toasters

toasters@lists.teaparty.net

13531 discussions

RE: Problem with WINS registration on 5.3.4R3
by Bond, Andrew 04 Apr '00

04 Apr '00

Maybe the WINS server is not accessible to the filer (in which case there should be errors in the messages file). Some versions of ONTAP allow you to run "cifs testdc" and this will report whether the name was registered by WINS or by broadcast because the WINS server was not available (I think this shows as "WINS" or "WINS broadcast" respectively). It is also possible to disable wins registration for specific IP address using the -wins flag to ifconfig. Have you opened a call with NetApp support? There are other tools to debug this.... Regards, Andrew > -----Original Message----- > From: owner-dl-toasters(a)netapp.com > [mailto:owner-dl-toasters@netapp.com]On Behalf Of Johan Gillquist > Sent: 04 April 2000 07:16 > To: netapp mailing list > Subject: Problem with WINS registration on 5.3.4R3 > > > Hi all, > > I have have a problem with my F720 running 5.3.4R3, CIFS and NFS. The > toaster does not seem to register itself in WINS. When I run a "cifs > sessions" command it states "WINS servers: xxx.xxx.x....." > and it's the > correct address. When a check the WINS server there is no entry's for > the filer. If I enter a static entry in the WINs database it > works fine > > Suggestions ? > > /Johan >

1 0

Problem with WINS registration on 5.3.4R3
by Johan Gillquist 04 Apr '00

04 Apr '00

Hi all, I have have a problem with my F720 running 5.3.4R3, CIFS and NFS. The toaster does not seem to register itself in WINS. When I run a "cifs sessions" command it states "WINS servers: xxx.xxx.x....." and it's the correct address. When a check the WINS server there is no entry's for the filer. If I enter a static entry in the WINs database it works fine Suggestions ? /Johan

1 0

RE: Legato NetApp Client Pak = weak
by Graydon Dodson 03 Apr '00

03 Apr '00

> Is anyone else using the Legato Client Pak for NetApp? The fact that it > doesn't support the GUI doesn't bother me, but I *do* have a problem that I > have to manually restart the daemon on the filer after every reboot. There > doesn't seem to be a way to start if from the rc file, as you have to type > the filer's password. We have had MAJOR problems using Legato's client pack for NetApp. Both with filler panics and with bad backups. My recommendation would be to avoid it. > Does anyone have insight into whether or not Legato will be expanding this > product? Legato's answer to us was to give us a free license to BudTool. But it's future is far from certain. Thanks. Graydon Dodson grdodson(a)lexmark.com Lexmark International Inc.

1 0

Legato NetApp Client Pak = weak
by Timothy Demarest 03 Apr '00

03 Apr '00

Is anyone else using the Legato Client Pak for NetApp? The fact that it doesn't support the GUI doesn't bother me, but I *do* have a problem that I have to manually restart the daemon on the filer after every reboot. There doesn't seem to be a way to start if from the rc file, as you have to type the filer's password. Does anyone have insight into whether or not Legato will be expanding this product? Ti

1 0

RE: How are you mounting CIFS homedirs with Win9x?
by Mohler,Anissa 03 Apr '00

03 Apr '00

You have hit the issue exactly... basically can't do that in 95 - not on Filers or NT Servers. Here are the two options I explored in my implementation of a filer in a CIFS only environment at my old job. I ended up using option #2 with batch files to ease the pain. I just couldn't get kixtart to run consistently. You can do some searches on MS's web site exact command to add in your batch file to automatically assign the user home directory in the NT Profile. Unfortunately I don't have the batch file myself anymore. Two options: #1 You can use a program like WINSET.EXE and a logon script processor like KiXtart to set the environment variable then map a drive for the 95 user in the same manner he would map the drive for the NT users. (Microsoft's web site has some good KB articles on this). #2 You can manually (or through batch file) set the Home Directory property (in NT User Manager for Domains under User Properties/Profile) to the desired home directory for each user. E.g. You would have to set the Home Directory property for your 95 users to Connect Z: (or whatever letter) to \\filer\username. Then you can use the native NT login script to map the 95 users to NET USE Z: /HOME This topic is pretty well covered on Microsoft's web site. Another option you can try is just place a shortcut to the filer on the users desktop. Then they could doubleclick on the filer and they will see their user home directory at the top level. Not as slick as some of the other options, but it should work as well. Hope this helps. :-----Original Message----- :From: Timothy Demarest [mailto:demarest@arraycomm.com] :Sent: Friday, March 31, 2000 10:05 AM :To: toasters(a)mathworks.com :Subject: How are you mounting CIFS homedirs with Win9x? : : : :I was wondering how everyone is mounting CIFS file systems :from Windows 9x :clients? Having worked through the NIS netgroup issue (Bug :#19922, patched :in 5.3.4R3P2 -- thanks everyone), I'm moving on to getting the :new filer :working with PCs. : :I have a CIFS homedir for each user. The filer shares this directory so :that it appears as the user name. In a Windows NT login :script, I can use :the "%username%" directive to mount the user's home dir as H:. :However, you :can't use "%username%" with Windows 9x clients. : :One option is to set the home directory in the user's :environment profile :(in User Manager), but this requires changing each user's profile. The :logon script is simple and elegant. : :I'd really like to mount the CIFS homedir directly, rather :than having the :user open the drive, then open the folder with their :associated username in :order to access files in their home directory. : :Any hints? : :Tim : :

