toasters

toasters@lists.teaparty.net

13531 discussions

Re: Backup solution.
by Graydon Dodson 16 Oct '00

16 Oct '00

A word of advice on the Spectra Logic Gator. We have had one for about two months now and it is not yet a mature product. (but it is making progress) In that time I have had to upgrade S/W in the unit at least four times. One upgrade (done at the recommendation of Spectra Logic) turned out to not be compatible with the EC level of my H/W requiring a couple of RMAs. Today it is still possible to confuse the unit if care is not taken. (what else can you expect from a product running Windoes CE?) Having said the above. I like the idea behind the gator and believe that in time Spectra Logic will get the kinks worked out. The ability to logically partition the unit as well as the future possibility of backing a netapp over FC-AL makes it very appealing. But you might want to give the unit a little time to mature first. In my experience the fastest way to back up a filler today is direct attach SCSI. FC-AL is not supported till OnTap 6.x which is still only early access. (I will not run early access code on production fillers) Gig ethernet may look appealing but I cannot see how placing an ethernet layer between the NetApp and the tape drive is going to speed anything up. There are two fundamental bottlenecks for backups, the speed of the tape drive and the load on the filler. Adding more stuff in the middle cannot speed things up. While we are on the subject, I have an upgrade the "ndmputil" package that supports large tape libraries and includes a tool to retrieve drive & tape stats from AIT tape drives. I hope to get the upgrade to NetApp for posting on NOW as soon as I get the documentation updated. Graydon Dodson grdodson(a)lexmark.com Lexmark International Inc. ------------- Begin Forwarded Message ------------- From: Jeff Bryer <bryer%sfu.ca(a)interlock.lexmark.com> Subject: Re: Backup solution. To: foo%eek.org(a)interlock.lexmark.com (foo) Date: Mon, 16 Oct 2000 14:28:27 -0700 (PDT) Cc: toasters%mathworks.com(a)interlock.lexmark.com > The questions are: > > 1.) Which tape library to buy (leaning towards SpectraLogic Bullfrog with 4 drives/40 tapes). > We were looking at the Gator 12000 from SpectraLogic ourselves. Then decided on the 64000 for it's ability to scale up. We haven't got it hooked up yet, so I can't give you a performance report just yet. > 2.) To connect the library to the network via gig-e or to the netapp's via FCAL. > We're going with direct attach SCSI to a pair of F740s and a pair of SUN E450s. May direct attach to a 3rd F740 (if we buy a third). In a year or so, I may want to switch over to a SAN environment (swap out our S-QIPS with F-QIPS in the Gator), but I'm not comfortable enough with technology currently to consider that for the immediate future. > 3.) Which software to buy and what to make it do (eh NDMP blah). > We're evaluating Time Navigator (Quadratec), Backup Express (SyncSort), and Quick Restore (Workstation Solutions). Legato's NDMP support currently is not worth considering, and Veritas hasn't been forthcoming in working with us. We're looking to handle 1 - 1 1/2TBs in the short term, and scale up if we start backing up servers across campus. -- Jeff Bryer bryer(a)sfu.ca Systems Administrator (604) 291-4935 Academic Computing, Simon Fraser University ------------- End Forwarded Message -------------

1 0

Re: Backup solution.
by mehlel001＠hawaii.rr.com 16 Oct '00

16 Oct '00

How about backing up to another filer? You could use "Snap Mirror". The main reson I suggest this is because to restore that much data off of tape would take days! We have over 40 filers are are looking to backup directly to a "Backup Filer" using Snap Mirror and MAYBE have a tape robot off of the "backup filer" that would go to tape maybe once a month and be used as a DR device ofr offsite storeage etc. Anybody do anything simular to this? Thanks in advance, Lance >Ok, so I've been trying to figure out the "optimal" backup solution for my environment for awhile and haven't had much luck. I >know this topic gets beaten to death, but if you can indulge me for a moment that'd be great. > >I have 2 760's with a 1/2 TB each (in a cluster). We'll be adding a 720 with ~1/2 TB very soon, and expect our overall storage >to at least double in the next 6-12 months. We have approximately 25 servers (sparcs of various flavors) using the storage. >We'd like to have a unified solution for backing up the netapp's as well as all of the servers. > >The questions are: > >1.) Which tape library to buy (leaning towards SpectraLogic Bullfrog with 4 drives/40 tapes). > >2.) To connect the library to the network via gig-e or to the netapp's via FCAL. > >3.) Which software to buy and what to make it do (eh NDMP blah). > >Did I miss anything? > >So if you had this environment today what kind of solution would your propose? > >-Brian >

