traitel, Eyal 22 Aug '01

22 Aug '01

Currently it will end up with the same page. Both F740 and F740c use the same platform (alpha), so they are the same. It could be that in the future this will change, so I guess a good practice is to choose the right model in the first place in order not to get an incorrect version. Eyal. ----------------------------------------------------------------- eTraitel - I'm the new eBuzzword around !!! Filer Escalation Engineer CNA, MCSE, CSA, LCA, NetApp CA Network Appliance BV Hoofddorp, The Netherlands Office: +31 23 567 9685 Cellular: +31 6 5497 2568 Get answers NOW! - NetApp On the Web - http://now.netapp.com ----------------------------------------------------------------- -----Original Message----- From: Allen, Pat [mailto:pat@mbari.org] Sent: Wednesday, August 22, 2001 18:46 To: toasters(a)mathworks.com Subject: RE: Netapp models? OK... this begs another question.... On the NOW site, the page where you download software has a pulldown menu where you select the platform that you want. Is there any difference between DOT for a F740 and a F740c (for example)? Both pages list the same version of DOT. Does it really matter which platform I select to get my software? Thanks! -----Original Message----- From: Traitel, Eyal [mailto:eyal@netapp.com] Sent: Wednesday, August 22, 2001 8:58 AM To: 'Tim Longo'; toasters(a)mathworks.com Subject: RE: Netapp models? The ending "c" is naming convention to mean that this is a pair of F760 clustered together. Physical differenece is that is 2xF760 with cluster kit. Hope it answers the question, Eyal. ----------------------------------------------------------------- eTraitel - I'm the new eBuzzword around !!! Filer Escalation Engineer CNA, MCSE, CSA, LCA, NetApp CA Network Appliance BV Hoofddorp, The Netherlands Office: +31 23 567 9685 Cellular: +31 6 5497 2568 Get answers NOW! - NetApp On the Web - http://now.netapp.com ----------------------------------------------------------------- -----Original Message----- From: Tim Longo [mailto:tlongo@avaya.com] Sent: Wednesday, August 22, 2001 17:03 To: toasters(a)mathworks.com Subject: Netapp models? What's the difference between an F7xx and an F7xxc? How would I know which is which?

1 0

RE: Netapp models?
by Allen, Pat 22 Aug '01

22 Aug '01

OK... this begs another question.... On the NOW site, the page where you download software has a pulldown menu where you select the platform that you want. Is there any difference between DOT for a F740 and a F740c (for example)? Both pages list the same version of DOT. Does it really matter which platform I select to get my software? Thanks! -----Original Message----- From: Traitel, Eyal [mailto:eyal@netapp.com] Sent: Wednesday, August 22, 2001 8:58 AM To: 'Tim Longo'; toasters(a)mathworks.com Subject: RE: Netapp models? The ending "c" is naming convention to mean that this is a pair of F760 clustered together. Physical differenece is that is 2xF760 with cluster kit. Hope it answers the question, Eyal. ----------------------------------------------------------------- eTraitel - I'm the new eBuzzword around !!! Filer Escalation Engineer CNA, MCSE, CSA, LCA, NetApp CA Network Appliance BV Hoofddorp, The Netherlands Office: +31 23 567 9685 Cellular: +31 6 5497 2568 Get answers NOW! - NetApp On the Web - http://now.netapp.com ----------------------------------------------------------------- -----Original Message----- From: Tim Longo [mailto:tlongo@avaya.com] Sent: Wednesday, August 22, 2001 17:03 To: toasters(a)mathworks.com Subject: Netapp models? What's the difference between an F7xx and an F7xxc? How would I know which is which?

1 0

Netapp models?
by Tim Longo 22 Aug '01

22 Aug '01

What's the difference between an F7xx and an F7xxc? How would I know which is which?

3 3

RE: Netapp models?
by Traitel, Eyal 22 Aug '01

22 Aug '01

Be careful - Do not mix F760c with C760. F760c = clustered pair of F760 C760 = NetCache model name ----------------------------------------------------------------- eTraitel - I'm the new eBuzzword around !!! Filer Escalation Engineer CNA, MCSE, CSA, LCA, NetApp CA Network Appliance BV Hoofddorp, The Netherlands Office: +31 23 567 9685 Cellular: +31 6 5497 2568 Get answers NOW! - NetApp On the Web - http://now.netapp.com ----------------------------------------------------------------- -----Original Message----- From: Mike Horwath [mailto:drechsau@tiny.net] Sent: Wednesday, August 22, 2001 17:45 To: Tim Longo Cc: toasters(a)mathworks.com Subject: Re: Netapp models? On Wed, Aug 22, 2001 at 11:03:23AM -0400, Tim Longo wrote: > > > What's the difference between an F7xx and an F7xxc? How would I know > which is which? The 'c' model is the cache version used for proxy caching. -- Mike Horwath Admin & Manager @ VISI.com WORK: drechsau(a)visi.com IRC: Drechsau http://www.visi.com/ HOME: drechsau(a)geeks.org The only Minnesota ISP with public statistics: http://noc.visi.com/ Garbage In -- Gospel Out. - berkeley fortune(6)