2 1

Use of Netapp as DMZ "glue" device - possible?
by Jim Klun 7183 03 Apr '00

03 Apr '00

I am contemplating this architecture: INTERNET | CISCO PIX FW ------- DMZ FTP SERVER | | ------------------------------ | | | | NT CLIENT NETAPP The DMZ FTP Server would have an a UDP NFS mount THROUGH THE FIREWALL into the internal network. Inbound guest FTP would be directed there via the Netapp NFS mount. The NT box would have a covering SMB mount over the same filesystem. After a guest client delivers data to the FTP server it would be moved from the in to the out directory where it would be picked up by the internal NT box. Intent: Maintain an Internet-visible FTP server in the DMZ and yet provide easy access for inside servers. Question: 1. Does this make sense? Does anyopne use filers this way or in related ways? 2. Are there known exploits against filers doing UDP NFS as I describe above. Could the Netapp be attacked if the FTP box were hacked? 3. Related question: Can admin access to the filer be to ONLY the console port or ONLY a single interface? Thanks - Jim Klun * James A. Klun Sterling Commerce 4600 Lakehurst Ct. Dublin, Oh 43016 * * jklun(a)stercomm.com 614-793-7183 voice 614-793-7092 fax * * ---- "very like a whale" ---- * * ---- Hamlet, Act 2, Scene 2 ---- *

6 7

EMC Celerra vs NetApp Filer
by Gilles TOURPE 03 Apr '00

03 Apr '00

Hello, Did anybody get in the same site, an EMC Celerra system and a NETAPP F7xx ( at least 740 ) Filer ? I would be interrested in Performances, administration, supports issues ... I'm very curious about the Celerra solution. On the paper it looks smart and really really NAS-oriented ( quite funny for SAN-guys :-) ). Some tests we did we pretty concluant also ( and complex also ... ). Any thinking, remarks ???? Thanx Gilles ------------------------------------------------------------ Gilles TOURPE - System and Network Administrator APS BNP ARBITRAGE - Salle des marches ACTIONS - PARIS - France Phone : (33)-1-40-14-24-03 e-mail : gilles(a)bnpcn.com

2 1

Fwd: NetCache/NetApp Release 3.4
by Jamie Lawrence 03 Apr '00

03 Apr '00

From Bugtraq. -j >Approved-By: aleph1(a)SECURITYFOCUS.COM >Delivered-To: bugtraq(a)lists.securityfocus.com >Delivered-To: BUGTRAQ(a)SECURITYFOCUS.COM >X-Hate: Where do you want to go to die? >Date: Thu, 30 Mar 2000 10:16:08 +0200 >Reply-To: Michal Zalewski <lcamtuf(a)DIONE.IDS.PL> >Sender: Bugtraq List <BUGTRAQ(a)SECURITYFOCUS.COM> >From: Michal Zalewski <lcamtuf(a)DIONE.IDS.PL> >Subject: NetCache/NetApp Release 3.4 >X-To: BUGTRAQ(a)SECURITYFOCUS.COM >To: BUGTRAQ(a)SECURITYFOCUS.COM > >NetCache by Network Appliance (www.netapp.com) is one of the most popular >heavy-duty commercial proxy/cache application. In fact, it's rather poorly >written. We can't see the source code, but, some side effects instead. >There's a lot of them, but I'll try to focus on something called 'internal >requests' - requests that access proxy itself and are handled specially. > >For example, by connecting to proxy server and sending 'GET >http://proxy_server_itself:8080/disk_objects/help', we'll (usually) get >error message with this URL and a few stupid characters appended at the >end of it. It won't happen if you specify anything after 'help' - so I >believe it's something like broken sscanf() used to determine which help >object user wants. But that's not the point. Try appending slash and >approx 10k of 'A' letters to our request. In case of any other request >treated as 'external', it might result only in error message. But in this >case, with something around 9850 characters, our connection to proxy >server is immediately dropped... sounds familiar? I believe it's an >overflow. > >Another way to access it (urm, I mean, cause crash) is something like: >'GET disk_object://xx/AAAAA...'. Btw. I'm wondering there's anything >interesting available to download this way? There are some pictures, >disk_object://xx/help/graphics/help.gif and so on, but I haven't access to >NetCache box to check what else is inside... Aaaah, almost forgotten! Any >file within disk_object hierarchy might be downloaded as-is by appending >'/' to URL - for example, 'disk_object://xx/help/graphics/help.gif/' will >return text/plain dump of this GIF. This means, NetCache fails to classify >this file, so if there's any script or other special object, it won't be >recognized as something 'special'? > >-- > >Just to make everything clear - I haven't shell access to running NetCache >box, so I cannot verify I'm absolutely right - eg. if there's anything >interesting within disk_objects or what extactly happens, but I think >there's absolutely something wrong, and there's no excuse for poor, even >commercial, code. > >#include <stddiscl.h> // Standard Disclaimer applies > >_______________________________________________________ >Michal Zalewski * [lcamtuf(a)ags.pl] <=> [AGS WAN SYSADM] >[dione.ids.pl SYSADM] <-> [http://lcamtuf.na.export.pl] >[+48 22 551 45 93] [+48 603 110 160] bash$ :(){ :|:&};: >=-----=> God is real, unless declared integer. <=-----=

