toasters

toasters@lists.teaparty.net

13531 discussions

RE: Web site
by Bailey, Jourdan 02 Feb '00

02 Feb '00

The NOW web site is available. Jourdan Bailey Technical Support Engineer Telephone: 1-408-822-4792 Hotline: 1-888-4-NETAPP Get answers NOW! - NetAPP On the Web - http://now.netapp.com > -----Original Message----- > From: Dave Atkin [mailto:dla1@york.ac.uk] > Sent: Wednesday, February 02, 2000 5:18 AM > To: toasters(a)mathworks.com > Subject: Web site > > > Well I'm pleased that Netapp have improved their website. The > only slight > downside is that attempting go into "software download" produces the > message > > Content-type: text/html > Software error: > Can't tie > /www/external/now-legacy.netapp.com//data//now.db//company2agr > eement.db: > No such file or directory at > /usr/local/lib/perl5.004_04/netapp/WebAdmin/NowToolFuncs.pm > line 1019.<BR> > Please send mail to this site's webmaster for help. > > I guess I'll have to wait a while for 5.3.4R3 ! > > Dave Atkin > ------------------------------------------------------ > Dave Atkin, Head of Technical Services > Computing Service, University of York, YORK YO10 5DD > Phone: +44-1904-433804 (ddi) Fax: +44-1904-433740 > Email: D.Atkin(a)york.ac.uk > ------------------------------------------------------ > >

1 0

Web site
by Nguyen, Alan 02 Feb '00

02 Feb '00

Our "Software download" site has been fixed. The cause is unknown at this point. We will troubleshoot to prevent future problems. Alan. -----Original Message----- From: Dave Atkin [mailto:dla1@york.ac.uk] Sent: Wednesday, February 02, 2000 5:18 AM To: toasters(a)mathworks.com Subject: Web site Well I'm pleased that Netapp have improved their website. The only slight downside is that attempting go into "software download" produces the message Content-type: text/html Software error: Can't tie /www/external/now-legacy.netapp.com//data//now.db//company2agreement.db: No such file or directory at /usr/local/lib/perl5.004_04/netapp/WebAdmin/NowToolFuncs.pm line 1019.<BR> Please send mail to this site's webmaster for help. I guess I'll have to wait a while for 5.3.4R3 ! Dave Atkin ------------------------------------------------------ Dave Atkin, Head of Technical Services Computing Service, University of York, YORK YO10 5DD Phone: +44-1904-433804 (ddi) Fax: +44-1904-433740 Email: D.Atkin(a)york.ac.uk ------------------------------------------------------

1 0

RE: Web site
by Bailey, Jourdan 02 Feb '00

02 Feb '00

I will alert our internal helpdesk, I have also noticed this problem. Jourdan Bailey Technical Support Engineer Telephone: 1-408-822-4792 Hotline: 1-888-4-NETAPP Get answers NOW! - NetAPP On the Web - http://now.netapp.com > -----Original Message----- > From: Dave Atkin [mailto:dla1@york.ac.uk] > Sent: Wednesday, February 02, 2000 5:18 AM > To: toasters(a)mathworks.com > Subject: Web site > > > Well I'm pleased that Netapp have improved their website. The > only slight > downside is that attempting go into "software download" produces the > message > > Content-type: text/html > Software error: > Can't tie > /www/external/now-legacy.netapp.com//data//now.db//company2agr > eement.db: > No such file or directory at > /usr/local/lib/perl5.004_04/netapp/WebAdmin/NowToolFuncs.pm > line 1019.<BR> > Please send mail to this site's webmaster for help. > > I guess I'll have to wait a while for 5.3.4R3 ! > > Dave Atkin > ------------------------------------------------------ > Dave Atkin, Head of Technical Services > Computing Service, University of York, YORK YO10 5DD > Phone: +44-1904-433804 (ddi) Fax: +44-1904-433740 > Email: D.Atkin(a)york.ac.uk > ------------------------------------------------------ > >