1 0

RE: Netapp models?
by Cormier, Darleen 22 Aug '01

22 Aug '01

Hi Tim, Any models listed with a "c" implies a clustored system. -----Original Message----- From: Tim Longo [mailto:tlongo@avaya.com] Sent: Wednesday, August 22, 2001 8:03 AM To: toasters(a)mathworks.com Subject: Netapp models? What's the difference between an F7xx and an F7xxc? How would I know which is which?

2 1

RE: Netapp models?
by Traitel, Eyal 22 Aug '01

22 Aug '01

The ending "c" is naming convention to mean that this is a pair of F760 clustered together. Physical differenece is that is 2xF760 with cluster kit. Hope it answers the question, Eyal. ----------------------------------------------------------------- eTraitel - I'm the new eBuzzword around !!! Filer Escalation Engineer CNA, MCSE, CSA, LCA, NetApp CA Network Appliance BV Hoofddorp, The Netherlands Office: +31 23 567 9685 Cellular: +31 6 5497 2568 Get answers NOW! - NetApp On the Web - http://now.netapp.com ----------------------------------------------------------------- -----Original Message----- From: Tim Longo [mailto:tlongo@avaya.com] Sent: Wednesday, August 22, 2001 17:03 To: toasters(a)mathworks.com Subject: Netapp models? What's the difference between an F7xx and an F7xxc? How would I know which is which?

1 0

[Fwd: Snapshot Cheats?]
by Alfred Lerch 22 Aug '01

22 Aug '01

Hi folks, I have a somewhat similar issue; to reduce the size of snapshots I'd love to remove some files from the snapshots but I don't want to delete the whole snapshot - is there any way to do so ? regards alfred -------- Original Message -------- Subject: Snapshot Cheats? Date: Thu, 16 Aug 2001 10:19:21 +0200 From: "Kennard, Dave M SITI-ISEL-31" <Dave.M.Kennard(a)is.shell.com> To: "'toasters(a)mathworks.com'" <toasters(a)mathworks.com> List, Is there any way to cheat with snapshots so that data in them can be deleted? I have stupidly let a volume on my filer grow to capacity and want to free up space my moving files to another volume, if I just move the files I will get no gain in free disk space because the files will exist in the snapshot right? Could I cheat this by taking a snapshot of the volume before I delete the files then delete the snapshot after the file are deleted? Regards Dave Kennard

1 0

Write timeouts to CIFS share, oplock problem?
by Brian Tao 22 Aug '01

22 Aug '01

Is this type of problem (Windows applications hanging when trying to write out files, or the file being saved is mysteriously deleted from the filesystem) indicative of oplock issues between a Windows client and a filer? I have a few things to look at on the Netapp side, but I want to make sure there isn't some other easy explanation for this. Client is running W2K from a laptop on a 100t port. Filer is on the same switched network also on a 100t port. Filer is an F740 running 5.3.6R1. Oplocks are disabled on the particular filesystems the user is writing to. One filesystem has NTFS-style security and the other uses mixed mode security. The network itself and the filer interface are not congested or seeing packet errors. Do people generally leave oplocks enabled on CIFS shares? Is this feature enabled on NT-based file servers? Reading http://now.netapp.com/NOW/knowledge/docs/ontap/rel536r2/html/sag/cifs25.htm gives me the distinct feeling that oplocks == write-caching with no error checking, which sounds like trouble waiting to happen. First report: > Problem: Excel spreadsheet located in (****** directory) is taking > over 20 min to save. > > (email notes below) > I have a single worksheet Excel spreadsheet stored on CORP-NA1 > (****** directory) that takes over twenty minutes to save. During > that time, my Excel application and my Windows Explorer (which is > open to that directory) appear to be 'hung'. Given enough time, > both applications come back and the entries in the spreadsheet > appear to be saved. > > I am listing this as over 20 minutes, because I have not waited > around long enough to find out how long it actually takes. After 20 > minutes, I've gone off to meetings, and an hour and a half later, it > is working again. Second report (same user): > Sorry for the delay. > > This problem is ongoing. In fact, minutes ago, a new twist > appeared. As Alex was exiting this shared spreadsheet, I was > opening it. It was opened, I accepted a change that had been made > earlier by someone else, and went to save it. At that point, after > a long (three to five minute) wait, it gave a general error message, > and the data file was deleted from CORP-NA1. I was doing this from > a desktop PC on the LAN (not using the VPN). At the same time, > Sandi was online and was having a long delay in opening the > sub-directories under ******, and also saw the data file was > deleted. > > I re-saved the data file, and re-established the shared workbook > privileges. > > I'm still guessing that some sort of NetBIOS timeout is occurring. -- Brian Tao (BT300, taob(a)risc.org) "Though this be madness, yet there is method in't"