4 5

RE: Budtool v4.6.1 NDMP breaks with DOT 5.3.5P2
by Francett, Robert D (SAIC) 31 Mar '00

31 Mar '00

Our Veritas backups via NDMP also broke on reboot following upgrade to DOT5.3.5P2 from 5.3.4R2 (note Veritas not sanctioning 5.3.5anything yet to my knowledge). Got backups restarted, but not easy nor straight-forward. NetApp looking at it. We're downgrading to 5.3.4R3P2 per NetApp. Bob > ---------- > From: Jeff Krueger[SMTP:jkrueger@qualcomm.com] > Reply To: jkrueger(a)qualcomm.com > Sent: Friday, March 31, 2000 2:09 PM > To: Tim McDonough > Cc: Toaster > Subject: Re: Budtool v4.6.1 NDMP breaks with DOT 5.3.5P2 > > Tim: > > We're in the exact same situation. The first problem is that our tape > library device name is now called "mc0" instead of "spt0". This is > configured in the jbmgr_config file. > > Let us know if you figure this one out. We'll do the same if we figure > something out! =) > > -- Jeff > > > From Tim McDonough on Fri, 31 Mar 2000 13:00:54 PST: > > > >Howdy All: > > > > We were placed on the upgrade treadmill last week and > >upgraded > > our F740 cluster from v5.3.2D1 ( NDMP worked ) -- v5.3.4R3 > > -- v5.3.5P2 ( NDMP no worky ) > > > > Subsequently, our direct attached Budtool NDMP backups > >stopped > > working and timed out with each request. > > > > Anyone else seen this, solved this? > > > > Legato and NetApp are scratching their heads when they're not > > > > pointing their fingers at one another. > > > > Tim > > > > > > >

1 0

RE: Budtool v4.6.1 NDMP breaks with DOT 5.3.5P2
by Melvin, Grant 31 Mar '00

31 Mar '00

Hi Jeff, We've built in backwards compatibility, so that the jbmgr_config file can use "mco" or "spt0" even tough sysconfig -m only reports mc0. Hope this helps. Tim, I'd like to understand your issues with 5.3.5P2. Can you send me the case # logged with either NetApp or Legato support (or both)? Otherwise, please send me the symptoms. Cheers, Grant -----Original Message----- From: Jeff Krueger [mailto:jkrueger@qualcomm.com] Sent: Friday, March 31, 2000 3:10 PM To: Tim McDonough Cc: Toaster Subject: Re: Budtool v4.6.1 NDMP breaks with DOT 5.3.5P2 Tim: We're in the exact same situation. The first problem is that our tape library device name is now called "mc0" instead of "spt0". This is configured in the jbmgr_config file. Let us know if you figure this one out. We'll do the same if we figure something out! =) -- Jeff >From Tim McDonough on Fri, 31 Mar 2000 13:00:54 PST: > >Howdy All: > > We were placed on the upgrade treadmill last week and >upgraded > our F740 cluster from v5.3.2D1 ( NDMP worked ) -- v5.3.4R3 > -- v5.3.5P2 ( NDMP no worky ) > > Subsequently, our direct attached Budtool NDMP backups >stopped > working and timed out with each request. > > Anyone else seen this, solved this? > > Legato and NetApp are scratching their heads when they're not > > pointing their fingers at one another. > > Tim > > >

2 1

Budtool v4.6.1 NDMP breaks with DOT 5.3.5P2
by Tim McDonough 31 Mar '00

31 Mar '00

Howdy All: We were placed on the upgrade treadmill last week and upgraded our F740 cluster from v5.3.2D1 ( NDMP worked ) -- v5.3.4R3 -- v5.3.5P2 ( NDMP no worky ) Subsequently, our direct attached Budtool NDMP backups stopped working and timed out with each request. Anyone else seen this, solved this? Legato and NetApp are scratching their heads when they're not pointing their fingers at one another. Tim

