toasters

toasters@lists.teaparty.net

13531 discussions

2 volumes or 1
by Pesce, Nicholas (FUSA) 12 Sep '00

12 Sep '00

Gentlemen, I need your opinion on a creation I am about to unleash on our unsuspecting production site where I work. I currently have a 740 filer with 2 shelves and 7 disks. 3 disks are in the first shelf and configured for volume 0 root volume 4 disks are inthe second shelf and configured for volume 1 Now, since I didn't put this box together. I was wondering if it is better to make all of these disks one volume. There are heavy reads and writes to the volume1, but I don't know if having a heavily used volume combined with root functions will kill my performance. Besides this will make a larger volume for more distributed writes (8 total disks versus 4). Also I will be saving 1 disk of data space because I won't need the second parity disk. Addition information is that they are fibre channel shelves (of course). And that the data is not important (washed away daily). Assuming that the root volume functions do not critically affect my box (and I don't believe that they do, as most of it is running from memory) I believe my choice is correct. Just wondering if you guys agree with this hypothesis or not. Nicholas Pesce Internet Production Services Phone: 847 488-6384 Pager: 888 785 3455 Fax: 847 488 3434

4 5

RE: [toasters] Problems with quotas (or Hurrah ! my first core du mp )
by Fox, Adam 12 Sep '00

12 Sep '00

This looks like a bug in 5.3.6R1. If you do a user level quota and the user is not in the /etc/passwd file or NIS map, it panics the filer at the quota on. Worse, it keeps panicing at each reboot since part of the reboot sequence is to initialize the quotas. To fix. Boot floppy, then boot without rc (option 2). You'll need to upgrade ONTAP to 5.3.6R2 which has the fix. This one seemed to creep in right at 5.3.6R1 and was fixed with anything after that. Previous released do not seem to have the problem. Hope this helps. -- Adam Fox NetApp Professional Services, NC adamfox(a)netapp.com > -----Original Message----- > From: David Schairer [mailto:njal@concentric.net] > Sent: Tuesday, September 12, 2000 3:08 PM > To: kubek > Cc: toasters(a)mathworks.com > Subject: Re: [toasters] Problems with quotas (or Hurrah ! my > first core > dump ) > > > Boot via floppy, remove the /etc/quotas file. I'm 99% sure that > quota initialization will abort with a warning if it's not there. > > --DRS > > On Tue, Sep 12, 2000 at 08:48:37PM +0200, kubek wrote: > > Hi, > > I just populated our F720 but noticed that I forgot some > > "quota off -- quota on" > > command sequence in order to enable the last > modification i did > > on quotas... > > > > ... I'm still waiting for the output of the "quota on" > command as > > since then the filer is cyclicly trying to dump core (see > > screenshot at the end of this message) : > > > > After each boot the filer notices that the "quota on" > command was > > interrupted by a reboot, it tries to initialize the quotas > > and then ... reboot > > > > Does someone on this list know how to break this cycle ? Either > > by modifying the quota file (throught the floppy boot menu) or > > by forcing a reboot *without* quota initialization ? > > > > Thanks, > > Jean-Marie > > > > ----- Screenshot of the error message > > > > NetApp Release 5.3.6R1: Wed Jun 14 18:21:14 PDT 2000 > > Copyright (c) 1992-2000 Network Appliance, Inc. > > Starting boot on Tue Sep 12 17:49:29 GMT 2000 > > Scanning for disk drives: .......Tue Sep 12 17:49:51 GMT > [de0]: Ethernet e0: Link up. > > > > Configuring disk drives: 2.5 2.1 2.6 2.2 2.0 2.3 2.4 > > Disk 2.0 is reserved for "hot spare" > > > > 1 disk is reserved for "hot spare". > > Restoring parity from NVRAM > > Loading volume vol0 > > Tue Sep 12 17:49:55 GMT [rc]: wafl: Reboot interrupted > quota initialization. > > Will attempt to reinitialize quotas on volume vol0 > after WAFL has started. > > Replaying WAFL log > > > > PANIC: ../common/wafl/quota.c:1863: Assertion failure. on > release NetApp Release 5.3.6R1 on Tue Sep 12 17:49:57 2000 > > > > version: NetApp Release 5.3.6R1: Wed Jun 14 18:21:14 PDT 2000 > > cc flags: 1 > > dumping core: ......Old core present on disk --- not dumped. > > >

1 0

[toasters] Problems with quotas (or Hurrah ! my first core dump )
by kubek 12 Sep '00