2 1

RE: Tape SAN Help
by Collie, Michael 21 Aug '01

21 Aug '01

Hello Earle, Firstly, attached is the URL http://www.netapp.com/osn/info/config.html#fcsan that captures NetApp certified tape SAN configurations. Regarding the configuration you outlined below, the 3rd Party Engineering group (responsible for certifying tape SAN solutions) has no visibility to the Overland Data Library. This library is not NetApp certified. Further, we do not support having any other (non NetApp) device connected to the same FC switch as filers. We believe that if non-NetApp devices must be connected to the same FC switch as with filers, those devices must then be configured in a separate zone from the filers. We are in the process of beginning filer/none filer interoperable switch tests using zoning. The bottom line is, we are not full fabric compliant at this point. The library should be configured as NL-Port before the filer can see the tape drives. You did not mentioned which bridge is in the library. However, we currently support Pathlight and Crossroads bridges in STK L180, L700 and ADIC Scalar 1000 respectively. In summary, the configuration you mentioned is not a supported tape SAN configuration, but hope that the additional information I provided will help you to understand how our supported configuration works. We expect to start talks with Overland very soon on certifying their libraries. Michael. -----Original Message----- From: Ambrose_Earle(a)shamrockfoods.com [mailto:Ambrose_Earle@shamrockfoods.com] Sent: Tuesday, August 21, 2001 3:40 PM To: toasters(a)mathworks.com Subject: Tape SAN Help We are having enormous problems configuring our new tape libraries with our filers. We keep getting different answers from people and I hope that someone on this list has something similar in production and they would be willing to share their information. Here's what we hope to have running: 2 F840c's 1 Brocade 2400 2 internal fibre channel bridges inside an Overland Data library a few other servers on the switch NetVault with dynamic drive sharing option So, should everything be fabric, do we need to zone, any other ideas? Thanks for your time,

1 0

RE: Tape SAN Help
by Parisse, Dave 21 Aug '01

21 Aug '01

Ambrose, Network Appliance has multiple Fibre Channel Solutions for TapeSAN that I urge you to look at. These can be found in our Data Protection Portal at http://www.netapp.com/solutions/data_protection.html and click on the FC SAN on the left. Unfortunately, we do not have an Overland Data Library in our current tapeSAN solutions, but I do want to suggest you make a request through your Netapp Sales Representative specifying library model, etc. so the feedback can be incorporated into future solutions. Our TapeSAN solution certification processes allow us to work with the partners to eliminate any 'enormous problems' you the customer would have deploying those certified tapeSAN configurations. Dave parisse(a)netapp.com Third Party Engineering Manager Network Appliance, Inc. -----Original Message----- From: Ambrose_Earle(a)shamrockfoods.com [mailto:Ambrose_Earle@shamrockfoods.com] Sent: Tuesday, August 21, 2001 3:40 PM To: toasters(a)mathworks.com Subject: Tape SAN Help We are having enormous problems configuring our new tape libraries with our filers. We keep getting different answers from people and I hope that someone on this list has something similar in production and they would be willing to share their information. Here's what we hope to have running: 2 F840c's 1 Brocade 2400 2 internal fibre channel bridges inside an Overland Data library a few other servers on the switch NetVault with dynamic drive sharing option So, should everything be fabric, do we need to zone, any other ideas? Thanks for your time,

1 0

RE: busy snapshot
by Waters, G Scott DSTI 21 Aug '01

21 Aug '01

we had the same thing happen recently. The only way to clear this is to bounce the filer and then delete the snapshot. There is a bug report on the NOW site for this. - Scott W. -----Original Message----- From: Leonard, Roger [mailto:Roger.Leonard@marconi.com] Sent: Tuesday, August 21, 2001 5:09 PM To: 'toasters(a)mathworks.com' Subject: busy snapshot my veritas server crashed during an ndmp backup of my 740 and there is a snapshot_for_backup that is busy and will not let me delete it. it says it is in use by dump or cifs share. any idea how to clear this out? Roger D. Leonard