3 2

RE: benchmark
by Paquette, Trevor 31 Mar '00

31 Mar '00

Oh brother.. asked and answered.. "Each vendor will have to be asked what method they use in the real world." The answer is "ask them.." Subject Closed. > -----Original Message----- > From: Jay Orr [SMTP:orrjl@stl.nexen.com] > Sent: Friday, March 31, 2000 7:30 AM > To: Paquette, Trevor > Cc: Karl Swartz; toasters(a)mathworks.com > Subject: RE: benchmark > > Yes, and THAT WAS MY ORIGINAL QUESTION THAT YOU RESPONDED TO - what do > Netapps and Sun do? > > > On Thu, 30 Mar 2000, Paquette, Trevor wrote: > > Each vendor either sticks with powers of 2 (as they should) or they > round > > it. > > Each vendor will have to be asked what method they use in the real > world. > > > > > > > > Yes, we all know the theory, but I'm talking real world. As the > > > thread reads, some vendors use 1000 KB = 1MB, while some don't. I am > > > asking what is implemented. Do the majority of the unix vendors adhere > to > > > the proper base 2 theory of metrics or do they use base 10 combos > > > (i.e. the 1GB = 1000 MB)? > > > > > > On Thu, 30 Mar 2000, Paquette, Trevor wrote: > > > > > > > Here we go.. > > > > > > > > A bit is the smallest unit of data in a computer; has a single > binary > > > value > > > > of either 0 or 1 > > > > A byte is a unit of information that is eight bits long. (half a > byte, > > > or 4 > > > > bits is called a nibble) > > > > > > > > Computer storage is measured in byte multiples; which are based on > > > powers of > > > > 2. > > > > Because humans deal mainly with the decimal system, the resulting > number > > > is > > > > usually 'rounded off' > > > > for humans to better comprehend. > > > > > > > > 1 Kbyte = 1024 bytes = 2^10 bytes = 1,024 > bytes > > > > 1 Megabyte = 1024 Kbytes = 2^20 bytes = 1,048,576 > bytes > > > > 1 Gigabyte = 1024 Mbytes = 2^30 bytes = 1,073,741,824 > bytes > > > > 1 Terabyte = 1024 Gbytes = 2^40 bytes = 1,099,511,627,776 > bytes > > > > 1 Petabyte = 1024 Tbytes = 2^50 bytes = 1,125,899,906,842,624 > bytes > > > > 1 Exabyte = 1024 Pbytes = 2^60 bytes = 1,152,921,504,606,846,976 > bytes > > > > = approx. 1 billion GB > > > > > > Humm... Fun.. Do you know off hand how Solaris and netapp defines > a MB > > > or > > > > > GB? In moving around files I started thinking about this, since > > > depending > > > > > on what tool you use you get file sizes in "bytes" "kbytes" "MB" > or > > > "GB" > > > > > (a la gnu df, solaris df, etc...) > > > > > > > > > > On Wed, 29 Mar 2000, Karl Swartz wrote: > > > > > > > > > > > > Uh, if I can ask a simple (stupid?) question - I was looking > at > > > the > > > > > source > > > > > > > code for postmark and noticed this : > > > > > > > > > > > > > > #define KILOBYTE 1024 > > > > > > > #define MEGABYTE (1000*KILOBYTE) > > > > > > > > > > > > > > Isn't a megabite 1024*1024? > > > > > > > > > > > > It depends on who you ask. Long ago, disk vendors decided that > they > > > > > > could make their disks look bigger if they defined MB to be 2000 > > > > > > sectors (assuming 512 byte sectors, or 8000 sectors for 128 byte > > > > > > sectors), and GB became 1000 MB. Of course they also talked > about > > > > > > unformatted capacity to further inflate the size over reality. > > > > > > > > > > > > In a similar vein, telco people talk about 1K = 1000. The 64K > B- > > > > > > channel in an ISDN line is 64000 bits per second, not 65536. > > > > > > > > > > > > I'm not sure about networking, but I think FDDI and 100base-T > both > > > > > > have a data rate of 100 * 10^6 bits/second, not 100 * 2^20. > (Not to > > > be > > > > > > confused with the signalling rate, which as I recall is 125 * > 10^6 > > > for > > > > > > FDDI with some flavor of 4/5 encoding, and possibly something > > > similar > > > > > > for 100base-T.) > > > > > > > > > > > > What's right for Postmark? Good question. It looks like Jeff > > > picked > > > > > > the standard for disk sizes, but it's not clear that's the > standard > > > > > > most applicable to this case. Aren't standards wonderful? :-) > > > > > > > > > > > > -- > > > > > > Karl Swartz Network Appliance Engineering > > > > > > Work: kls(a)netapp.com http://www.netapp.com/ > > > > > > Home: kls(a)chicago.com http://www.chicago.com/~kls/ > > > > > > > > > > > > > > > > ----------- > > > > > Jay Orr > > > > > Systems Administrator > > > > > Fujitsu Nexion Inc. > > > > > St. Louis, MO > > > > > > > > > > ----------- > > > Jay Orr > > > Systems Administrator > > > Fujitsu Nexion Inc. > > > St. Louis, MO > > > > ----------- > Jay Orr > Systems Administrator > Fujitsu Nexion Inc. > St. Louis, MO