1 0

RE: FTP uploads from the UK (fwd)
by Sam Schorr 02 Feb '00

02 Feb '00

Hip, Hip, Hooray!!!!! -----Original Message----- From: Nguyen, Alan [mailto:alan.nguyen@netapp.com] Sent: Tuesday, February 01, 2000 8:59 PM To: 'toasters(a)mathworks.com' Subject: RE: FTP uploads from the UK (fwd) After about a month and a half of testing with customers like yourself, Network Appliance has a new FTP site. The new FTP site will be available for all of our customer's use beginning tomorrow (2/2/2000) at 6PM PST. The transition from the current FTP site to the new will be transparent and the FTP site will not have any downtime. Please read below for more details regarding architecture. Again, thank you for your patience. If you have any questions, please reply to me at anguyen(a)netapp.com <mailto:anguyen@netapp.com> Alan. Mgr., IT Server Ops Network Appliance -----Original Message----- From: Nguyen, Alan Sent: Thursday, December 02, 1999 10:49 AM To: 'toasters(a)mathworks.com' Subject: FW: FTP uploads from the UK (fwd) To All, First, I must apologize for NetApp's poor FTP site performance. The responsibility for this in my group. The situation is our internet connection is highly-utilized and thus your access to our FTP site is affected by this. Actions have been taken to increase internet bandwidth in the next few weeks. However, despite this, it will be superseded by our new FTP implementation. We have designed a new FTP site to improve security, system management, increase internet bandwidth/presence (i.e. ISP hosted site), and system redundancy. Because our current infrastructure is very out-dated, we have taken the "big-bang" approach rather than making incremental improvements. In a little more detail, our new FTP site is at a web-hosted site with 100MB bandwidth capacity vs current T1, and is configured with two fail-over dual-processor Unix systems with cluster fail-over (F760) Filer pair for storage. In short, the new site is very robust and will scale to meet to our customers' needs. What has been done thus far? Contract has been signed with new web-hosting site. Networking infrastructure is in place. Programming is complete for account and directory management. The new systems are racked at the new web-hosted site. As I type this, we are plugging in the cables for the new systems. What's the schedule? In the next two days, we will perform internal/external FTP tests. On Monday (Dec 6th), we will start live tests with customers assuming everything goes well this week. After which, the new site will go live. I will announce to this list when we are live. If you have any questions, please direct them to me anguyen(a)netapp.com <mailto:anguyen@netapp.com> or reply to this list as I will be joining this list soon. Again, thank you for your patience. Alan Nguyen Network Appliance Mgr., IT Server Operations

1 0

console cable pinout
by Ronan Mullally 02 Feb '00

02 Feb '00

After almost two years without a toaster I've finally come out of the wilderness and am once again the proud "owner" (ok, admin) of a nice shiny new toaster. Pity they're not as shiny as they used to be in the Good Ol' Days :( I'm trying to connect this beastie to a console server. Specifically a Linux box with a Cyclades Ze multiport serial card. The Cyclades presents its ports using an expansion box with RJ45 female ports. I haven't been able to get this to talk to the toaster's console port :( I've tried various combinations of straight-through and null-modem adaptors with standard ethernet and roll-over drop cables with no joy. Can anybody point me in the direction of the pinouts I need in a RJ45 - DB-9 adaptor that will work with a roll-over drop cable? For reference, I've got the Cyclades ports talking to cisco kit (3600s) using a roll-over cable. I've also got them talking to Foundry switches using the same roll-over cable with an adaptor using the pinout below. RJ45 DB-9 Signal Pinout Pinout Signal TxD 3 --------------------- 2 RxD RxD 6 --------------------- 3 TxD Gnd 4 --------------------- 5 Gnd RTS 1 ----+ +----- 7 RTS CTS 5 ----+ +----- 8 CTS DTR 2 ----+ +----- 4 DTR DCD 7 ----| |----- 1 DCD DSR 8 ----+ +----- 6 DSR -Ronan

2 2

Web site
by Dave Atkin 02 Feb '00

02 Feb '00

Well I'm pleased that Netapp have improved their website. The only slight downside is that attempting go into "software download" produces the message Content-type: text/html Software error: Can't tie /www/external/now-legacy.netapp.com//data//now.db//company2agreement.db: No such file or directory at /usr/local/lib/perl5.004_04/netapp/WebAdmin/NowToolFuncs.pm line 1019.<BR> Please send mail to this site's webmaster for help. I guess I'll have to wait a while for 5.3.4R3 ! Dave Atkin ------------------------------------------------------ Dave Atkin, Head of Technical Services Computing Service, University of York, YORK YO10 5DD Phone: +44-1904-433804 (ddi) Fax: +44-1904-433740 Email: D.Atkin(a)york.ac.uk ------------------------------------------------------

1 0

RE: Toasters in Compaq Racks and Unicode Files
by Dani Tal 01 Feb '00

01 Feb '00

It looks like Hebrew has a bug. When you try to create a folder, the initial name is "new folder", then when I try to rename it to something in Hebrew , I get a message: \\filer\home_dir <\\filer\\home_dir> does not exist , do you want to create it ? If I answer yes I get " can not create or replace xxx. Can not create a file when that file already exists . So for me it is only English file name on the filer. Dani Tal Manager Information Systems Siemens Data Communication ltd. Tel. +972-4-9953200 X 0403 FAX. +972-4-9888097 Mobile +972-53797814 -----Original Message----- From: Bruce Sterling Woodcock [SMTP:sirbruce@ix.netcom.com] Sent: Mon, 31 January, 2000 16:41 To: Andrew-M.Webster(a)db.com; toasters(a)mathworks.com Subject: Re: Toasters in Compaq Racks and Unicode Files > Also, can anyone relay any experiences (good or bad) with using Non-English > files and directories for CIFS clients on Toasters. This isn't exactly what you're asking, but I know that Windows copying files over CIFS used to choke if it ran into a file with a space at the end of the name. I don't know if that was ever resolved. Bruce

1 0

RE: Toasters in Compaq Racks and Unicode Files
by Hawley, Rob 01 Feb '00

01 Feb '00

The product is intended to be used (and is certainly tested) in an environment where the filer is accessed by many different language systems. Any limitations described here are limitations of the clients and protocols and are not limitations introduced by the filer. The language setting is relevant in a cifs only environment since it controls how we generate 8.3 names, system, and domain names. For Win9x clients it will also be how we encode share names. >If setting the language is needed what impact does this have with multiple >language clients, ie one kanji, one US English and maybe a laptop user from >Korea? In such a mixed, cifs-only environment I would set the language to "C", and use only ASCII strings for system and domain names. System and domain names are transmitted using the OEM character set selected by the language setting on the system volume. If you are using NT, then you are not restricted in the share names, but bear in mind that an English client installed with the normal fonts will not be able to display a Japanese share name correctly. The filer supports non-ASCII system and domain names, but limitations on the clients effectively require that everyone agree on the language being used. For example if you use the name nihongo (in the correct Kanji) for your domain name, Japanese systems will correctly display the domain name. (We have such a domain in our development lab). English systems will see the name as a series of European characters since they cannot interpret the OEM character strings. As long as you are using an NT backup program running on NT you should not have a problem. Win9x does not implement a full version of UNICODE so I would not expect it to work correctly there. -----Original Message----- From: Alan R. White [mailto:arw@tipper.demon.co.uk] Sent: Monday, January 31, 2000 11:39 AM To: toasters(a)mathworks.com Cc: Andrew-M.Webster(a)db.com Subject: Re: Toasters in Compaq Racks and Unicode Files Hi, Sorry 'bout the bunch more questions below - I'd like to understand what this means in more detail if there's knowledgable folks out there. Is unicode mode enabled automatically on the wafl fs? Is the language setting relevant in a CIFS only environment or is it only to allow translation between single-byte clients and wide charsets? If setting the language is needed what impact does this have with multiple language clients, ie one kanji, one US English and maybe a laptop user from Korea? Does it have any impact if the user consistently uses the same language/nt build on the client? How about backup/restore via CIFS - are we OK as long as the CIFS backup client understands unicode? Any idea on maximum filename lengths? The Win32 SDK says 32,000 unicode characters if prefixed with "\\?\" - anyone stress tested this? Hope I get it with answers to all these questions - talk about opening pandoras box Andrew ;-) TIA, Alan ----- Original Message ----- From: Joan Pearson <jpearson(a)netapp.com> To: <Andrew-M.Webster(a)db.com> Cc: <toasters(a)mathworks.com> Sent: Monday, January 31, 2000 5:11 PM Subject: Re: Toasters in Compaq Racks and Unicode Files > Hi, > Some requests from NT do not supply Unicode file names, but > instead use the PC character set. Setting the language indicates > which PC character set is in use, so that it can be properly translated > to Unicode. > > The volume's language should be set as soon as possible after > bringing up a 5.3 release. > > Joan Pearson > Network Appliance > > > At 10:47 PM 1/30/00 , you wrote: > >Hi There, > > > > Has anyone out there installed F 760es in Compaq racks? Any special tips, > >problems etc.... > > > > Also, can anyone relay any experiences (good or bad) with using Non-English > >files and directories for CIFS clients on Toasters. > > > > And finally, the System. Admin Guide talks about setting a language of a > >volume with the "vol lang volume langcode" command. Does this having any > >affect for Unicode clients, ie NT? > > > >Any help appreciated. > > > >-- > >Andrew Webster > >Windows NT Consultant > >Sanno Move Project > >Deutsche Securities Ltd. - Tokyo > >Tel. +81-3-5408-4787 > >Fax. +81-3-5401-6998 > >Andrew-M.Webster(a)db.com > > >