2 1

Tape SAN Help
by Ambrose_Earle＠shamrockfoods.com 21 Aug '01

21 Aug '01

We are having enormous problems configuring our new tape libraries with our filers. We keep getting different answers from people and I hope that someone on this list has something similar in production and they would be willing to share their information. Here's what we hope to have running: 2 F840c's 1 Brocade 2400 2 internal fibre channel bridges inside an Overland Data library a few other servers on the switch NetVault with dynamic drive sharing option So, should everything be fabric, do we need to zone, any other ideas? Thanks for your time,

1 0

Opera denied !
by Farid Hamjavar 21 Aug '01

21 Aug '01

Greetings, Just checking .... Is it me or now.netapp.com denies access to those of us using Opera web browser? I use Opera 5.12 (the windows version) and all cookies stuff and all SSL stuff (V2 and V3) are turned on and enabled. here are error messages: http://now.netapp.com/nowsmforms/login.fcc The certificate's chain was not ordered properly. Thanks, Farid

2 1

busy snapshot
by Leonard, Roger 21 Aug '01

21 Aug '01

my veritas server crashed during an ndmp backup of my 740 and there is a snapshot_for_backup that is busy and will not let me delete it. it says it is in use by dump or cifs share. any idea how to clear this out? Roger D. Leonard

1 0

RE: virus protection on Filer
by Ziller, Frank 21 Aug '01

21 Aug '01

Take a look at Symantec CarrierScan. -----Original Message----- From: ken pigg [mailto:kenp@uptime.net] Sent: Tuesday, August 21, 2001 4:40 PM To: 'Edward Lee'; toasters(a)mathworks.com Subject: RE: virus protection on Filer http://www.netapp.com/partners/catalog.cgi/company/178 kp -----Original Message----- From: Edward Lee [mailto:edwardlee60@hotmail.com] Sent: Tuesday, August 21, 2001 1:33 PM To: toasters(a)mathworks.com Subject: virus protection on Filer Can anyone recommend a good antivirus program that can do real-time scans of the Filer. I have a 760 and running version 6.1 Thanks, -Eddie _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

1 0

RE: virus protection on Filer
by Benn, Paul 21 Aug '01

21 Aug '01

You can get a solution specifically for the filer from the following vendors: McAfee Symantec Trend Micro Regards, Paul Benn Network Appliance -----Original Message----- From: Edward Lee [mailto:edwardlee60@hotmail.com] Sent: Tuesday, August 21, 2001 11:33 AM To: toasters(a)mathworks.com Subject: virus protection on Filer Can anyone recommend a good antivirus program that can do real-time scans of the Filer. I have a 760 and running version 6.1 Thanks, -Eddie _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

1 0

RE: virus protection on Filer
by ken pigg 21 Aug '01

21 Aug '01

http://www.netapp.com/partners/catalog.cgi/company/178 kp -----Original Message----- From: Edward Lee [mailto:edwardlee60@hotmail.com] Sent: Tuesday, August 21, 2001 1:33 PM To: toasters(a)mathworks.com Subject: virus protection on Filer Can anyone recommend a good antivirus program that can do real-time scans of the Filer. I have a 760 and running version 6.1 Thanks, -Eddie _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

1 0

virus protection on Filer
by Edward Lee 21 Aug '01

21 Aug '01

Can anyone recommend a good antivirus program that can do real-time scans of the Filer. I have a 760 and running version 6.1 Thanks, -Eddie _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