12 Sep '00

Hi, I just populated our F720 but noticed that I forgot some "quota off -- quota on" command sequence in order to enable the last modification i did on quotas... ... I'm still waiting for the output of the "quota on" command as since then the filer is cyclicly trying to dump core (see screenshot at the end of this message) : After each boot the filer notices that the "quota on" command was interrupted by a reboot, it tries to initialize the quotas and then ... reboot Does someone on this list know how to break this cycle ? Either by modifying the quota file (throught the floppy boot menu) or by forcing a reboot *without* quota initialization ? Thanks, Jean-Marie ----- Screenshot of the error message NetApp Release 5.3.6R1: Wed Jun 14 18:21:14 PDT 2000 Copyright (c) 1992-2000 Network Appliance, Inc. Starting boot on Tue Sep 12 17:49:29 GMT 2000 Scanning for disk drives: .......Tue Sep 12 17:49:51 GMT [de0]: Ethernet e0: Link up. Configuring disk drives: 2.5 2.1 2.6 2.2 2.0 2.3 2.4 Disk 2.0 is reserved for "hot spare" 1 disk is reserved for "hot spare". Restoring parity from NVRAM Loading volume vol0 Tue Sep 12 17:49:55 GMT [rc]: wafl: Reboot interrupted quota initialization. Will attempt to reinitialize quotas on volume vol0 after WAFL has started. Replaying WAFL log PANIC: ../common/wafl/quota.c:1863: Assertion failure. on release NetApp Release 5.3.6R1 on Tue Sep 12 17:49:57 2000 version: NetApp Release 5.3.6R1: Wed Jun 14 18:21:14 PDT 2000 cc flags: 1 dumping core: ......Old core present on disk --- not dumped.

2 4

RE: [toasters] Problems with quotas (or Hurrah ! my first core du mp )
by Pettit, Simon 12 Sep '00

12 Sep '00

Please check bug number 26472 This is probably caused by an error in the quotas file. this is fixed in 5.3.6R2 Quick suggestion is boot from floppy upgrade to 5.3.6R2 and find the error in quotas file!! Please check with support on this if you have a case open!! Regards Simon P. -----Original Message----- From: David Schairer [mailto:njal@concentric.net] Sent: 12 September 2000 20:08 To: kubek Cc: toasters(a)mathworks.com Subject: Re: [toasters] Problems with quotas (or Hurrah ! my first core dump ) Boot via floppy, remove the /etc/quotas file. I'm 99% sure that quota initialization will abort with a warning if it's not there. --DRS On Tue, Sep 12, 2000 at 08:48:37PM +0200, kubek wrote: > Hi, > I just populated our F720 but noticed that I forgot some > "quota off -- quota on" > command sequence in order to enable the last modification i did > on quotas... > > ... I'm still waiting for the output of the "quota on" command as > since then the filer is cyclicly trying to dump core (see > screenshot at the end of this message) : > > After each boot the filer notices that the "quota on" command was > interrupted by a reboot, it tries to initialize the quotas > and then ... reboot > > Does someone on this list know how to break this cycle ? Either > by modifying the quota file (throught the floppy boot menu) or > by forcing a reboot *without* quota initialization ? > > Thanks, > Jean-Marie > > ----- Screenshot of the error message > > NetApp Release 5.3.6R1: Wed Jun 14 18:21:14 PDT 2000 > Copyright (c) 1992-2000 Network Appliance, Inc. > Starting boot on Tue Sep 12 17:49:29 GMT 2000 > Scanning for disk drives: .......Tue Sep 12 17:49:51 GMT [de0]: Ethernet e0: Link up. > > Configuring disk drives: 2.5 2.1 2.6 2.2 2.0 2.3 2.4 > Disk 2.0 is reserved for "hot spare" > > 1 disk is reserved for "hot spare". > Restoring parity from NVRAM > Loading volume vol0 > Tue Sep 12 17:49:55 GMT [rc]: wafl: Reboot interrupted quota initialization. > Will attempt to reinitialize quotas on volume vol0 after WAFL has started. > Replaying WAFL log > > PANIC: ../common/wafl/quota.c:1863: Assertion failure. on release NetApp Release 5.3.6R1 on Tue Sep 12 17:49:57 2000 > > version: NetApp Release 5.3.6R1: Wed Jun 14 18:21:14 PDT 2000 > cc flags: 1 > dumping core: ......Old core present on disk --- not dumped. >

1 0

Using Omniback for CIFS backups
by Clarke, John L. (CIS) 12 Sep '00

12 Sep '00

> Hello, > > I am looking for anyone who is using Omniback II to back up a Network > Appliance via CIFS. We have an F740 mounted on the tape backup server as > shared disks and we are finding that Omniback does not want to back up all > the directories on the filer. > > Has anyone done this successfully? > > Thanks > John

1 0

comments on netapp filers cifs performance
by neil lehrer 12 Sep '00

12 Sep '00

hi, i'd appreciate hearing what you think of the cifs performance on netapp filers. i am considering a f760, our wkstations are primarily nt 4.0 sp5. thanks --