1 0

RE: win2k , WINS and serving CIFS
by Muhlestein, Mark 31 Jan '00

31 Jan '00

You shouln't have any problems with Win2k clients as long as you have the latest version of OnTap (5.3.4R3). Take a look at http://now.netapp.com/NOW/misc/head/win2000.shtml for a writeup on Windows 2000 and the filer. It's slightly out of date but you will get the basic idea. Mark Muhlestein -- mmm(a)netapp.com -----Original Message----- From: Farid Hamjavar [mailto:hamjavar@unm.edu] Sent: Friday, January 28, 2000 2:40 PM To: toasters(a)mathworks.com Subject: win2k , WINS and serving CIFS Hello, This is not entirely related to netapp but it somewhat is. Being a primarily unix environment, we have these two windows-nt boxes that really don't do much but act as WINS server so our clients can access the netapp filers. And that's how we serve CIFS. Not being a windows person, I like to ask what do we need to do with win2k coming to picture. Will a win2k client be able to access files via CIFS the way we do it now? Is that true that WINS service is history and no longer available? If so, how would we keep on giving CIFS service to our windows clients? Thanks, Farid

1 0

RE: Toasters in Compaq Racks
by Benn, Paul 31 Jan '00

31 Jan '00

NetApp racks and Compaq racks are both made by Rittal. The mounting flanges on the Compaq version, however, are set nearly flush to the door jam of the cabinet whereas the NetApp version of the rack has the mounting flanges set a bit farther back (closer to the vertical support beams). Though I'm pretty familiar with both rack styles I've never tried this. In my, somewhat educated opinion, there are a few possible issues: 1) The filer will need to be supported in the front as well as the rear. While mounting the disk shelves directly to the flange should be okay, mounting the filer using the front support only probably isn't a good idea given the thickness (rather, thinness) of the flange. The horizontal support rails that ship with a filer *might* work in the Compaq version. Similarly, the horizontal support rails that ship with the Compaq also *might* work (you likely have a number of these laying around since Proliants typically ship with their own slide rails or horizontal support rails). 2) Now if the above turns out to be a non-issue, you may have a problem closing the cabinet door since the filer's bezel sticks out farther than your typical Compaq Proliant does. 3) Even if the door does close, the bezel might be set too close to the Plexiglas door panel which might cause ventilation problems (you wouldn't want to suffocate your poor filer, now would you?). I've seen some shops, however, where they opt not to install the front door on their Compaq racks. 4) I don't recall if the cabinets are available in the same height. If so, you might be able to get away with mounting a NetApp version next to a Compaq version using a coupling kit. Since the internal vertical bracing seems to be different, you may not be able to install the angle brackets for extra support. It probably seems like I'm pitching the NetApp solution as a stand-alone rack (okay, so I am) but Rittal (http://www.rittal.com/) may have a better answer for you. Meanwhile, I'll poke around over here to see if I can come up with some better news. Regards, Paul Benn Network Appliance -----Original Message----- From: Andrew Webster [mailto:Andrew-M.Webster@db.com] Sent: Sunday, January 30, 2000 10:47 PM To: toasters(a)mathworks.com Subject: Toasters in Compaq Racks and Unicode Files Hi There, Has anyone out there installed F 760es in Compaq racks? Any special tips, problems etc....

1 0

Toasters in Compaq Racks and Unicode Files
by Andrew Webster 31 Jan '00

31 Jan '00

Hi There, Has anyone out there installed F 760es in Compaq racks? Any special tips, problems etc.... Also, can anyone relay any experiences (good or bad) with using Non-English files and directories for CIFS clients on Toasters. And finally, the System. Admin Guide talks about setting a language of a volume with the "vol lang volume langcode" command. Does this having any affect for Unicode clients, ie NT? Any help appreciated. -- Andrew Webster Windows NT Consultant Sanno Move Project Deutsche Securities Ltd. - Tokyo Tel. +81-3-5408-4787 Fax. +81-3-5401-6998 Andrew-M.Webster(a)db.com

4 4

Re: rsync and nfs file systems
by jrg＠acm.org 31 Jan '00

31 Jan '00

On Mon 31 Jan, 2000, "Bruce Sterling Woodcock" <sirbruce(a)ix.netcom.com> wrote: >It's not Solaris per se; it's been a standard UNIX-ism for a long time. Yup. Before 4.2BSD, there wasn't an rmdir and directories were made and removed with link/unlink (so not very atomic). Excerpt from SunOS 4.1.x man pages: unlink (8V): WARNINGS Only the super-user can unlink a directory, in which case the files it contains are lost. The files can, however, be recovered from the file system's lost+found directory after performing an fsck. unlink (2V): NOTES Applications should use rmdir(2V) to remove directories. Although root may use unlink() on directories, all users may use rmdir(). Of course, it's awkward for root - especially if something changes from a file to a directory whilst it's not looking - as there doesn't seem to be a way to call unlink whilst saying 'but not directories'. James.

1 0

logging quota exceeded messages
by R.P. Aditya 30 Jan '00

30 Jan '00

I'm unable to find out (on NOW) what facility and level disk quota exceeded messages are syslogged: Sat Jan 29 23:01:53 PST [de0]: uid 14602: disk quota exceeded on volume vol0 I don't want to see them in general...however I'm wary of missing other stuff. I currently have: *.info /dev/console *.warning /etc/messages *.err;kern.* @adminhost which I presume should catch most interesting events apart from the the quota exceeded messages... Advice is appreciated. Thanks, Adi

2 3

RE: Shelf Temp. Alarms...
by Datta, Diptish 30 Jan '00

30 Jan '00

Eyal, This sounds like bug 16767 - you can look it up on NOW. If you are not running 5.3.4 then this is the likely suspect. -diptish -----Original Message----- From: Eyal Traitel [mailto:eyal.traitel@motorola.com] Sent: Saturday, January 29, 2000 10:02 AM To: toasters(a)mathworks.com Cc: support(a)netapp.com; vine(a)netapp.com; dubi; Menachem Kaiser (r53417); Mati Giladi (r57914); Haim Leib (r52445) Subject: Shelf Temp. Alarms... Importance: High Hi ! Does anybody have this errors ALL THE TIME ?!? It seems that we've had enough of these... CAN SOMEONE SOLVE THIS ONCE AND FOR ALL ????? Eyal/Motorola. ****************************************** Eurologic EDM shelf 7 on host adapter 7 reports a critical condition Digital Temperature Sensor Element 1 reports critical status overtemperature failure temperature: 228 (C), 442 (F) Sat Jan 29 18:59:49 IST [statd]: *** NOTICE *** Fault reported on disk storage shelf attached to slot 7. Please check fans, power and temperature.