1 0

Solaris/Oracle/filer tuning question
by Chris Lamb 21 Aug '01

21 Aug '01

Howdy, all. This is more a Solaris question, but since there's a filer involved and y'all are so helpful, I'll start with you guys. :-) The setup: Sun E4500, Solaris 7 HW11/99+patches, GigE NetApp F760, 5.3.6R2, GigE Oracle 8.1.6.x Two volumes for Oracle, NFSv3/UDP, 32K rsize/wsize The problem(s): I/O from any single Oracle process tops out around 3-4MB/sec; Even with 'disk_asynch_io = false' in init.ora set to false, we are getting sporadic and seemingly nonsensical "resource temporarily unavailable" errors (had a case open for this one). The question: Are any of the Solaris kernel NFS tunables useful in increasing that apparent limit? Specifically: nfs:nfs3_max_threads nfs:nfs3_async_clusters nfs:nfs3_nra The DBA and our NetApp reps/support team have gone over our config, and things are generally stable. After turning off asynch_io we noticed a slight performance hit, but thought that the annoying "resource temporarily unavailable" errors had been solved. We had another one yesterday, first time in almost two months. :-/ In diagnosing the performance sluggishness, we noticed that for a non-partitioned full table scan a single Oracle "reader" process was limited to about 2.5-4MB/sec, while a full scan of a partitioned table (using 6 readers) was getting 18-22MB/sec. (Obvious workaround: increase the number of readers and writers, and partition the table with the bottleneck...) In each case, Solaris iostat was claiming that the NFS mount was "100% busy", although response times were generally in the sub-6ms range and the filer's sysstat showed less than 20% cpu usage (this is a production box, so I can't completely isolate the load for our testing). For _reads_ I just can't figure out why Solaris - or Oracle itself - are throttling performance like this... I've clocked > 50MB/sec reads in plain old NFS tests (bonnie, cpio, etc), and I think it's generally clear when we've saturated the PCI bus on the filer. This Oracle bottleneck is a mystery, but now I'm wondering if the two problems are related. The Solaris Tunable Parameters Reference Manual says of the "nfs:nfs3_max_threads" knob: "Controls the number of kernel threads that perform asynchronous I/O for the NFS version 3 _client_. [emphasis mine] Since NFS is based on RPC and RPC is inherently synchronous, separate execution contexts are required to perform NFS operations that are asynchronous from the calling thread." The default pool of threads is set to 8. There's some indication that this may be a _per filesystem_ setting, and is not global to all NFS client activity on the machine. Even if that's the case, then 8 threads per mount point might still be fairly restrictive on an 8-cpu machine, given the fat pipe and headroom still available on the filer. I generally shy away from mucking about with Solaris internals ("this isn't your father's SunOS" :-) but if that's what it takes to boost performance I'm all for it. Has anyone running in a similar environment had to tweak any Solaris kernel tunables? Are there other Oracle/NetApp tuning hints not discussed in the whitepapers on the NOW site? My DBA and I would be most grateful for any warnings/suggestions/hints. :-) I'll be happy to provide more detail off-list, then summarize results if anyone else is interested. Thanks, -- Chris -- Chris Lamb, Unix Guy MeasureCast, Inc. 503-241-1469 x247 <skeezics(a)measurecast.com>

1 0

RE: DOT 6.1.1 + NBU 4.6 == /vol/vol0 path may not exist
by Smith, A Michael (CS) 20 Aug '01

20 Aug '01

Is /vol/vol0 on your machine a snapmirror destination? If so try backing up an existing snapshot as ndmpd will not be able to create a snapshot for dump on a snapmirrored volume. The path that I use in the Backup Policy Mgmt interface under the policy of my choice I simply change the entry from "/vol/vol2" to "/vol/vol2/.snapshot/hourly.0" and bada-bing it worked. If it is not snapmirrored then another test to try would be to run a manual dump of /vol/vol0 and send the output to null: dump 0f null /vol/vol0 The output of that command will be useful for NetApp support folks. Mike Smith -----Original Message----- From: Jeffrey Krueger [mailto:jkrueger@qualcomm.com] Sent: Monday, August 20, 2001 10:41 AM To: toasters(a)mathworks.com Subject: DOT 6.1.1 + NBU 4.6 == /vol/vol0 path may not exist Anyone else having difficulty backing up /vol/vol0 with Veritas after upgrading to DOT 6.1.1? We get the following error message: 0997254443 1 4 8 claxton 750651 750651 0 fuelcell bptm WRN - Cannot backup filepath /vol/vol0, NDMP_HALT_INTERNAL_ERROR occurred (path may not exist), skipping We also found this upon deeper investigation: Aug 10 21:02:15 GMT [ndmpd:52]: ReplySequence 0 Aug 10 21:02:15 GMT [ndmpd:52]: Error NDMP_NO_ERR Aug 10 21:02:15 GMT [ndmpd:52]: Log message: Cannot backup WAFL directory -- memory management problem. Here's the interesting rub: other volumes on that machine (/vol/root and /vol/vol1) backup just fine. /vol/vol0 won't backup to a local tape drive, or remote via NDMP 3-way. This has manifested itself on more than one machine, but not all 6.1.1 filers. Really strange. We're submitting a case to NetApp, but just wondering if we're the only ones seeing this problem. -- Jeff -- ---------------------------------------------------------------------------- Jeff Krueger, NetApp CA E-Mail: jeff(a)qualcomm.com Senior Engineer Phone: 858-651-6709 Storage Lead (NAS/SAN) Fax: 858-651-6627 QUALCOMM, Inc. IT Engineering Web: www.qualcomm.com

2 1

Jump to page:

2026

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