4 3

RE: benchmark
by Paquette, Trevor 31 Mar '00

31 Mar '00

Common decency is taking this to private email.. which I am now going to do. Sorry to bother the rest of this list with crap.. > -----Original Message----- > From: Jay Orr [SMTP:orrjl@stl.nexen.com] > Sent: Friday, March 31, 2000 8:48 AM > To: Paquette, Trevor > Cc: toasters(a)mathworks.com > Subject: RE: benchmark > > > By restating someone's question as an answer, it is not "asked and > answered". I had put to this forum the question of what does Netapps and > Sun do, since there are many Netapp people on this and many people who use > Sun here. > > Your answers have been less then helpful - If you don't know then why did > you reply to the original posting? It was a SIMPLE QUESTION, I didn't ask > for the theory of powers of two, I didn't ask "who should I ask", I didn't > say "how does Trevor feel about this?" I simply posted to this informative > list a simple question of WHAT DOES NETAPP and SUN do in formulating what > they consider MB and GB. > > Please, if you have nothing useful to add to a discussion, don't > interject. That's just common decency... > > > On Fri, 31 Mar 2000, Paquette, Trevor wrote: > > > Oh brother.. > > > > asked and answered.. > > > > "Each vendor will have to be asked what method they use in the real > world." > > > > The answer is "ask them.." > > > > Subject Closed. > > > > > -----Original Message----- > > > From: Jay Orr [SMTP:orrjl@stl.nexen.com] > > > Sent: Friday, March 31, 2000 7:30 AM > > > To: Paquette, Trevor > > > Cc: Karl Swartz; toasters(a)mathworks.com > > > Subject: RE: benchmark > > > > > > Yes, and THAT WAS MY ORIGINAL QUESTION THAT YOU RESPONDED TO - what > do > > > Netapps and Sun do? > > > > > > > > > On Thu, 30 Mar 2000, Paquette, Trevor wrote: > > > > Each vendor either sticks with powers of 2 (as they should) or they > > > round > > > > it. > > > > Each vendor will have to be asked what method they use in the real > > > world. > > > > > > > > > > > > > > Yes, we all know the theory, but I'm talking real world. As the > > > > > thread reads, some vendors use 1000 KB = 1MB, while some don't. I > am > > > > > asking what is implemented. Do the majority of the unix vendors > adhere > > > to > > > > > the proper base 2 theory of metrics or do they use base 10 combos > > > > > (i.e. the 1GB = 1000 MB)? > > > > > > > > > > On Thu, 30 Mar 2000, Paquette, Trevor wrote: > > > > > > > > > > > Here we go.. > > > > > > > > > > > > A bit is the smallest unit of data in a computer; has a single > > > binary > > > > > value > > > > > > of either 0 or 1 > > > > > > A byte is a unit of information that is eight bits long. (half a > > > byte, > > > > > or 4 > > > > > > bits is called a nibble) > > > > > > > > > > > > Computer storage is measured in byte multiples; which are based > on > > > > > powers of > > > > > > 2. > > > > > > Because humans deal mainly with the decimal system, the > resulting > > > number > > > > > is > > > > > > usually 'rounded off' > > > > > > for humans to better comprehend. > > > > > > > > > > > > 1 Kbyte = 1024 bytes = 2^10 bytes = > 1,024 > > > bytes > > > > > > 1 Megabyte = 1024 Kbytes = 2^20 bytes = > 1,048,576 > > > bytes > > > > > > 1 Gigabyte = 1024 Mbytes = 2^30 bytes = > 1,073,741,824 > > > bytes > > > > > > 1 Terabyte = 1024 Gbytes = 2^40 bytes = > 1,099,511,627,776 > > > bytes > > > > > > 1 Petabyte = 1024 Tbytes = 2^50 bytes = > 1,125,899,906,842,624 > > > bytes > > > > > > 1 Exabyte = 1024 Pbytes = 2^60 bytes = > 1,152,921,504,606,846,976 > > > bytes > > > > > > = approx. 1 billion GB > > > > > > > > > > Humm... Fun.. Do you know off hand how Solaris and netapp > defines > > > a MB > > > > > or > > > > > > > GB? In moving around files I started thinking about this, > since > > > > > depending > > > > > > > on what tool you use you get file sizes in "bytes" "kbytes" > "MB" > > > or > > > > > "GB" > > > > > > > (a la gnu df, solaris df, etc...) > > > > > > > > > > > > > > On Wed, 29 Mar 2000, Karl Swartz wrote: > > > > > > > > > > > > > > > > Uh, if I can ask a simple (stupid?) question - I was > looking > > > at > > > > > the > > > > > > > source > > > > > > > > > code for postmark and noticed this : > > > > > > > > > > > > > > > > > > #define KILOBYTE 1024 > > > > > > > > > #define MEGABYTE (1000*KILOBYTE) > > > > > > > > > > > > > > > > > > Isn't a megabite 1024*1024? > > > > > > > > > > > > > > > > It depends on who you ask. Long ago, disk vendors decided > that > > > they > > > > > > > > could make their disks look bigger if they defined MB to be > 2000 > > > > > > > > sectors (assuming 512 byte sectors, or 8000 sectors for 128 > byte > > > > > > > > sectors), and GB became 1000 MB. Of course they also talked > > > about > > > > > > > > unformatted capacity to further inflate the size over > reality. > > > > > > > > > > > > > > > > In a similar vein, telco people talk about 1K = 1000. The > 64K > > > B- > > > > > > > > channel in an ISDN line is 64000 bits per second, not 65536. > > > > > > > > > > > > > > > > I'm not sure about networking, but I think FDDI and > 100base-T > > > both > > > > > > > > have a data rate of 100 * 10^6 bits/second, not 100 * 2^20. > > > (Not to > > > > > be > > > > > > > > confused with the signalling rate, which as I recall is 125 > * > > > 10^6 > > > > > for > > > > > > > > FDDI with some flavor of 4/5 encoding, and possibly > something > > > > > similar > > > > > > > > for 100base-T.) > > > > > > > > > > > > > > > > What's right for Postmark? Good question. It looks like > Jeff > > > > > picked > > > > > > > > the standard for disk sizes, but it's not clear that's the > > > standard > > > > > > > > most applicable to this case. Aren't standards wonderful? > :-) > > > > > > > > > > > > > > > > -- > > > > > > > > Karl Swartz Network Appliance Engineering > > > > > > > > Work: kls(a)netapp.com http://www.netapp.com/ > > > > > > > > Home: kls(a)chicago.com http://www.chicago.com/~kls/ > > > > > > > > > > > > > > > > > > > > > > ----------- > > > > > > > Jay Orr > > > > > > > Systems Administrator > > > > > > > Fujitsu Nexion Inc. > > > > > > > St. Louis, MO > > > > > > > > > > > > > > > > ----------- > > > > > Jay Orr > > > > > Systems Administrator > > > > > Fujitsu Nexion Inc. > > > > > St. Louis, MO > > > > > > > > > > ----------- > > > Jay Orr > > > Systems Administrator > > > Fujitsu Nexion Inc. > > > St. Louis, MO > > > > ----------- > Jay Orr > Systems Administrator > Fujitsu Nexion Inc. > St. Louis, MO