1 0

Shelf Temp. Alarms...
by Eyal Traitel 29 Jan '00

29 Jan '00

Hi ! Does anybody have this errors ALL THE TIME ?!? It seems that we've had enough of these... CAN SOMEONE SOLVE THIS ONCE AND FOR ALL ????? Eyal/Motorola. ****************************************** Eurologic EDM shelf 7 on host adapter 7 reports a critical condition Digital Temperature Sensor Element 1 reports critical status overtemperature failure temperature: 228 (C), 442 (F) Sat Jan 29 18:59:49 IST [statd]: *** NOTICE *** Fault reported on disk storage shelf attached to slot 7. Please check fans, power and temperature.

2 1

RE: win2k , WINS and serving CIFS
by ZANGA, Michael, GCM 29 Jan '00

29 Jan '00

Wins presenting mantains Netbios names. Win2k no longer has this dependency so NEW clients i.e. Win2k will only DNS for resolution. Downloevel clients will still need WINS but you can upgrde it to the Win2k version and make it more stable and integrated with DNS. -----Original Message----- From: Farid Hamjavar [mailto:hamjavar@unm.edu] Sent: Friday, January 28, 2000 5:40 PM To: toasters(a)mathworks.com Subject: win2k , WINS and serving CIFS Hello, This is not entirely related to netapp but it somewhat is. Being a primarily unix environment, we have these two windows-nt boxes that really don't do much but act as WINS server so our clients can access the netapp filers. And that's how we serve CIFS. Not being a windows person, I like to ask what do we need to do with win2k coming to picture. Will a win2k client be able to access files via CIFS the way we do it now? Is that true that WINS service is history and no longer available? If so, how would we keep on giving CIFS service to our windows clients? Thanks, Farid

1 0

RE: [NETAPP LOG #95883.00000]Shelf Temp. Alarms...
by Cremers, Ronald 29 Jan '00

29 Jan '00

Hello Eyal, thanks for sharing this, the annoying reported messages should well be fixed, once and for all, in Data OnTap 5.3.4 and up. If you are running that, and still see this error the shelf in question should be replaced. See bugs online for bug 16767: The SCSI-3 Environmental Services (SES) driver on occasion emits an erroneous overtemperature failure containing an impossible temperature. An overtemperature failure would normally be preceded by an overtemperature warning; the warning is not issued. The error is typically quickly reported as having been corrected. A typical report: Eurologic EDM shelf 5 on host adapter 8 reports a critical condition Digital Temperature Sensor Element 1 reports critical status overtemperature failure temperature: 217 (C), 422 (F) Sometime the temperature reported is very low rather than very high. It is very apparent that this fault should be ignored. A temperature reported so far above fault conditions cannot be reached in a ten second interval. WORKAROUND: There is no workaround. Examine the messages file to see if there are any persistent temperature reports. Constant faults would indicate the temperature sensor has reached a hard error state. In that case the shelf must be replaced. FIX: The problem is corrected in 5.3.4 Ronald Cremers, NetApp TSE, Europe. )-----Original Message----- )From: owner-dl-support(a)netapp.com )[mailto:owner-dl-support@netapp.com]On )Behalf Of Eyal Traitel )Sent: 29 January 2000 19:02 )To: toasters(a)mathworks.com )Cc: support(a)netapp.com; vine(a)netapp.com; dubi; Menachem Kaiser )(r53417); )Mati Giladi (r57914); Haim Leib (r52445) )Subject: [NETAPP LOG #95883.00000]Shelf Temp. Alarms... ) ) )Hi ! ) )Does anybody have this errors ALL THE TIME ?!? )It seems that we've had enough of these... ) )CAN SOMEONE SOLVE THIS ONCE AND FOR ALL ????? ) )Eyal/Motorola. ) ) )****************************************** )Eurologic EDM shelf 7 on host adapter 7 reports a )critical condition ) Digital Temperature Sensor ) Element 1 reports critical status ) overtemperature failure ) temperature: 228 (C), 442 (F) ) )Sat Jan 29 18:59:49 IST [statd]: *** NOTICE *** ) Fault reported on disk storage shelf )attached to slot 7. ) Please check fans, power and temperature. ) ) ) )

1 0

RE: max file system size
by mds＠gbnet.net 28 Jan '00

28 Jan '00

On Fri 28 Jan, 2000, tkaczma(a)gryf.net wrote: > On Fri, 28 Jan 2000, Jason Middlebrooks wrote: > > NetApp person. Besides, I though they were 36GB drives. As per > performance, of course, there is bus arbitration overhead to start. Also Bus arbitration may be reduced by reducing the number of active devices on a loop? At least that would be my intuition. Has anyone quantified any effects this might carry through to the head's performance? Analytic, simulated or empiracle evidence would all be acceptable! > you're talking about putting in huge drives, which maximum input/output is > still about the same as the smaller drives. Drive performance hasn't > improved all that much while capacity has increased significantly. That's always been one of my concerns - in fact I think this has cropped up on the list before at least once. Each successive volume increment in SCSI disks seemed to about double capacity while only raising IO's/s by say 20% over each generation. As capacities grow this gradient mismatch between the two characteristics means data gets relatively less accessible per spindle. This is why *more* spindles are often recommended for high-performance/bandwidth requirements. So far so orthodox. It'd be nice to see what bus arbitration, RAID calculation overheads and fs-layout&maintenance take away by counterpoint due to such increased spindle-counts. If anything. My intuition is that spindles are a key factor in these algorithms, whereas larger linear arrays of sectors at each spindle are less significant. Any volunteers to argue either way? Recently I've been pleasantly surprised by the performance of Seagate/Sun 9GB disks - able to transfer 2-3x what old 4.5GB disks could, but then I suppose this is a generation or two further on and 4.5GB disks aren't really being produced. Fibre-channel product almost certainly gets highest-performance controllers, interfaces and firmware so maybe Tom and I do them a disservice by suspecting them of the same capacity-curve mismatch described above. Evidence about all this would be welcome. I'd bet we've almost all of us, at some point, bought at least one filer intended for heavy use (usenet, Oracle, etc.) and had to intuit how our disk-size, spindle-count and loop/chain-count decisions would impact our actual performance. Often this might even override questions of money (perish the thought!). If someone in NetApp engineering or testing hasn't had their own curiosity piqued by this vexed question yet and drawn up matrices describing the tradeoffs, or even shown convincingly that there are no real tradeoffs to be made, then perhaps someone might sound them out about it soon? Ever living in hope.. >-- End of excerpt from tkaczma(a)gryf.net -- -Mark ... an Englishman in London ...

1 0

win2k , WINS and serving CIFS
by Farid Hamjavar 28 Jan '00

28 Jan '00

Hello, This is not entirely related to netapp but it somewhat is. Being a primarily unix environment, we have these two windows-nt boxes that really don't do much but act as WINS server so our clients can access the netapp filers. And that's how we serve CIFS. Not being a windows person, I like to ask what do we need to do with win2k coming to picture. Will a win2k client be able to access files via CIFS the way we do it now? Is that true that WINS service is history and no longer available? If so, how would we keep on giving CIFS service to our windows clients? Thanks, Farid