1 0

RE: NIS groups not showing up
by mehlel001＠hawaii.rr.com 16 Oct '00

16 Oct '00

Thanks to all for your help. As it turns out the problem is if a group line on your NIS+ server exceeds 1024 chars the filer will stop processing the NIS entry once it sees that line. So all groups after that are non-existant! I understand this is fixed in 6.0 but don't know for sure. If not there should atleast be a warning in the log file that says you may have exceeded the 1024 char limit in a particular NIS map. >Hi Lance, > > When the group cache is created, if there are duplicate entries, a message will >be logged. Are there any such messages logged? > >Thanks, >-Devi > > > >-----Original Message----- >From: Lance R. Mehle [mailto:mehlel001@hawaii.rr.com] >Sent: Saturday, October 14, 2000 3:13 PM >To: Toasters (E-mail) >Subject: NIS groups not showing up > > >Hi all, > >I have a problem that I was wondering if any of you have seen. > >We have a little over 2000 groups in our NIS+ environment. When I check the >nis_cache file on the filer about 399 are missing. A ypgroup also shows the >group(s) missing from the filer although a ypmatch can find the group(s). >It also appears that the same groups are effected all the time. > >The groups are fine on the UNIX side. > >Help! > >Lance >

1 0

Backup solution.
by foo 16 Oct '00

16 Oct '00

Ok, so I've been trying to figure out the "optimal" backup solution for my environment for awhile and haven't had much luck. I know this topic gets beaten to death, but if you can indulge me for a moment that'd be great. I have 2 760's with a 1/2 TB each (in a cluster). We'll be adding a 720 with ~1/2 TB very soon, and expect our overall storage to at least double in the next 6-12 months. We have approximately 25 servers (sparcs of various flavors) using the storage. We'd like to have a unified solution for backing up the netapp's as well as all of the servers. The questions are: 1.) Which tape library to buy (leaning towards SpectraLogic Bullfrog with 4 drives/40 tapes). 2.) To connect the library to the network via gig-e or to the netapp's via FCAL. 3.) Which software to buy and what to make it do (eh NDMP blah). Did I miss anything? So if you had this environment today what kind of solution would your propose? -Brian

2 1

RE: cifs.home_dir values
by Louis Brune 16 Oct '00

16 Oct '00

If I may submit a summary: From paul.benn(a)netapp.com: options cifs.home_dir vol\vol0\homes,vol\vol1\more_homes Data ONTAP 6.0 has another new option as well called "cifs.home_dir_namestyle." This option allows you to change how a username is used for mapping the home directory. Here are the three options with descriptions: ntname (or nothing): This is the same as with ONTAP 5.x. The user's home directory name must match the user's NT domain account name. mapped: The user's home directory name must match the user's mapped UNIX user name. domain: Under the root home directory (set with the cifs.home_dir option), you need to create one subdirectory for each domain that the filer trusts. Each user's directory is then created under their respective domain directory. The user's home directory name must match the user's NT domain account name and the domain directory must match the user's domain. and from Anissa.Mohler(a)netapp.com: Actually the syntax will be: options cifs.home_dir /vol/vol0/homes,/vol/vol1/more_homes If you type it wrong you will get a friendly error message :) There are many different ways of using home directories in ONTAP 6.x like Paul mentioned. It is one of the cooler new features from a CIFS point of view. Meanwhile, back in 5.3.x-land, there was from Eli.Lopez(a)netapp.com, jkrueger(a)qualcomm.com, and Moshel.Linzer(a)nsc.com: > > > On Wed, Sep 20, 2000 at 09:23:04AM +0200, Eli Lopez wrote: > > > > BTW, you can create a soft link on the filer called QTREE1 pointing to > > > > /vol/vol1/QTREE1 and mount using the link name. > > > > > > Cool! I didn't know the NetApp could internally resolve symlinks for > > > NFS/CIFS exports/shares and still give us the "right stuff" from rquotad > > > and what not. > > > > Yeah, the same trick allows us to share our CIFS homedirs under a single root, > > without actually moving the user's home directory. > > Whoa. What's this? I knew symlinks if resolved on the same > filesystem on a filer work correctly under CIFS, but this seems > different. You can symlink across *volumes* on the same filer and it is > transparent to NFS and CIFS clients??? > Yes, the difference is that you are SHARING THE SYMLINK, not putting a symlink inside a share. So without quoting all of Chris' email, you can set your cifs homedir to a directory full of symlinks (userid -> /path/to/homedir), and each user will get a share to his own homedir upon login! Cool undocumented feature. And it appears to be true that some 6.X release will have multiple homedir paths, but this is still more versatile! So we ask: Will support for that most excellent symlink trick persist into ONTAP >=6? -- Moderation in all things, except love. Louis

1 0

Re: FILERS IN A HIGH-TRANSACTION SYBASE APP ENVIRONMENT
by Rick Hulsey 16 Oct '00

16 Oct '00

We did some performance testing on HP K class boxes running HP-UX 10.20 and 11.0. We found that the Netapp over Gig-E was on average about 30% faster than the HP Mod-20 disk arrays (AKA CLARiiON 20 slot) configured in RAID 5 for our application queries. We are planning to wait for more testing on Sybase 12.5 to make the move to the filer because the move to 12.5 will be a dump and load of our existing databases anyway. Rick Hulsey Southwest Airlines 2425 Wyman Mail Stop 4DC Dallas, Tx., 75235 rick.hulsey(a)wnco.com (214) 792-7188 Office (972) 880-6882 Cell (800) 915-3747 Pager >>> Joe Luchtenberg <joe.luchtenberg(a)data-line.com> 10/13/00 03:40PM >>> Is anybody using NetApp filers for backend storage in a high-transaction Sybase app environment? I have a customer considering this configuration and would appreciate any insight/lessons learned from the field. Thanks. Joe Joe Luchtenberg Dataline, Inc. New number and email, please update: 757.858.0600 858.0606 fax joe.luchtenberg(a)data-line.com