2 1

How are you mounting CIFS homedirs with Win9x?
by Timothy Demarest 31 Mar '00

31 Mar '00

I was wondering how everyone is mounting CIFS file systems from Windows 9x clients? Having worked through the NIS netgroup issue (Bug #19922, patched in 5.3.4R3P2 -- thanks everyone), I'm moving on to getting the new filer working with PCs. I have a CIFS homedir for each user. The filer shares this directory so that it appears as the user name. In a Windows NT login script, I can use the "%username%" directive to mount the user's home dir as H:. However, you can't use "%username%" with Windows 9x clients. One option is to set the home directory in the user's environment profile (in User Manager), but this requires changing each user's profile. The logon script is simple and elegant. I'd really like to mount the CIFS homedir directly, rather than having the user open the drive, then open the folder with their associated username in order to access files in their home directory. Any hints? Tim

1 0

RE: benchmark
by Paquette, Trevor 31 Mar '00

31 Mar '00

Each vendor either sticks with powers of 2 (as they should) or they round it. Each vendor will have to be asked what method they use in the real world. > -----Original Message----- > From: Jay Orr [SMTP:orrjl@stl.nexen.com] > Sent: Thursday, March 30, 2000 1:53 PM > To: Paquette, Trevor > Cc: Karl Swartz; toasters(a)mathworks.com > Subject: RE: benchmark > > > Yes, we all know the theory, but I'm talking real world. As the > thread reads, some vendors use 1000 KB = 1MB, while some don't. I am > asking what is implemented. Do the majority of the unix vendors adhere to > the proper base 2 theory of metrics or do they use base 10 combos > (i.e. the 1GB = 1000 MB)? > > On Thu, 30 Mar 2000, Paquette, Trevor wrote: > > > Here we go.. > > > > A bit is the smallest unit of data in a computer; has a single binary > value > > of either 0 or 1 > > A byte is a unit of information that is eight bits long. (half a byte, > or 4 > > bits is called a nibble) > > > > Computer storage is measured in byte multiples; which are based on > powers of > > 2. > > Because humans deal mainly with the decimal system, the resulting number > is > > usually 'rounded off' > > for humans to better comprehend. > > > > 1 Kbyte = 1024 bytes = 2^10 bytes = 1,024 bytes > > 1 Megabyte = 1024 Kbytes = 2^20 bytes = 1,048,576 bytes > > 1 Gigabyte = 1024 Mbytes = 2^30 bytes = 1,073,741,824 bytes > > 1 Terabyte = 1024 Gbytes = 2^40 bytes = 1,099,511,627,776 bytes > > 1 Petabyte = 1024 Tbytes = 2^50 bytes = 1,125,899,906,842,624 bytes > > 1 Exabyte = 1024 Pbytes = 2^60 bytes = 1,152,921,504,606,846,976 bytes > = > > approx. 1 billion GB > > > > > > > > > -----Original Message----- > > > From: Jay Orr [SMTP:orrjl@stl.nexen.com] > > > Sent: Thursday, March 30, 2000 10:52 AM > > > To: Karl Swartz > > > Cc: toasters(a)mathworks.com > > > Subject: Re: benchmark > > > > > > > > > Humm... Fun.. Do you know off hand how Solaris and netapp defines a MB > or > > > GB? In moving around files I started thinking about this, since > depending > > > on what tool you use you get file sizes in "bytes" "kbytes" "MB" or > "GB" > > > (a la gnu df, solaris df, etc...) > > > > > > On Wed, 29 Mar 2000, Karl Swartz wrote: > > > > > > > > Uh, if I can ask a simple (stupid?) question - I was looking at > the > > > source > > > > > code for postmark and noticed this : > > > > > > > > > > #define KILOBYTE 1024 > > > > > #define MEGABYTE (1000*KILOBYTE) > > > > > > > > > > Isn't a megabite 1024*1024? > > > > > > > > It depends on who you ask. Long ago, disk vendors decided that they > > > > could make their disks look bigger if they defined MB to be 2000 > > > > sectors (assuming 512 byte sectors, or 8000 sectors for 128 byte > > > > sectors), and GB became 1000 MB. Of course they also talked about > > > > unformatted capacity to further inflate the size over reality. > > > > > > > > In a similar vein, telco people talk about 1K = 1000. The 64K B- > > > > channel in an ISDN line is 64000 bits per second, not 65536. > > > > > > > > I'm not sure about networking, but I think FDDI and 100base-T both > > > > have a data rate of 100 * 10^6 bits/second, not 100 * 2^20. (Not to > be > > > > confused with the signalling rate, which as I recall is 125 * 10^6 > for > > > > FDDI with some flavor of 4/5 encoding, and possibly something > similar > > > > for 100base-T.) > > > > > > > > What's right for Postmark? Good question. It looks like Jeff > picked > > > > the standard for disk sizes, but it's not clear that's the standard > > > > most applicable to this case. Aren't standards wonderful? :-) > > > > > > > > -- > > > > Karl Swartz Network Appliance Engineering > > > > Work: kls(a)netapp.com http://www.netapp.com/ > > > > Home: kls(a)chicago.com http://www.chicago.com/~kls/ > > > > > > > > > > ----------- > > > Jay Orr > > > Systems Administrator > > > Fujitsu Nexion Inc. > > > St. Louis, MO > > > > ----------- > Jay Orr > Systems Administrator > Fujitsu Nexion Inc. > St. Louis, MO