2 2

RE: AIT libraries
by Premanshu Jain 12 Sep '00

12 Sep '00

Yes Qualstar's are really good. We are implementing them now. We could dump 105 GB (Unix format data) on a AIT2 tape. This compression depends on the type of data you are backing up. The CIFS data blocks dumped only upto 60-70 GB per tape. -----Original Message----- From: Larry Goodman [mailto:Larry.Goodman@Intense3D.com] Sent: Tuesday, September 12, 2000 6:45 AM To: toasters(a)mathworks.com Subject: AIT libraries If any of you kind toaster'ites have experience with AIT2 tape libraries, I'd like to here your advice about the different brands available. I've been researching these devices and they all seem to claim pretty much the same specs (duh, they're all using the same Sony drive!). I'm particularly interested in the durability of the robotics. I've been looking at libraries from Spectra Logic, Qualstar, Adic and Overland Data, usually providing from 12 to 24 slots (single drive). The Qualstar folks claim that they're the only supplier that uses a "lead-screw" design instead of belts for better reliability. Any comments? Thanks, Larry.Goodman(a)Intense3d.com

1 0

RE: 2 volumes or 1
by Walters, Mike 11 Sep '00

11 Sep '00

You don't mention what size drives they are (9/18/36GB). As a rule, however, I would suggest that you are correct in thinking you would be better off combining disks to make a single volume. You can use qtrees to segragate data (& give more control if some is NFS & some CIFS, for example). I believe that your 'heavy reads & writes' will benefit from the larger raid group. You would obviously need to destroy vol1 before adding disks to vol0. Just one minor point - I presume you do have a spare disk as well - if not, & you have a disk fail, the filer will squawk at hourly intervals complaining, then if you don't add a drive, shutdown after 24 hours (options raid.timeout defaults to 24 [hours] to configure this) on the basis that you don't want to run degraded for any significant time. Cheers Mike UK SE -----Original Message----- From: Pesce, Nicholas (FUSA) [mailto:Nicholaspesce@firstusa.com] Sent: 11 September 2000 21:30 To: 'toasters(a)mathworks.com' Subject: 2 volumes or 1 Gentlemen, I need your opinion on a creation I am about to unleash on our unsuspecting production site where I work. I currently have a 740 filer with 2 shelves and 7 disks. 3 disks are in the first shelf and configured for volume 0 root volume 4 disks are inthe second shelf and configured for volume 1 Now, since I didn't put this box together. I was wondering if it is better to make all of these disks one volume. There are heavy reads and writes to the volume1, but I don't know if having a heavily used volume combined with root functions will kill my performance. Besides this will make a larger volume for more distributed writes (8 total disks versus 4). Also I will be saving 1 disk of data space because I won't need the second parity disk. Addition information is that they are fibre channel shelves (of course). And that the data is not important (washed away daily). Assuming that the root volume functions do not critically affect my box (and I don't believe that they do, as most of it is running from memory) I believe my choice is correct. Just wondering if you guys agree with this hypothesis or not. Nicholas Pesce Internet Production Services Phone: 847 488-6384 Pager: 888 785 3455 Fax: 847 488 3434

1 0

RE: Restricting CIFS
by Hawley, Rob 11 Sep '00

11 Sep '00

The WINS server must be able to ping each interface on the filer that you want to register with WINS. In situations like you describe where you have a private network that should not be registered with WINS, then you need to turn off WINS registration on that interface. You do that by adding a -wins parameter to the ifconfig for the interface. ~~~~~~~~~~~~~~~~~ mailto: hawleyr(a)netapp.com US Mail: Network Appliance Inc. Windows Networking Engineering 495 E Java Dr Phone/pager: 408-822-6661 Sunnyvale, CA 94089 http://www.netapp.com -----Original Message----- From: Graydon Dodson [mailto:grdodson@lexmark.com] Sent: Monday, September 11, 2000 1:09 PM To: toasters(a)mathworks.com Subject: Restricting CIFS Anyone know of a way to restrict CIFS on a filer to certain network interfaces. Specificaly we need to turn it off on one interface. This interface is to a private net and we keep getting: No WINS servers are responding; still trying warnings clogging up the log. We want CIFS on the other interfaces and it can easily contact the wins servers on those nets. Thanks. Graydon Dodson grdodson(a)lexmark.com Lexmark International Inc.