1 0

RE: Minimum Disks
by Mohler, Jeff 28 Jan '00

28 Jan '00

Four, you have to have drive IDs 3,4,11,12 in-place for FCAL shelf services to function properly. Add IDs 2 and 10 if you have a cluster. -----Original Message----- From: July at Zerowait [mailto:july@zerowait.com] Sent: Friday, January 28, 2000 11:52 AM To: toasters(a)mathworks.com Subject: Minimum Disks On a F740 with two shelves, what is the minimum number of disks a person can use? -- ---------------------------------------------------------------- July A. J. Linett, VP Tech Mktg 302.266.9408 july(a)zerowait.com Zerowait: Best Of Breed Internet Core Technologies Web Acceleration ** Data Storage ** Load Balancing www.zerowait.com*www.loadbalancing.net*www.nas-san.com -------------------------------------------------------------------

1 0

Minimum Disks
by July at Zerowait 28 Jan '00

28 Jan '00

On a F740 with two shelves, what is the minimum number of disks a person can use? -- ---------------------------------------------------------------- July A. J. Linett, VP Tech Mktg 302.266.9408 july(a)zerowait.com Zerowait: Best Of Breed Internet Core Technologies Web Acceleration ** Data Storage ** Load Balancing www.zerowait.com*www.loadbalancing.net*www.nas-san.com -------------------------------------------------------------------

1 0

RE: max file system size
by Reidenbach, Edward 28 Jan '00

28 Jan '00

Jason' response is correct. -----Original Message----- From: Timothy A. McCarthy [mailto:tmac@netapp.com] Sent: Friday, January 28, 2000 7:37 AM To: Jason Middlebrooks Cc: toasters(a)mathworks.com Subject: Re: max file system size The F760 could have at most: 168 9GB disks for total Raw/unformatted capacity of 1.4TB (Minimum 3 loops of 56 disks) or 84 18GB disks for a total Raw/unformatted capacity of 1.4TB (Minimum 2 loops, can have no more than 56 drives on one loop) or 42 36GB disks for a total Raw/unformatted capacity of 1.4TB (Minimum 1 loop. More loops can be added if desired).

1 0

RE: max file system size
by Bond, Andrew 28 Jan '00

28 Jan '00

Actually no you can't - 56 * 36 = 2TB, which is not currently supported. You can have 42 * 36GB drives on an F760 (42 * 34.5GB formatted = 1.4TB). I don't believe there is any performance compromise here; there may possibly be a benefit in splitting the disks over two loops for particular access patterns (for example large sequential access), but I don't believe we have done any profiling of this yet. Regards, Andrew > -----Original Message----- > From: Jason Middlebrooks [mailto:Jason_Middlebrooks@datalink.com] > Sent: 28 January 2000 15:44 > To: Bond, Andrew > Cc: toasters(a)mathworks.com > Subject: RE: max file system size > > > So on a F760 I can have 56 32GB drives on one fibre loop? Is any > performance lost here? > > Thanks > Jason > > > > > > "Bond, Andrew" <abond(a)netapp.com> on 01/28/2000 09:48:33 AM > > To: Jason Middlebrooks/Datalink > cc: toasters(a)mathworks.com > Subject: RE: max file system size > > > > > This question should really be addressed to your local NetApp > sales person > or > SE, however: > > > Is the maximum storage per fibre loop still 464.4GB? > > No. > > The maximum storage per F760 is presently approx. 1.4TB raw > disk space. > This > can be on one loop or more than one loop, in whatever config > you require, > subject to a maximum of 56 devices per loop. > > Regards, > Andrew > > > >

1 0

max file system size
by Jason Middlebrooks 28 Jan '00

28 Jan '00

Is the maximum storage per fibre loop still 464.4GB? So I can not have 8 shelves full of 18 GB drives but I can have the following: 56 9 GB drives 28 18 GB drives 14 36 GB drives If so this sucks. The drives are improving but the total storage space is not increasing. When is this going to change? Jason

1 0

RE: max file system size
by Bond, Andrew 28 Jan '00

28 Jan '00

This question should really be addressed to your local NetApp sales person or SE, however: > Is the maximum storage per fibre loop still 464.4GB? No. The maximum storage per F760 is presently approx. 1.4TB raw disk space. This can be on one loop or more than one loop, in whatever config you require, subject to a maximum of 56 devices per loop. Regards, Andrew

1 0

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