3 2

Re: nas storage
by bar-lev benjamin 31 Mar '00

31 Mar '00

We did not really get an answer to the question raised. I am curious too as I am currently evaluating SANergy and I am very impressed with what I see. __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com

1 0

Creating a RAID
by July at Zerowait 31 Mar '00

31 Mar '00

Hello: The filer in question is an inherited box, an F760 with two shelves of 9GB SCSI drives, one shelf of 9GB FC-AL drives all in one RAID and one VOL. The single VOL also contains the root, I think. I've heard that you are supposed to separate SCSI and FC-AL drives into separate RAID groups, but this is how the box came and we can't change it. Anyway, my task is to add two shelves of 36GB drives to this unit. I know I need to create a separate RAID group for the 36GB drives. In the man pages I see instructions for adding disks to an existing volume and how to create a volume. I don't see anything that tells me how to create a new RAID group. I must be missing something obvious. Can someone help? Thank you. July ---------------------------------------------------------------- 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 -------------------------------------------------------------------

6 7

RE: benchmark
by Paquette, Trevor 30 Mar '00

30 Mar '00

Here we go.. A bit is the smallest unit of data in a computer; has a single binary value of either 0 or 1 A byte is a unit of information that is eight bits long. (half a byte, or 4 bits is called a nibble) Computer storage is measured in byte multiples; which are based on powers of 2. Because humans deal mainly with the decimal system, the resulting number is usually 'rounded off' for humans to better comprehend. 1 Kbyte = 1024 bytes = 2^10 bytes = 1,024 bytes 1 Megabyte = 1024 Kbytes = 2^20 bytes = 1,048,576 bytes 1 Gigabyte = 1024 Mbytes = 2^30 bytes = 1,073,741,824 bytes 1 Terabyte = 1024 Gbytes = 2^40 bytes = 1,099,511,627,776 bytes 1 Petabyte = 1024 Tbytes = 2^50 bytes = 1,125,899,906,842,624 bytes 1 Exabyte = 1024 Pbytes = 2^60 bytes = 1,152,921,504,606,846,976 bytes = approx. 1 billion GB > -----Original Message----- > From: Jay Orr [SMTP:orrjl@stl.nexen.com] > Sent: Thursday, March 30, 2000 10:52 AM > To: Karl Swartz > Cc: toasters(a)mathworks.com > Subject: Re: benchmark > > > Humm... Fun.. Do you know off hand how Solaris and netapp defines a MB or > GB? In moving around files I started thinking about this, since depending > on what tool you use you get file sizes in "bytes" "kbytes" "MB" or "GB" > (a la gnu df, solaris df, etc...) > > On Wed, 29 Mar 2000, Karl Swartz wrote: > > > > Uh, if I can ask a simple (stupid?) question - I was looking at the > source > > > code for postmark and noticed this : > > > > > > #define KILOBYTE 1024 > > > #define MEGABYTE (1000*KILOBYTE) > > > > > > Isn't a megabite 1024*1024? > > > > It depends on who you ask. Long ago, disk vendors decided that they > > could make their disks look bigger if they defined MB to be 2000 > > sectors (assuming 512 byte sectors, or 8000 sectors for 128 byte > > sectors), and GB became 1000 MB. Of course they also talked about > > unformatted capacity to further inflate the size over reality. > > > > In a similar vein, telco people talk about 1K = 1000. The 64K B- > > channel in an ISDN line is 64000 bits per second, not 65536. > > > > I'm not sure about networking, but I think FDDI and 100base-T both > > have a data rate of 100 * 10^6 bits/second, not 100 * 2^20. (Not to be > > confused with the signalling rate, which as I recall is 125 * 10^6 for > > FDDI with some flavor of 4/5 encoding, and possibly something similar > > for 100base-T.) > > > > What's right for Postmark? Good question. It looks like Jeff picked > > the standard for disk sizes, but it's not clear that's the standard > > most applicable to this case. Aren't standards wonderful? :-) > > > > -- > > Karl Swartz Network Appliance Engineering > > Work: kls(a)netapp.com http://www.netapp.com/ > > Home: kls(a)chicago.com http://www.chicago.com/~kls/ > > > > ----------- > Jay Orr > Systems Administrator > Fujitsu Nexion Inc. > St. Louis, MO

2 1

RE: NOW website
by Davis, Barry 30 Mar '00

30 Mar '00

We've been making lots of changes lately. I thought you might be interested to hear about them. We've added a natural language search engine, chat capability, case notification, release comparison tool, newsletters, etc....Check out the "Headlines" section at http://now.netapp.com/NOW/products/head/ for more info. P.S. Look for the "..." links at the end of each description for the full details. Yes, we will be changing that also. ;-) --------------------------------------- Barry Davis - Director of eService Network Appliance, Inc. Email: barry.davis(a)netapp.com Get answers NOW! - NetApp On the Web - http://now.netapp.com > -----Original Message----- > From: Jay Orr [mailto:orrjl@stl.nexen.com] > Sent: Thursday, March 23, 2000 8:28 AM > To: toasters(a)mathworks.com > Subject: NOW website > > > Dunno if anyone had noticed, but as of the 11th they got rid of > JavaScript. > > I just logged in with LYNX and it let me. Still gives lots of cookies, > though. I'm going to check it out some more, but nice to see > they finally > took care of this. > > > ----------- > Jay Orr > Systems Administrator > Fujitsu Nexion Inc. > St. Louis, MO >