1 0

RE: Restricting CIFS
by Benn, Paul 11 Sep '00

11 Sep '00

you can disable wins on just that interface by entering: filer> ifconfig <interface> -wins -----Original Message----- From: Graydon Dodson [mailto:grdodson@lexmark.com] Sent: Monday, September 11, 2000 1:09 PM To: toasters(a)mathworks.com Subject: Restricting CIFS Anyone know of a way to restrict CIFS on a filer to certain network interfaces. Specificaly we need to turn it off on one interface. This interface is to a private net and we keep getting: No WINS servers are responding; still trying warnings clogging up the log. We want CIFS on the other interfaces and it can easily contact the wins servers on those nets. Thanks. Graydon Dodson grdodson(a)lexmark.com Lexmark International Inc.

1 0

Restricting CIFS
by Graydon Dodson 11 Sep '00

11 Sep '00

Anyone know of a way to restrict CIFS on a filer to certain network interfaces. Specificaly we need to turn it off on one interface. This interface is to a private net and we keep getting: No WINS servers are responding; still trying warnings clogging up the log. We want CIFS on the other interfaces and it can easily contact the wins servers on those nets. Thanks. Graydon Dodson grdodson(a)lexmark.com Lexmark International Inc.

2 1

RE: Restricting CIFS
by Krause, Oliver 11 Sep '00

11 Sep '00

You cannot restrict CIFS to an interface. you get rid of the error message if you configure the private interface with the "-wins" option. Then the filer doesn't try to register this ip on the wins server. Oliver > -----Original Message----- > From: Graydon Dodson [mailto:grdodson@lexmark.com] > Sent: Montag, 11. September 2000 22:09 > To: toasters(a)mathworks.com > Subject: Restricting CIFS > > > > Anyone know of a way to restrict CIFS on a filer to certain > network interfaces. > Specificaly we need to turn it off on one interface. This > interface is to > a private net and we keep getting: > > No WINS servers are responding; still trying > > warnings clogging up the log. We want CIFS on the other > interfaces and it can > easily contact the wins servers on those nets. > > > Thanks. > > Graydon Dodson > grdodson(a)lexmark.com > Lexmark International Inc. >

1 0

RE: creating a share with vba on a filer
by Uren, Jim 11 Sep '00

11 Sep '00

Hi Ulrich, from a DOS prompt, typing net helpmsg 2116 reveals this-- The device or directory does not exist. EXPLANATION You specified an unknown device or directory. ACTION Check the spelling of the device or directory name. so checking the pathname as suggested below seems like a good idea. as I recall, if you wish to use the "vol" form of the filename, you will want to put C$ in the pathname, such as: \\filer\c$\vol\vol0\sharename\directory but the \\filer\share\directory format should work just fine. check to see if c:\> dir \\cobfile01\users$ succeeds or not for example. Jim U > -----Original Message----- > From: Hawley, Rob [mailto:rob.hawley@netapp.com] > Sent: Monday, September 11, 2000 7:31 AM > To: 'Ulrich Neumann'; toasters(a)mathworks.com > Subject: RE: creating a share with vba on a filer > > > Your program looks correct. We do support using level 502 to > add a share. > Please confirm a couple of points that could be the source of > the problem: > > 1. The user running the script is a member of either the > filer's Administrators > group or is a member of Power Users. That is required for NT > compatible > security. > 2. That the path vol\nt01\users is the correct physical path > to where the user's > directory should be created. That is there are no other > directories in the path. > > If neither of these resolve your problem, then please contact > NetApp customer > support. They will need a packet trace of this exchange and a > cifs sessions -s > of the user attempting to execute the script. CS can help you > if you are > unfamiliar with how to get a packet trace. > > ~~~~~~~~~~~~~~~~~~~ > mailto: hawleyr(a)netapp.com US Mail: Network Appliance Inc. > Windows Networking Engineering 495 E Java Dr > Phone/pager: 408-822-6661 Sunnyvale, CA 94089 > http://www.netapp.com > > > -----Original Message----- > From: Ulrich Neumann [mailto:ulrich.neumann@brose.de] > Sent: Monday, September 11, 2000 2:01 AM > To: toasters(a)mathworks.com > Subject: creating a share with vba on a filer > > << File: ShareErr2116Filer.txt >> we just turned our > user-home-shares from a > nt-server to a filer. > for creating new user-shares, we use a selfmade program in vb. > unfortunately this program woun't work now with the filer. > > neither "c:\vol\nt01\users\username" nor "\\cobfile01\users$\username" is working. there's always the error 2116.(see sub makeshare in attached code) is there anybody out there who has a good idea to solve this problem? greetings from brose GmbH & Co KG Client-Server-Technologien Ulrich Neumann