1 0

Top 5 Web Sites
by mmcdonald＠123go.com 16 Oct '00

16 Oct '00

1 0

RE: cifs.home_dir values
by Muhlestein, Mark 15 Oct '00

15 Oct '00

And if you aren't ready to upgrade to 6.0, you can still create symbolic links which point across volumes. Mark -----Original Message----- From: Mohler, Anissa [mailto:Anissa.Mohler@netapp.com] Sent: Saturday, October 14, 2000 9:56 AM To: Benn, Paul; 'Coder, James (STP)'; 'toasters(a)mathworks.com' Subject: RE: cifs.home_dir values Actually the syntax will be: options cifs.home_dir /vol/vol0/homes,/vol/vol1/more_homes If you type it wrong you will get a friendly error message :) There are many different ways of using home directories in ONTAP 6.x like Paul mentioned. It is one of the cooler new features from a CIFS point of view. They are covered extensively in the ONTAP 6.0 Web Based Training (shameless plug) ... or you can always read the System Administrators Guide for ONTAP 6 on the NOW site ;) Enjoy! :-----Original Message----- :From: Benn, Paul [mailto:paul.benn@netapp.com] :Sent: Friday, October 13, 2000 10:48 PM :To: 'Coder, James (STP)'; 'toasters(a)mathworks.com' :Subject: RE: cifs.home_dir values : : :Hello James, : :If you're using Data ONTAP 6.0 or later you can specify :multiple directories on :one filer. The option you need to set is called :"cifs.home_dir" and you can use :a comma separated list with a maximum of 255 characters. :Here's a sample :command: : :options cifs.home_dir vol\vol0\homes,vol\vol1\more_homes : :Data ONTAP 6.0 has another new option as well called :"cifs.home_dir_namestyle." :This option allows you to change how a username is used for :mapping the home :directory. Here are the three options with descriptions: : :ntname (or nothing): :This is the same as with ONTAP 5.x. The user's home directory :name must match :the user's NT domain account name. : :mapped: :The user's home directory name must match the user's mapped :UNIX user name. : :domain: :Under the root home directory (set with the cifs.home_dir :option), you need to :create one subdirectory for each domain that the filer trusts. :Each user's :directory is then created under their respective domain :directory. The user's :home directory name must match the user's NT domain account :name and the domain :directory must match the user's domain. : : :Best Regards, : :Paul Benn :Network Appliance : : :-----Original Message----- :From: Coder, James (STP) [mailto:james.coder@guidant.com] :Sent: Friday, October 13, 2000 7:53 PM :To: 'toasters(a)mathworks.com' :Subject: cifs.home_dir values : : :I have a 760 being used for home file server in Solaris / windows :environment. This is replacing my e3500 filer server running :samba. Can :you have more than one volume listed in for cifs.home_dir? If :so, how do :you set it up? : :thx in advance - jcc : : :James C. Coder :UNIX Administrator :Guidant Corporation :Phone: 651-582-4797 :Email: james.coder(a)guidant.com : :

1 0

NIS groups not showing up
by Lance R. Mehle 14 Oct '00

14 Oct '00

Hi all, I have a problem that I was wondering if any of you have seen. We have a little over 2000 groups in our NIS+ environment. When I check the nis_cache file on the filer about 399 are missing. A ypgroup also shows the group(s) missing from the filer although a ypmatch can find the group(s). It also appears that the same groups are effected all the time. The groups are fine on the UNIX side. Help! Lance

1 0

RE: cifs.home_dir values
by Mohler, Anissa 14 Oct '00

14 Oct '00

Actually the syntax will be: options cifs.home_dir /vol/vol0/homes,/vol/vol1/more_homes If you type it wrong you will get a friendly error message :) There are many different ways of using home directories in ONTAP 6.x like Paul mentioned. It is one of the cooler new features from a CIFS point of view. They are covered extensively in the ONTAP 6.0 Web Based Training (shameless plug) ... or you can always read the System Administrators Guide for ONTAP 6 on the NOW site ;) Enjoy! :-----Original Message----- :From: Benn, Paul [mailto:paul.benn@netapp.com] :Sent: Friday, October 13, 2000 10:48 PM :To: 'Coder, James (STP)'; 'toasters(a)mathworks.com' :Subject: RE: cifs.home_dir values : : :Hello James, : :If you're using Data ONTAP 6.0 or later you can specify :multiple directories on :one filer. The option you need to set is called :"cifs.home_dir" and you can use :a comma separated list with a maximum of 255 characters. :Here's a sample :command: : :options cifs.home_dir vol\vol0\homes,vol\vol1\more_homes : :Data ONTAP 6.0 has another new option as well called :"cifs.home_dir_namestyle." :This option allows you to change how a username is used for :mapping the home :directory. Here are the three options with descriptions: : :ntname (or nothing): :This is the same as with ONTAP 5.x. The user's home directory :name must match :the user's NT domain account name. : :mapped: :The user's home directory name must match the user's mapped :UNIX user name. : :domain: :Under the root home directory (set with the cifs.home_dir :option), you need to :create one subdirectory for each domain that the filer trusts. :Each user's :directory is then created under their respective domain :directory. The user's :home directory name must match the user's NT domain account :name and the domain :directory must match the user's domain. : : :Best Regards, : :Paul Benn :Network Appliance : : :-----Original Message----- :From: Coder, James (STP) [mailto:james.coder@guidant.com] :Sent: Friday, October 13, 2000 7:53 PM :To: 'toasters(a)mathworks.com' :Subject: cifs.home_dir values : : :I have a 760 being used for home file server in Solaris / windows :environment. This is replacing my e3500 filer server running :samba. Can :you have more than one volume listed in for cifs.home_dir? If :so, how do :you set it up? : :thx in advance - jcc : : :James C. Coder :UNIX Administrator :Guidant Corporation :Phone: 651-582-4797 :Email: james.coder(a)guidant.com : :

1 0

RE: cifs.home_dir values
by Benn, Paul 14 Oct '00

14 Oct '00

Hello James, If you're using Data ONTAP 6.0 or later you can specify multiple directories on one filer. The option you need to set is called "cifs.home_dir" and you can use a comma separated list with a maximum of 255 characters. Here's a sample command: options cifs.home_dir vol\vol0\homes,vol\vol1\more_homes Data ONTAP 6.0 has another new option as well called "cifs.home_dir_namestyle." This option allows you to change how a username is used for mapping the home directory. Here are the three options with descriptions: ntname (or nothing): This is the same as with ONTAP 5.x. The user's home directory name must match the user's NT domain account name. mapped: The user's home directory name must match the user's mapped UNIX user name. domain: Under the root home directory (set with the cifs.home_dir option), you need to create one subdirectory for each domain that the filer trusts. Each user's directory is then created under their respective domain directory. The user's home directory name must match the user's NT domain account name and the domain directory must match the user's domain. Best Regards, Paul Benn Network Appliance -----Original Message----- From: Coder, James (STP) [mailto:james.coder@guidant.com] Sent: Friday, October 13, 2000 7:53 PM To: 'toasters(a)mathworks.com' Subject: cifs.home_dir values I have a 760 being used for home file server in Solaris / windows environment. This is replacing my e3500 filer server running samba. Can you have more than one volume listed in for cifs.home_dir? If so, how do you set it up? thx in advance - jcc James C. Coder UNIX Administrator Guidant Corporation Phone: 651-582-4797 Email: james.coder(a)guidant.com

1 0

RE: Trivia: what do the splodges on the LCD represent?
by Mohler, Jeff 14 Oct '00

14 Oct '00

The scale is "realative" to recent readings. If the last reading or so was 12, and the current one is 3000, it'll peg. If the last reading or so was 3000, and the current one is 12, it'll sag. It shifts..its part of the MEI. (M)anagement (E)ntertainment (I)nterface. -----Original Message----- From: Frank Smith [mailto:frank@artair.com] Sent: Friday, October 13, 2000 9:51 PM To: Guy Harris Cc: toasters(a)mathworks.com Subject: Re: Trivia: what do the splodges on the LCD represent? --On Friday, October 13, 2000 7:04 PM -0700 Guy Harris <guy(a)netapp.com> wrote: >> And the 10 splodges isn't related to ops at all; it's the CPU usage, >> 0 to 100%. > > Nope, it's just a bar graph of the number displayed to the left of the > bar graph. > And what is the scale? Ten splodges equals how many ops? It looks to me more like some sort of autoscaling based on a rolling average, as I have seen the bar graph shoot off to the right during slack times when the number is relatively low compared to its rated max ops/sec. Frank -- Frank Smith fsmith(a)hoovers.com Systems Administrator Voice: 512-374-4673 Hoover's Online Fax: 512-374-4501

1 0

RE: Trivia: what do the splodges on the LCD represent?
by Mohler, Jeff 14 Oct '00

14 Oct '00

Its just representing activity over the most recent "period". That period is probly relative to the last measurement displayed. -----Original Message----- From: Jamie Lawrence [mailto:jal@thirdage.com] Sent: Friday, October 13, 2000 2:22 PM To: Chris Thompson; toasters(a)mathworks.com Subject: Re: Trivia: what do the splodges on the LCD represent? At 08:45 PM 10/13/00 +0100, Chris Thompson wrote: >During normal operation, the LCD on an F7xx displays each second a >number, which I believe to be the number of NFS operations in the >last second, followed by a row of 0 to 10 splodges --- at any rate, >I have never seen more than 10. > >The hardware installation guide refers to this as "a bar graph" but >doesn't go into details. What is actually being measured here? The >number of splodges is correlated with the number, but not very >strongly, and the number has a far larger dynamic range. After staring at that thing and wondering myself, it looks to me as if it is sort of a sliding rate-of-change meter. When there is a big surge in #/Ops over a short period of time, bar jumps up a lot. Ditto for a big decrease in ops, shaving the bar. However, I've seen the bar nearly pegged with a relatively small number of ops (say, 400 or so) and hanging around 50% when the machine in question was pushing as many as I ever have seen it push. In general, I think the main purpose is to have Cool Blinky Lights. -j