1 0

RE: NetApp and HP MC Service Guard
by Dave Mills 30 Mar '00

30 Mar '00

Alan, I have installed and config'd Service Guard a few times. The data disks are shared between systems with an exclusive "lock" which guarantees that only 1 node can access the shared disks at a time. The locking mechanism is LVM based. Specifically, only certified HP disks support the locking. Since the Netapps is an NFS Server, it is outside the scope of LVM and would not play well in a SG environment. NFS locks are 'advisory' - my process hopes your process adheres to pre-extablished rules. Netapps has cluster configs which allow for redundancy of the filer and its DASD. This alone would be redundancy of DASD, where SG attempts to eliminate ALL single points of failure. Another potential issue. NFS clients (HP-UX included) are not usually happy when the NFS server hangs, panics,... I worked on a SG environment where each node in the cluster was both a NFS client and NFS server to the other nodes. This sort of became a nightmare to implement. HP has an NFS Toolkit for Service Guard which supports ONLY an NFS server, not the client. On the other hand (the non-supported hand ;>) ), if you configured SG properly, the adoptive node (adopting the application during failover) would do an NFS mount to the filer and be able to access the data, if it were not corrupted. If you do decide to go ahead with this unsupported environment, I would be very interested in hearing the results. Dave Mills Cypress Systems -----Original Message----- From: Alan Beauchemin [mailto:Alan.Beauchemin.B@bayer.com] Sent: Thursday, March 30, 2000 8:22 AM To: - *toasters(a)mathworks.com Subject: NetApp and HP MC Service Guard I have a project involving four HP systems requiring continuous availability. I would like to use MC Service Guard but I am not sure if I can use this software with a NetApp. Is anyone using this failover software in a NetApp / HP environment? I would also be curious of other methods of high availability from the server side with a NetApp? Thanks Alan

1 0

RE: rc registry registry rc
by Thompson, Tim 30 Mar '00

30 Mar '00

> All options and variables, persistent or not, should be > documented and modifiable via plain-text files. > ... > I decided to keep a "master" file under SCCS control, and run The desire to have all configuration aspects accessible in a single text-based way - which can make change control easier (something we've heard previously on toasters) and which can make it easier to implement management applications - is an explicit goal of the registry mechanism. The current implementation does not manage all configuration aspects, so a user interface to it is perhaps premature. We used it to implement a few features (automatically-persistent options, saving of DHCP info, easier creation of filerview screens, etc.) in the last few releases. But it's not quite ready to be the user-accessible-all-configuration-in-one-place mechanism, so there is no user interface to it, yet. ...Tim Thompson...Network Appliance, Inc...tjt(a)netapp.com...

1 0

benchmark
by Rainchik, Aleksandr (MED, Non GE) 30 Mar '00

30 Mar '00

Hi, I was using Postmark to measure NFS performance in Solaris <-> NetApp configuration. Very good tool, I think. http://www.netapp.com/tech_library/3022.html http://www.netapp.com/tech_library/postmark.html I even ported it to Windows NT :) But, I think there should be something more Windows-native. I just want to measure remote disk performance, NFS or CIFS connected. NetBench seems to be to compicated... Any suggestions? Thank you.

5 5

NetApp and HP MC Service Guard
by Alan Beauchemin 30 Mar '00

30 Mar '00

I have a project involving four HP systems requiring continuous availability. I would like to use MC Service Guard but I am not sure if I can use this software with a NetApp. Is anyone using this failover software in a NetApp / HP environment? I would also be curious of other methods of high availability from the server side with a NetApp? Thanks Alan

1 0

RE: rc registry registry rc
by Thompson, Tim 30 Mar '00

30 Mar '00

> we get these messages on reboot of one netapp > ** Option wafl.default_unix_user is being set to "pcuser" in /etc/rc, > ... > ** Option wafl.default_unix_user is being set to "root" in /etc/rc, > anyone know what causes such schizo behaviour? I'm getting additional info from John, offline, to attempt to see what's going on. (Besides the obvious action of opening a trouble ticket on the issue.) > wher is a netapp's registry? It's a mechanism introduced in 5.3 for making options and other things persistent. It should be invisible, for the most part. ...Tim Thompson...Network Appliance, Inc...tjt(a)netapp.com...

3 2

Server capacity and performance
by Brian Hostetter 30 Mar '00

30 Mar '00

Hello, I am running into a problem where I am getting NFS not responding errors from our f740 (512M ram, 32M nvram). After the error pops up it is immediatly available again, but performance is getting really bad on our filer. My question is does the amout of disk space used effect filer performance substantially? Yesterday we got to 96% full (2x18G shelves). I trimmed it down as much as possible, but am still seeing horrible performance. I have ordered a new 36G shelf, but that may be a few weeks away and I would like to find a stop gap solution. Could my performance problems be related to space, or are they probably just sheer volume of ops problems. TIA, --Brian

4 3

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