1 0

Re: Deleting snapshot
by Steve Kappel 11 Sep '00

11 Sep '00

Hi Stephen, Thanks for jumping in here and ending this thread. >It definitely does not seem appropriate to blame >the NDMP client. It is a filer specific issue. >Apologies to Steve Kappel. But I bet his >blood pressure wasn't high enough. ;) Naw, if I let a little thing like this bother me I'd be dead already ;) I'd rather not see this list turn into a blame game of any kind. We all have our good days and our not so good days. I would urge readers who have problems like this to use the support organizations of both NetApp and VERITAS. For at least the VERITAS side, if you are unhappy with an answer you get from the support organization please ask the support engineer to escalate the problem - use my name if you have to :) __________________________________________________________________________ Steve Kappel steve.kappel(a)veritas.com VERITAS Software steve.kappel(a)iname.com (Personal)

1 0

RE: creating a share with vba on a filer
by Hawley, Rob 11 Sep '00

11 Sep '00

Your program looks correct. We do support using level 502 to add a share. Please confirm a couple of points that could be the source of the problem: 1. The user running the script is a member of either the filer's Administrators group or is a member of Power Users. That is required for NT compatible security. 2. That the path vol\nt01\users is the correct physical path to where the user's directory should be created. That is there are no other directories in the path. If neither of these resolve your problem, then please contact NetApp customer support. They will need a packet trace of this exchange and a cifs sessions -s of the user attempting to execute the script. CS can help you if you are unfamiliar with how to get a packet trace. ~~~~~~~~~~~~~~~~~~~ mailto: hawleyr(a)netapp.com US Mail: Network Appliance Inc. Windows Networking Engineering 495 E Java Dr Phone/pager: 408-822-6661 Sunnyvale, CA 94089 http://www.netapp.com -----Original Message----- From: Ulrich Neumann [mailto:ulrich.neumann@brose.de] Sent: Monday, September 11, 2000 2:01 AM To: toasters(a)mathworks.com Subject: creating a share with vba on a filer << File: ShareErr2116Filer.txt >> we just turned our user-home-shares from a nt-server to a filer. for creating new user-shares, we use a selfmade program in vb. unfortunately this program woun't work now with the filer. neither "c:\vol\nt01\users\username" nor "\\cobfile01\users$\username" is working. there's always the error 2116.(see sub makeshare in attached code) is there anybody out there who has a good idea to solve this problem? greetings from brose GmbH & Co KG Client-Server-Technologien Ulrich Neumann

1 0

creating a share with vba on a filer
by Ulrich Neumann 11 Sep '00

11 Sep '00

we just turned our user-home-shares from a nt-server to a filer. for creating new user-shares, we use a selfmade program in vb. unfortunately this program woun't work now with the filer. neither "c:\vol\nt01\users\username" nor "\\cobfile01\users$\username" is working. there's always the error 2116.(see sub makeshare in attached code) is there anybody out there who has a good idea to solve this problem? greetings from brose GmbH & Co KG Client-Server-Technologien Ulrich Neumann

1 0

Re: dlt autoloaders
by Darrell Fuhriman 09 Sep '00

09 Sep '00

Brian Tao <taob(a)risc.org> writes: > Out of curiosity, what's the aggregate throughput on your backups > to 12 DLT7000's? They should be able to do, what, 5MB/sec if you can > stream data to them consistently? 12 drives would be 60MB/sec... are You can push much more than that -- we sustain 10-12MB/sec to a DLT7000. > you actually seeing that much throughput over the network? Yes, we see 60MB/sec, and have pushed more in tests. Darrell

1 0

Re: gigabit ethernet performance
by Eyal Traitel 09 Sep '00

09 Sep '00

Only on 6.0 buddy... Eyal. --- Brian Tao <taob(a)risc.org> wrote: > On Thu, 7 Sep 2000, Eyal Traitel wrote: > > > > Have you considered Jumbo packets ? It could help you... > > How do you set a jumbo frame size on a Netapp? I recall trying a > few combinations of ifconfig invocations, but nothing would bump it > up > to 9K. > -- > Brian Tao (BT300, taob(a)risc.org) > "Though this be madness, yet there is method in't" > ===== Yours, Eyal Traitel eTraitel(a)yahoo.com, Home: 972-3-5290415 (Tel Aviv) *** eTraitel - it's the new eBuzzword ! *** __________________________________________________ Do You Yahoo!? Yahoo! Mail - Free email you can access from anywhere! http://mail.yahoo.com/