1 0

RE: cifs.home_dir values
by White, Lance 14 Oct '00

14 Oct '00

See http://now.netapp.com/NOW/knowledge/docs/ontap/rel60/html/sag/cifs19.htm#14… for info on setting up multiple directories. (Note that this is for OnTap 6.0 which will run on your F760 (assuming you pass the WAT test). -----Original Message----- From: Coder, James (STP) [mailto:james.coder@guidant.com] Sent: Friday, October 13, 2000 7:53 PM To: 'toasters(a)mathworks.com' Subject: cifs.home_dir values I have a 760 being used for home file server in Solaris / windows environment. This is replacing my e3500 filer server running samba. Can you have more than one volume listed in for cifs.home_dir? If so, how do you set it up? thx in advance - jcc James C. Coder UNIX Administrator Guidant Corporation Phone: 651-582-4797 Email: james.coder(a)guidant.com

1 0

cifs.home_dir values
by Coder, James (STP) 14 Oct '00

14 Oct '00

I have a 760 being used for home file server in Solaris / windows environment. This is replacing my e3500 filer server running samba. Can you have more than one volume listed in for cifs.home_dir? If so, how do you set it up? thx in advance - jcc James C. Coder UNIX Administrator Guidant Corporation Phone: 651-582-4797 Email: james.coder(a)guidant.com

1 0

FILERS IN A HIGH-TRANSACTION SYBASE APP ENVIRONMENT
by Joe Luchtenberg 13 Oct '00

13 Oct '00

Is anybody using NetApp filers for backend storage in a high-transaction Sybase app environment? I have a customer considering this configuration and would appreciate any insight/lessons learned from the field. Thanks. Joe Joe Luchtenberg Dataline, Inc. New number and email, please update: 757.858.0600 858.0606 fax joe.luchtenberg(a)data-line.com

1 0

RE: F760/F840, ONTAP 5.3.x/6.x gripes
by Reidenbach, Edward 13 Oct '00

13 Oct '00

Thanks for the feedback on the Data ONTAP release model. There were numerous factors we considered in the decision making process for both Data ONTAP 6.0 and 5.3.6H1. Factors including * time to market * time frame for separation of code lines * feature/function and regression/reliability testing scope * project schedule differences * and the risk profile of the various alternatives were critical in the decision making process. We believe the resulting release model decisions provided the highest quality solution for you in the most timely manner. Follow-on maintenance releases for both Data ONTAP 5.3.6H1 and 6.0 will address the platform specific issues associated with the above mentioned releases. Ed Reidenbach Marketing Manager, Data ONTAP, Multiprotocol and High Availability Network Appliance Phone: 408-822-6167 Email: edr(a)netapp.com -----Original Message----- From: Todd C. Merrill [mailto:tmerrill@mathworks.com] Sent: Thursday, October 12, 2000 7:11 AM To: toasters(a)mathworks.com Subject: F760/F840, ONTAP 5.3.x/6.x gripes Can someone from NetApp comment on these following concerns? It seems...inconvenient for those who upgraded to ONTAP 6 on F760s to have to revert back to 5.3.6.x in order to upgrade to the F840s. The release of ONTAP 6 couldn't wait until a port to the x86 code was ready for the F840s? And, also, why won't 5.3.6H1 run on an alpha-based filer? I couldn't test 5.3.6H1 *anywhere* in my environment, except on my new F840 which I needed immediately on a high-visibilty filer. I upgraded and went live with 5.3.6H1 and crossed my fingers that there would be no problems with this new release. Very unnerving. I'm now running with two different ONTAP versions on a group of filers, and hoping I don't run into any inconsistencies across my environment due to this. Again, a port of 5.3.6H1 couldn't have been made for the alpha-based filers coincident with the x86-based F840? Lastly, could the boot process be made smarter to load the appropriate microkernel, x86 or alpha, if *both* were included on disk? These gyrations to install and download x86- or alpha-based code from *floppy* (slllooowwwww) just seems so...so...kludgy. Probably not, but I'm just asking. Makes a simple head-swap upgrade not so simple. In summary, in comparison to the usual elegant and quick upgrades and downgrades for previous ONTAP versions, these points above just increased my downtime [0] and decreased my comfort level. (BTW, the 840 is taking the load of our previous 760 without even breaking a sweat. Yipee!) [0] Just one of these migrations or ugprade/downgrades takes you far below a 99.999% uptime metric. Until next time... The Mathworks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

1 0

F760/F840, ONTAP 5.3.x/6.x gripes
by Todd C. Merrill 12 Oct '00

12 Oct '00

Can someone from NetApp comment on these following concerns? It seems...inconvenient for those who upgraded to ONTAP 6 on F760s to have to revert back to 5.3.6.x in order to upgrade to the F840s. The release of ONTAP 6 couldn't wait until a port to the x86 code was ready for the F840s? And, also, why won't 5.3.6H1 run on an alpha-based filer? I couldn't test 5.3.6H1 *anywhere* in my environment, except on my new F840 which I needed immediately on a high-visibilty filer. I upgraded and went live with 5.3.6H1 and crossed my fingers that there would be no problems with this new release. Very unnerving. I'm now running with two different ONTAP versions on a group of filers, and hoping I don't run into any inconsistencies across my environment due to this. Again, a port of 5.3.6H1 couldn't have been made for the alpha-based filers coincident with the x86-based F840? Lastly, could the boot process be made smarter to load the appropriate microkernel, x86 or alpha, if *both* were included on disk? These gyrations to install and download x86- or alpha-based code from *floppy* (slllooowwwww) just seems so...so...kludgy. Probably not, but I'm just asking. Makes a simple head-swap upgrade not so simple. In summary, in comparison to the usual elegant and quick upgrades and downgrades for previous ONTAP versions, these points above just increased my downtime [0] and decreased my comfort level. (BTW, the 840 is taking the load of our previous 760 without even breaking a sweat. Yipee!) [0] Just one of these migrations or ugprade/downgrades takes you far below a 99.999% uptime metric. Until next time... The Mathworks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

3 4

SAN's and backups (Was: Re: Veritas NetBackup + NDMP == Doh!!)
by mds＠gbnet.net 12 Oct '00

12 Oct '00

On Wed 11 Oct, 2000, Marion Hakanson <hakanson(a)cse.ogi.edu> wrote: > I think things are headed toward the use of SAN-connected tape > drives and libraries for this type of setup. The backup software > tells the SAN switch to connect up the appropriate number of drives > to the backup client (NetApp), and then uses NDMP to initiate the > backup on the client, using "direct attached" tape drives. When > that backup is done, you "direct attach" the drives to the next client. I can think of a rather neat scenario along those lines: as described tape drives and disk drives are all on a switched FC fabric along with one or more NetApp filers, and possibly an NDMP backup host or two. When a backup host starts a backup running, the NetApp starts causing blocks to be transferred directly from the disks it manages to the tape drive(s) that the backup host controls, mingled with a little metadata it originates itself, while sending the lion's share of metadata to the NDMP backup host. Nothing should go across any piece of wire any more times than is absolutely necessary. The same scenario could go for any normal open system (Unix, NT, AS400, whatever) in a similar fabric, but I'm not sure who's best placed to actually make it work. I'd like to think NetApp were. See, until these scenarios start coming true, SAN's aren't much more than an overworked buzzword for big flipping messes made of disks and fibre. I still think SAN's are an about-to-be-useful item, but I've yet to see much real progress. > Heck, some libraries (ATL, for one) are becoming available with NDMP > servers right in them, ethernet interface included. Dunno where > you put the file history or tape indexes, but it's getting interesting. With the nominated backup controller boxes, which keep the metadata on-line (on disk) so that restores, archival, cloning etc. remain first-order activities, as much as running databases and maintaining other applications. Backups should never be afterthoughts as they so often are. Sorry, got a bit bitter and twisted there. 8) > Speaking of NAS & tape libraries, does anybody have any experience > with the Grau Data "Infinistore Virtual Disk System" (www.graudata.com)? > It's apparently a NAS box with a RAID cache in front of a tape library. Sounds like fun. Crafted well, the scenario I describe would ideally have the IO balanced to keep the tape drives well-fed, the backup window small, the filer heads serving out of the front-door performantly and the backup controllers adequately busy. Otherwise, what would be the point? Strikes me that they're angling toward HSM which is related but not the same. > The brief article I saw in Server/Workstation Expert magazine mentioned > that an "entry level" system starts at around $100k, so I didn't read > beyond that (:-). We sure wish NetApp had something like this, > since somewhere around 80% of our filer's capacity is tied up in > files that haven't been accessed in a year or more. And that final comment makes me think you'd quite like an HSM product with NetApp-style qualities. I've often thought that it'd be nice if NetApp could make such boxes, by way of extension beyond the filers, but I don't know if anyone's tested the market to see what sort of demand there might be. >-- End of excerpt from Marion Hakanson -- -Mark ... an Englishman in Derby ...