1 0

Re: gigabit ethernet performance
by Eyal Traitel 09 Sep '00

09 Sep '00

Marcus, Have you considered Jumbo packets ? It could help you... Eyal. Marcus Smiley <marcus.smiley(a)ericsson.com> wrote: Does anyone have any benchmarks/suggestions for poor performace with gigabit ethernet between Sun enterprise 220R and NetApp 740 with Intel GBII card? I have an open ticket w/Netapp and it's been open forever. These machines are directly attached (private network), and I've tried a mkfile and copying that file back during the day and at night (no load) and it makes no difference. The performance is ~ 10 MB/s, no matter what. The Sun is running Solaris 8, with latest GB Ethernet 3.0 drivers and patches. Also, the Sun box's only function is Legato server, so I've even stopped all nsr daemons while the test copy/mkfile is in progress, & it makes no difference. I've tried NFSv2 vs. NFSv3, with 32k size packets and 8k size packets. I've tried tcp and udp, with the same results. The /etc/system file on the E220R looks like this, for the ge adapter: * Increase size of STREAMS syncronized queues to increase network * performance. Release notes for qfe card recommends setting this * to 25 per 64Mbytes of RAM in the system. It also prevents * receive overrun on the GbE interface. msmiley 7/20/00 set sq_max_size=775 set ge:ge_adv_1000autoneg_cap=0 set ge:ge_adv_1000fdx_cap=1 set ge:ge_adv_pauseTX=1 Also, I've set tcp/udp specific stuff per the Netapp tech suppt. folks: ndd -set /dev/udp udp_recv_hiwat 65535 ndd -set /dev/udp udp_xmit_hiwat 65535 ndd -set /dev/tcp tcp_recv_hiwat 65535 ndd -set /dev/tcp tcp_xmit_hiwat 65535 The netapp is a 740 (5.3.6R1P1) with over 100GB of free space available on the volume I'm copying to/from. Interestingly enough, I have done a packet trace (pktt) on the GBII card and it looks OK. Also, the throughput from my 100BaseT card to the netapp (thru a catalyst 5500) does ~ 7.5 MB/s. *Any* input here would be much appreciated. -Marc Marcus Smiley | Ericsson Wireless Communications, Inc. Unix Sys Admin | 6210 Spine Rd. Boulder, CO 80301 | Ph (303) 473-6903 Yours, Eyal Traitel eTraitel(a)yahoo.com, Home: 972-3-5290415 (Tel Aviv) *** eTraitel - it's the new eBuzzword ! *** --------------------------------- Do You Yahoo!? Yahoo! Mail - Free email you can access from anywhere!

2 1

gigabit ethernet performance
by Marcus Smiley 09 Sep '00

09 Sep '00

Does anyone have any benchmarks/suggestions for poor performace with gigabit ethernet between Sun enterprise 220R and NetApp 740 with Intel GBII card? I have an open ticket w/Netapp and it's been open forever. These machines are directly attached (private network), and I've tried a mkfile and copying that file back during the day and at night (no load) and it makes no difference. The performance is ~ 10 MB/s, no matter what. The Sun is running Solaris 8, with latest GB Ethernet 3.0 drivers and patches. Also, the Sun box's only function is Legato server, so I've even stopped all nsr daemons while the test copy/mkfile is in progress, & it makes no difference. I've tried NFSv2 vs. NFSv3, with 32k size packets and 8k size packets. I've tried tcp and udp, with the same results. The /etc/system file on the E220R looks like this, for the ge adapter: * Increase size of STREAMS syncronized queues to increase network * performance. Release notes for qfe card recommends setting this * to 25 per 64Mbytes of RAM in the system. It also prevents * receive overrun on the GbE interface. msmiley 7/20/00 set sq_max_size=775 set ge:ge_adv_1000autoneg_cap=0 set ge:ge_adv_1000fdx_cap=1 set ge:ge_adv_pauseTX=1 Also, I've set tcp/udp specific stuff per the Netapp tech suppt. folks: ndd -set /dev/udp udp_recv_hiwat 65535 ndd -set /dev/udp udp_xmit_hiwat 65535 ndd -set /dev/tcp tcp_recv_hiwat 65535 ndd -set /dev/tcp tcp_xmit_hiwat 65535 The netapp is a 740 (5.3.6R1P1) with over 100GB of free space available on the volume I'm copying to/from. Interestingly enough, I have done a packet trace (pktt) on the GBII card and it looks OK. Also, the throughput from my 100BaseT card to the netapp (thru a catalyst 5500) does ~ 7.5 MB/s. *Any* input here would be much appreciated. -Marc Marcus Smiley | Ericsson Wireless Communications, Inc. Unix Sys Admin | 6210 Spine Rd. Boulder, CO 80301 | Ph (303) 473-6903

2 1

dlt autoloaders
by neil lehrer 09 Sep '00

09 Sep '00

hi, i'd like recommendations for a dlt autoloader to be attached to an f760 and controlled by legato. thanks. --

5 6

RE: Deleting snapshot
by Premanshu Jain 09 Sep '00

09 Sep '00

Ndmpd status, doesn't revealany active ndmp sessions. These snapshots date back to 17 Aug. Any other way out..avoiding reboot prem -----Original Message----- From: Steve Kappel [mailto:steve.kappel@raistlin.min.ov.com] Sent: Friday, September 08, 2000 1:22 PM To: mrb(a)netapp.com; PrJain(a)shastanets.com Cc: toasters(a)mathworks.com Subject: RE: Deleting snapshot No, not a chance that it is the backup software (NDMP client). Consider this - if your network link goes down your backup (NDMP client) is aborted. Since you have no network link there is no way the backup product can do anything in NDMP to abort or cleanup the backup running in the filer. The filer has to be able to clean up from the NDMP session socket "going away". Since you have "failed backup processes" the NDMP session is closed on the client side. So it is OnTap that is not functioning correctly. You might try looking for the NDMP sessions with "ndmpd status" on the filer and then kill any that are running with "ndmpd kill <sessionid>". >From: Premanshu Jain <PrJain(a)shastanets.com> >To: "'mrb(a)netapp.com'" <mrb(a)netapp.com>, Premanshu Jain <PrJain(a)shastanets.com> >Cc: toasters(a)mathworks.com >Subject: RE: Deleting snapshot >Date: Fri, 8 Sep 2000 11:09:01 -0700 > >NDMP or Veritas Netbackup. All these snapshots were a part of failed backups >prem > >-----Original Message----- >From: J. Bryan Kelsch [mailto:mrb@netapp.com] >Sent: Friday, September 08, 2000 11:13 AM >To: Premanshu Jain >Cc: toasters(a)mathworks.com >Subject: Re: Deleting snapshot > > >This is not a function the filer, it is an error in the backup software that >is not cleaning up properly. > > > > >> >> >> >>You have to reboot first. >> >> >>On Fri, 8 Sep 2000 10:42:26 -0700 , you wrote: >> >>>I have some snapshots created by failed backup processes and shows them >busy >>>still. How can I delete them..? >>> >>> >>>Volume sw >>>working.................................................................. . >.. >>>......................................................................... . >.. >>>.................................... >>> >>> %/used %/total date name >>>---------- ---------- ------------ -------- >>> 3% ( 3%) 2% ( 2%) Sep 07 18:04 bot_sw.1 >>> 4% ( 3%) 3% ( 2%) Sep 07 16:12 hourly.0 >>> 4% ( 4%) 3% ( 3%) Sep 07 12:10 hourly.1 >>> 5% ( 4%) 3% ( 3%) Sep 07 08:10 hourly.2 >>> 5% ( 4%) 4% ( 3%) Sep 07 00:11 nightly.0 >>> 6% ( 4%) 4% ( 3%) Sep 06 20:11 hourly.3 >>> 7% ( 4%) 5% ( 3%) Sep 06 16:12 hourly.4 >>> 7% ( 4%) 5% ( 3%) Sep 06 12:10 hourly.5 >>> 8% ( 4%) 6% ( 3%) Sep 06 00:08 nightly.1 >>> 19% (15%) 17% (12%) Aug 20 21:52 snapshot_for_backup.32 (busy) >>> 20% (15%) 17% (12%) Aug 19 22:46 snapshot_for_backup.30 (busy) >>> 21% (15%) 19% (12%) Aug 17 04:08 snapshot_for_backup.23 (busy) > > >------------------------ >J. Bryan Kelsch, PSE >Network Appliance >5220 Spring Valley Road - Suite 350 >Dallas, TX 75240 >972.759.7528 Office >817.821.0350 Cell >888.374.6422 Pager __________________________________________________________________________ Steve Kappel steve.kappel(a)veritas.com VERITAS Software steve.kappel(a)iname.com (Personal)

3 2

multiple raid grps in filesystem on filer
by neil lehrer 08 Sep '00

08 Sep '00

hi, i understand that having more than 1 raid grp in a filesystem makes the filer more fault tolerant of a full raid grp failure. so lets say there are two raid grps in the filesystem. if you lose a raid grp [2 drives fail] how do you restore it? i can understand restoring the whole filesystem or a qtree or a dir, but if files and dirs can go across more than 1 raid grp in a filesystem what is the advantage of short restore of the broken raid grp? how do you put back just that raid group? i know you can replace the bad drives and then recreate the group but how do you restore the data for just that raid grp? thanks