2 1

RE: Migrating 2 x F740 => F840
by Knight, Graham 12 Oct '00

12 Oct '00

If you have two volumes that will be merged on to one F840 and they have the same name, I would suggest attaching only the volume you'd like to use as the root of the F840 first. Then follow the procedure at the link below. After the F840 is up and running on the new volume then add the second volume. When the second vol0 is added the system will recognize it as a foreign volume and leave it offline. At that point you can rename it and bring it online with the other volume. Graham -----Original Message----- From: Brian Tao [mailto:taob@risc.org] Sent: Wednesday, October 11, 2000 8:52 PM To: toasters(a)mathworks.com Subject: Migrating 2 x F740 => F840 Is it possible to "merge" the disk shelves from two or more F740's onto a single F840 without losing the data? It sounds like it might be as simple as having to rename volumes (to avoid having multiple /vol/vol0's) when attaching the shelves to the F840. Has anyone tried this? Any requirements or caveats to watch out for? -- Brian Tao (BT300, taob(a)risc.org) "Though this be madness, yet there is method in't"

1 0

RE: Migrating 2 x F740 => F840
by Ljungberg, Anders 12 Oct '00

12 Oct '00

Hello Brian, Well first you have to decide which volume to be your new root. (or you may be using a brand new one) ( vol options volname root) The volumes will come up as "foreign" and named /vol/vol0 and /vol/vol0(2) Bring them online with "vol online" and change the names with "vol rename" Anders Ljungberg -----Original Message----- From: Brian Tao [mailto:taob@risc.org] Sent: Thursday, October 12, 2000 6:12 AM To: toasters(a)mathworks.com Subject: Re: Migrating 2 x F740 => F840 On Wed, 11 Oct 2000, Brian Tao wrote: > > Is it possible to "merge" the disk shelves from two or more F740's > onto a single F840 without losing the data? It sounds like it might > be as simple as having to rename volumes (to avoid having multiple > /vol/vol0's) when attaching the shelves to the F840. Has anyone tried > this? Any requirements or caveats to watch out for? Here is the document on NOW describing the migration procedure (including kernel and bootblock update) going from an F700 series to an F840: http://now.netapp.com/Knowledgebase/solutionarea.asp?id=3.0.6869703.2865011 It doesn't specifically cover adding another volume of disks that was once considered to be the /vol/vol0 boot volume on different filer, but I imagine that is easily solved by renaming the volume once it is on the destination filer? -- Brian Tao (BT300, taob(a)risc.org) "Though this be madness, yet there is method in't"

1 0

Migrating 2 x F740 => F840
by Brian Tao 12 Oct '00

12 Oct '00

Is it possible to "merge" the disk shelves from two or more F740's onto a single F840 without losing the data? It sounds like it might be as simple as having to rename volumes (to avoid having multiple /vol/vol0's) when attaching the shelves to the F840. Has anyone tried this? Any requirements or caveats to watch out for? -- Brian Tao (BT300, taob(a)risc.org) "Though this be madness, yet there is method in't"

1 1

RE: Veritas NetBackup + NDMP == Doh!!
by Alvarado, Michael 11 Oct '00

11 Oct '00

Here are two documents that overview the new library with the Gigabit Ethernet interface along with the ported NDMP tape server. http://www.atlp.com/pdf/ctrjuly2000.pdf ATL supports NDMP/NAS/GbE http://is.pennnet.com/home/articles.cfm?ARTICLE_ID=81787&VERSION_NUM=1&PUBL… ON_ID=23&Section=CurrentIssue -----Original Message----- From: Joe Luchtenberg [mailto:joe.luchtenberg@data-line.com] Sent: Wednesday, October 11, 2000 2:06 PM To: 'Marion Hakanson'; Scott Mikusko Cc: Cormier, Darleen; 'Jeffrey Krueger'; toasters(a)mathworks.com Subject: RE: Veritas NetBackup + NDMP == Doh!! Didn't take time to check the archives, but I believe the NAS w/SAN centralized tape backup has been addressed here before. The "certified" solution incorporates NetApp/Veritas/Vixel/Spectra Logic. Joe Joe Luchtenberg Dataline, Inc. New number and email, please update: 757.858.0600 858.0606 fax joe.luchtenberg(a)data-line.com > -----Original Message----- > From: Marion Hakanson [SMTP:hakanson@cse.ogi.edu] > Sent: Wednesday, October 11, 2000 2:27 PM > To: Scott Mikusko > Cc: Cormier, Darleen; 'Jeffrey Krueger'; toasters(a)mathworks.com > Subject: Re: Veritas NetBackup + NDMP == Doh!! > > > . . . > > This is where allowing NDMP to dump to a remote tape library is very > > important, because systems such as I support require more than 4 drives > > and having local tapes drives on filers is not really feasible. > > > > -Scott > > I think things are headed toward the use of SAN-connected tape > drives and libraries for this type of setup. The backup software > tells the SAN switch to connect up the appropriate number of drives > to the backup client (NetApp), and then uses NDMP to initiate the > backup on the client, using "direct attached" tape drives. When > that backup is done, you "direct attach" the drives to the next client. > > I think Veritas and NetApp claim to support something like this, but > only with certain SAN switching equipment. If you've got SCSI drives > (as opposed to FC), then you'll need a SCSI-to-FC converter while > you're at it. > > Heck, some libraries (ATL, for one) are becoming available with NDMP > servers right in them, ethernet interface included. Dunno where > you put the file history or tape indexes, but it's getting interesting. > > Speaking of NAS & tape libraries, does anybody have any experience > with the Grau Data "Infinistore Virtual Disk System" (www.graudata.com)? > It's apparently a NAS box with a RAID cache in front of a tape library. > > The brief article I saw in Server/Workstation Expert magazine mentioned > that an "entry level" system starts at around $100k, so I didn't read > beyond that (:-). We sure wish NetApp had something like this, > since somewhere around 80% of our filer's capacity is tied up in > files that haven't been accessed in a year or more. > > Regards, > > -- > Marion Hakanson <hakanson(a)cse.ogi.edu> > CSE Computing Facilities >

1 0

RE: Veritas NetBackup + NDMP == Doh!!
by Joe Luchtenberg 11 Oct '00

11 Oct '00

Didn't take time to check the archives, but I believe the NAS w/SAN centralized tape backup has been addressed here before. The "certified" solution incorporates NetApp/Veritas/Vixel/Spectra Logic. Joe Joe Luchtenberg Dataline, Inc. New number and email, please update: 757.858.0600 858.0606 fax joe.luchtenberg(a)data-line.com > -----Original Message----- > From: Marion Hakanson [SMTP:hakanson@cse.ogi.edu] > Sent: Wednesday, October 11, 2000 2:27 PM > To: Scott Mikusko > Cc: Cormier, Darleen; 'Jeffrey Krueger'; toasters(a)mathworks.com > Subject: Re: Veritas NetBackup + NDMP == Doh!! > > > . . . > > This is where allowing NDMP to dump to a remote tape library is very > > important, because systems such as I support require more than 4 drives > > and having local tapes drives on filers is not really feasible. > > > > -Scott > > I think things are headed toward the use of SAN-connected tape > drives and libraries for this type of setup. The backup software > tells the SAN switch to connect up the appropriate number of drives > to the backup client (NetApp), and then uses NDMP to initiate the > backup on the client, using "direct attached" tape drives. When > that backup is done, you "direct attach" the drives to the next client. > > I think Veritas and NetApp claim to support something like this, but > only with certain SAN switching equipment. If you've got SCSI drives > (as opposed to FC), then you'll need a SCSI-to-FC converter while > you're at it. > > Heck, some libraries (ATL, for one) are becoming available with NDMP > servers right in them, ethernet interface included. Dunno where > you put the file history or tape indexes, but it's getting interesting. > > Speaking of NAS & tape libraries, does anybody have any experience > with the Grau Data "Infinistore Virtual Disk System" (www.graudata.com)? > It's apparently a NAS box with a RAID cache in front of a tape library. > > The brief article I saw in Server/Workstation Expert magazine mentioned > that an "entry level" system starts at around $100k, so I didn't read > beyond that (:-). We sure wish NetApp had something like this, > since somewhere around 80% of our filer's capacity is tied up in > files that haven't been accessed in a year or more. > > Regards, > > -- > Marion Hakanson <hakanson(a)cse.ogi.edu> > CSE Computing Facilities >

1 0

RE: Veritas NetBackup + NDMP == Doh!!
by Cormier, Darleen 11 Oct '00

11 Oct '00

Jeff, FYI, Workstation Solutions' Quick Restore package also allows all of the configuration options that Budtools allowed. If you want to take a look at their stuff, the url is www.worksta.com You're right that Veritas requires the drives in the library to be either directly attached to a filer, or attached to another filer. On the second issue, each F760 is able to support up to four tape drives, an F740 can support 2 and our newer F840 can support up to 8, so it depends on the model of filer you have. If your tape library supports splitting the internal bus, by daisy chaining a couple of drives together and cabling them directly to a filer, we have customers who are doing this, using Veritas to control the robotics. A backup package is required to control the robotics because we have no robotics control capability built-in to DataOnTap. Hope this helps. -----Original Message----- From: Jeffrey Krueger [mailto:jkrueger@qualcomm.com] Sent: Tuesday, October 10, 2000 7:04 PM To: toasters(a)mathworks.com Subject: Veritas NetBackup + NDMP == Doh!! We are currently using BudTool and implementing Veritas NetBackup as a replacement. We want to use NDMP to dump the filer's data off to a NetBackup server (big honkin' sun box w/ lots of library enclosed DLT7k's). Veritas says that in 3.2, they only support NDMP dump/restore to filer attached tape. In 3.4 they claim to support NDMP 3-way dump/restore from one NetApp to another NetApp with locally attached tape. So we are double hosed. Of course, BudTool effortlessly does NDMP to the filer attached tape, 3-way to another filer attached tape, and apparently right back into the BudTool host which is what we want with NetBackup. Anyone else trying to do the same thing? Any other ideas short of sitting on Veritas? On another topic entirely, does anyone have experience attaching many (2-4) DLT7k's per NetApp and using them as a backup tape server via 3-way NDMP? Something tells me that ONTAP and the filer are not designed to handle this well, but we'd like to see if anyone else is trying this. A possible solution for us would be using two or three filers with 4 drives attached to each as a pool of dump/restore servers for the other 20 or 30 filers. Thanks!! -- Jeff -- ---------------------------------------------------------------------------- Jeff Krueger E-Mail: jeff(a)qualcomm.com Senior Engineer Phone: 858-651-6709 NetApp Filers / UNIX Infrastructure Fax: 858-651-6627 QUALCOMM, Inc. IT Engineering Web: www.qualcomm.com

3 2

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