2 1

RE: Deleting snapshot
by Steve Kappel 08 Sep '00

08 Sep '00

No, not a chance that it is the backup software (NDMP client). Consider this - if your network link goes down your backup (NDMP client) is aborted. Since you have no network link there is no way the backup product can do anything in NDMP to abort or cleanup the backup running in the filer. The filer has to be able to clean up from the NDMP session socket "going away". Since you have "failed backup processes" the NDMP session is closed on the client side. So it is OnTap that is not functioning correctly. You might try looking for the NDMP sessions with "ndmpd status" on the filer and then kill any that are running with "ndmpd kill <sessionid>". >From: Premanshu Jain <PrJain(a)shastanets.com> >To: "'mrb(a)netapp.com'" <mrb(a)netapp.com>, Premanshu Jain <PrJain(a)shastanets.com> >Cc: toasters(a)mathworks.com >Subject: RE: Deleting snapshot >Date: Fri, 8 Sep 2000 11:09:01 -0700 > >NDMP or Veritas Netbackup. All these snapshots were a part of failed backups >prem > >-----Original Message----- >From: J. Bryan Kelsch [mailto:mrb@netapp.com] >Sent: Friday, September 08, 2000 11:13 AM >To: Premanshu Jain >Cc: toasters(a)mathworks.com >Subject: Re: Deleting snapshot > > >This is not a function the filer, it is an error in the backup software that >is not cleaning up properly. > > > > >> >> >> >>You have to reboot first. >> >> >>On Fri, 8 Sep 2000 10:42:26 -0700 , you wrote: >> >>>I have some snapshots created by failed backup processes and shows them >busy >>>still. How can I delete them..? >>> >>> >>>Volume sw >>>working................................................................... >.. >>>.......................................................................... >.. >>>.................................... >>> >>> %/used %/total date name >>>---------- ---------- ------------ -------- >>> 3% ( 3%) 2% ( 2%) Sep 07 18:04 bot_sw.1 >>> 4% ( 3%) 3% ( 2%) Sep 07 16:12 hourly.0 >>> 4% ( 4%) 3% ( 3%) Sep 07 12:10 hourly.1 >>> 5% ( 4%) 3% ( 3%) Sep 07 08:10 hourly.2 >>> 5% ( 4%) 4% ( 3%) Sep 07 00:11 nightly.0 >>> 6% ( 4%) 4% ( 3%) Sep 06 20:11 hourly.3 >>> 7% ( 4%) 5% ( 3%) Sep 06 16:12 hourly.4 >>> 7% ( 4%) 5% ( 3%) Sep 06 12:10 hourly.5 >>> 8% ( 4%) 6% ( 3%) Sep 06 00:08 nightly.1 >>> 19% (15%) 17% (12%) Aug 20 21:52 snapshot_for_backup.32 (busy) >>> 20% (15%) 17% (12%) Aug 19 22:46 snapshot_for_backup.30 (busy) >>> 21% (15%) 19% (12%) Aug 17 04:08 snapshot_for_backup.23 (busy) > > >------------------------ >J. Bryan Kelsch, PSE >Network Appliance >5220 Spring Valley Road - Suite 350 >Dallas, TX 75240 >972.759.7528 Office >817.821.0350 Cell >888.374.6422 Pager __________________________________________________________________________ Steve Kappel steve.kappel(a)veritas.com VERITAS Software steve.kappel(a)iname.com (Personal)

2 1

RE: Deleting snapshot
by Premanshu Jain 08 Sep '00

08 Sep '00

Oh..This is CRAZY on part of NetAPPS...!!!!!!!!!! -----Original Message----- From: J. Bryan Kelsch [mailto:mrb@netapp.com] Sent: Friday, September 08, 2000 10:56 AM To: Premanshu Jain Subject: Re: Deleting snapshot You have to reboot first. On Fri, 8 Sep 2000 10:42:26 -0700 , you wrote: >I have some snapshots created by failed backup processes and shows them busy >still. How can I delete them..? > > >Volume sw >working.................................................................... . >........................................................................... . >.................................... > > %/used %/total date name >---------- ---------- ------------ -------- > 3% ( 3%) 2% ( 2%) Sep 07 18:04 bot_sw.1 > 4% ( 3%) 3% ( 2%) Sep 07 16:12 hourly.0 > 4% ( 4%) 3% ( 3%) Sep 07 12:10 hourly.1 > 5% ( 4%) 3% ( 3%) Sep 07 08:10 hourly.2 > 5% ( 4%) 4% ( 3%) Sep 07 00:11 nightly.0 > 6% ( 4%) 4% ( 3%) Sep 06 20:11 hourly.3 > 7% ( 4%) 5% ( 3%) Sep 06 16:12 hourly.4 > 7% ( 4%) 5% ( 3%) Sep 06 12:10 hourly.5 > 8% ( 4%) 6% ( 3%) Sep 06 00:08 nightly.1 > 19% (15%) 17% (12%) Aug 20 21:52 snapshot_for_backup.32 (busy) > 20% (15%) 17% (12%) Aug 19 22:46 snapshot_for_backup.30 (busy) > 21% (15%) 19% (12%) Aug 17 04:08 snapshot_for_backup.23 (busy) ------------------------ J. Bryan Kelsch, PSE Network Appliance 5220 Spring Valley Road - Suite 350 Dallas, TX 75240 972.759.7528 Office 817.821.0350 Cell 888.374.6422 Pager

2 1

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters