toasters

toasters@lists.teaparty.net

1 participants
13516 discussions

RE: Moving files but not .snapshots?
by Art Hebert 13 Mar '03

13 Mar '03

I take it when copying files using ndmpcopy on a single source and destination filer the data never uses the network. This seems like it should be a lot faster, anybody have an idea how much faster? Also, which release was the first with ndmpcopy on board? Is there any other ways to copy just on the filer itself beside ndmpcopy? thanks art -----Original Message----- From: Jeffrey.Burton(a)abbott.com [mailto:Jeffrey.Burton@abbott.com] Sent: Thursday, March 13, 2003 12:19 PM To: Jim Davis Cc: toasters(a)mathworks.com Subject: Re: Moving files but not .snapshots? You can use ndmpcopy from the filerhead CLI, granted you have a new enough DOT version. This creates a snapshot then copy's all the data and ACL's from source to target. First make sure that the ndmp option is turned on: ndmpd.enable on Then below will work: ndmpcopy /vol/sourcevol /vol/destvol You can copy at any level, not just the vol level. If the source and dest's vols are on a seperate filer just preceed the dir with "filename:" so "ndmpcopy sourcefiler:/vol/sourcevol destfiler:/vol/destvol". After the copy you can remove the temp file created at the destination end which is used for incremental updates (I forgot the name but it will be a huge file with an obviously generated name). Man page and NOW has good ndmp info. Thanks, Jeff Jim Davis <jdavis(a)cs.arizona.edu> Sent by: owner-toasters(a)mathworks.com 03/13/2003 01:47 PM To: toasters(a)mathworks.com cc: Subject: Moving files but not .snapshots? I want to move two big directories from an old volume to a new volume, but just the files -- not the snapshot subdirectories. So far I've tried find . -type d -name '.snapshot' -prune -o -print | cpio -pdm /newvol from the adminhost, which works but isn't blazingly fast. Are there any other ways people can suggest?

1 0

RE: Moving files but not .snapshots?
by Toal, Dave 13 Mar '03

13 Mar '03

my mistake, the /oldvol mounts should be different dirs under root. cd /oldvol1/dir1; find . ... | cpio -dmp /newvol1/dir1 & cd /oldvol2/dir2; find . ... | cpio -dmp /newvol2/dir2 & The /old/1, /old/2 would also work; you'd be reading from the same source mount though, unless the individual dirs were mounted separately. mount filer:/vol/old/1 /old/1 mount filer:/vol/old/2 /old/2 -----Original Message----- From: Toal, Dave To: 'Jim Davis '; 'toasters(a)mathworks.com ' Sent: 3/13/03 3:45 PM Subject: RE: Moving files but not .snapshots? Jim, If there are subdirectories in what you're copying, one way to speed things up is mounting the same pair of vols (old, new) multiple times, and running the find | cpio in each subdir. Use more than one cp host if you saturate i/o or cpu on the first one. cd /oldvol/dir1; find . ... | cpio -dmp /newvol1/dir1 & cd /oldvol/dir2; find . ... | cpio -dmp /newvol2/dir2 & > What I am seeking is a bit like folklore. How about a spam filter that bounces any posts that don't contain keywords? "filer, vol, ndmp, spam" ? -----Original Message----- From: Jim Davis To: toasters(a)mathworks.com Sent: 3/13/03 2:47 PM Subject: Moving files but not .snapshots? I want to move two big directories from an old volume to a new volume, but just the files -- not the snapshot subdirectories. So far I've tried find . -type d -name '.snapshot' -prune -o -print | cpio -pdm /newvol from the adminhost, which works but isn't blazingly fast. Are there any other ways people can suggest?

1 0

RE: Moving files but not .snapshots?
by Toal, Dave 13 Mar '03

13 Mar '03

Jim, If there are subdirectories in what you're copying, one way to speed things up is mounting the same pair of vols (old, new) multiple times, and running the find | cpio in each subdir. Use more than one cp host if you saturate i/o or cpu on the first one. cd /oldvol/dir1; find . ... | cpio -dmp /newvol1/dir1 & cd /oldvol/dir2; find . ... | cpio -dmp /newvol2/dir2 & > What I am seeking is a bit like folklore. How about a spam filter that bounces any posts that don't contain keywords? "filer, vol, ndmp, spam" ? -----Original Message----- From: Jim Davis To: toasters(a)mathworks.com Sent: 3/13/03 2:47 PM Subject: Moving files but not .snapshots? I want to move two big directories from an old volume to a new volume, but just the files -- not the snapshot subdirectories. So far I've tried find . -type d -name '.snapshot' -prune -o -print | cpio -pdm /newvol from the adminhost, which works but isn't blazingly fast. Are there any other ways people can suggest?

1 0

Re: Moving files but not .snapshots?
by Jeffrey.Burton＠abbott.com 13 Mar '03

13 Mar '03

You can use ndmpcopy from the filerhead CLI, granted you have a new enough DOT version. This creates a snapshot then copy's all the data and ACL's from source to target. First make sure that the ndmp option is turned on: ndmpd.enable on Then below will work: ndmpcopy /vol/sourcevol /vol/destvol You can copy at any level, not just the vol level. If the source and dest's vols are on a seperate filer just preceed the dir with "filename:" so "ndmpcopy sourcefiler:/vol/sourcevol destfiler:/vol/destvol". After the copy you can remove the temp file created at the destination end which is used for incremental updates (I forgot the name but it will be a huge file with an obviously generated name). Man page and NOW has good ndmp info. Thanks, Jeff Jim Davis <jdavis(a)cs.arizona.edu> Sent by: owner-toasters(a)mathworks.com 03/13/2003 01:47 PM To: toasters(a)mathworks.com cc: Subject: Moving files but not .snapshots? I want to move two big directories from an old volume to a new volume, but just the files -- not the snapshot subdirectories. So far I've tried find . -type d -name '.snapshot' -prune -o -print | cpio -pdm /newvol from the adminhost, which works but isn't blazingly fast. Are there any other ways people can suggest?

1 0

RE: antivirus software
by Wurster, Richard M DSTI 13 Mar '03

13 Mar '03

We have had very good results using Symantec's CarrierScan product, which is designed specifically for the filer. Rick Wurster SBCCOM / DSTI 410-436-3824 DSN 584-3824 <mailto:rick.wurster@sbccom.apgea.army.mil> -----Original Message----- From: David Papas [mailto:david.papas@newisys.com] Sent: Thursday, March 13, 2003 12:24 PM To: toasters(a)mathworks.com Subject: antivirus software Just curious which antivirus software, if any, folks are using on their filers out there. We're considering either Symantec or Trend's for an F820 and was wondering what people's experience with scanning filers has been like. Thanks all, David Papas

3 2

Writer/Researcher Requests your Help
by Steve Lawler 13 Mar '03

13 Mar '03

Dear Colleague, Please excuse the impersonal quality of this e-mail. I am writing to ask your help for an article I am writing about Life Practices. I am looking for examples of those simple things that we all do that are essential to expressing and maintaining our values as we face life's ups and downs - our Life Practices. I am especially interested in practices developed by those who have been in, lived through or dealt with the aftermath of war. What I am seeking is a bit like folklore. Something with a story and an easy to remember slogan or catchphrase. I am asking you to contribute out of your own personal experience not as an official representative of an organization. I am especially interested in those practices, which help you deal with others - in your family, at work, in your immediate community, in the larger global community, etc. This is the only e-mail you will receive from me unless you choose to participate in this project. So far I have gathered over 150 practices from people all over the world: These are examples of contributions I have received so far that may help clarify what I am looking for: ---------------------------------------------------------------------------- " 'Water what you want to grow' is my motto. Whenever I see something being done well I acknowledge it as soon as possible and in some tangible way. The least that I do is to complement the person doing the thing well. What I try to do more often is write a note to the person or give them a small gift." "My life changed when I went from seeing mistakes as awful to seeing them as the first chapter of a textbook on a new subject. I even had this made into a sign that hangs on the wall in my office. 'Mistakes are just chapter one in a textbook on a new subject' Now I get bothered if I don't have some sort of royal screw-up every few days. It's like I am not trying." "I learned from my grandfather to 'use the best materials you can afford.' This has provided a model for my life's work. Even in my late 70's I am still actively building things and more to the point of your book, I am still building my life out of "the best materials I can afford." ---------------------------------------------------------------------------- These practices come from people from many places and with a wide variety of backgrounds. I am grateful for that. I want to show the broadest possible view of Life Practices from around the globe. The article will feature some of the material I gather with some commentary. I will be grateful if you would send me a practice that you use as a foundation in your life. Please include some brief biographical information as well. Also, if you would like to share in what I am learning, please let me know as well. And please let me know if you have any additional questions. Thank you for your consideration. Steve Lawler swl(a)lawler.org About me: Steve Lawler an ethics consultant and writer based in St Louis, Missouri USA. He regularly gives workshops on issues of ethics, values and culture. He has been quoted in a number of related pieces in publications including the Wall Street Journal, The Christian Science Monitor and Cape Talk Radio - Cape Town. Lawler has worked with companies in the US and Europe - including Monsanto, Pulitzer Publishing, and the Danforth Foundation. He has been a lecturer at Washington University and at Webster University. He has written for numerous publications. Steve Lawler 47 Aberdeen Place, St. Louis, MO 63105 USA Voice: +1 (314) 753 7911 Fax: +1 (314) 727 9792 E-mail: swl(a)lawler.org Web Site: www.lawler.org

1 0

RE: antivirus software
by Nail, Larry 13 Mar '03

13 Mar '03

David, I did an analysis about 18 months ago & Trend came out on top for performance and managebility. Symantec CS was 2nd followed by McAfee. All 3 work fine & each have their pluses & minues. In the long run, it depends on what pricing you can pull out of the 3 of them and what you want for manageiblity. Carrier Scan wasn't really manageable, and McAfee requires EPO to centralize the stuff. Trend comes with their management utility in the form of an Information Server. I also noticed on busy systems (this was a F760 cluster running 99.9% CIFS) that CS & McAfee performance wasn't as good a Trend... Larry -----Original Message----- From: David Papas [mailto:david.papas@newisys.com] Sent: Thursday, March 13, 2003 11:24 AM To: toasters(a)mathworks.com Subject: antivirus software Just curious which antivirus software, if any, folks are using on their filers out there. We're considering either Symantec or Trend's for an F820 and was wondering what people's experience with scanning filers has been like. Thanks all, David Papas

1 0

Re: antivirus software
by John Duino 13 Mar '03

13 Mar '03

David Papas <david.papas(a)newisys.com> said: > Just curious which antivirus software, if any, folks are using on their > filers out there. We're considering either Symantec or Trend's for an > F820 and was wondering what people's experience with scanning filers has > been like. > > > Thanks all, > David Papas > I've been using and installing RAV antivirus (ravantivirus.com or from raeinternet.com) in various installations with great success. -- *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* | John Duino | | AeroHead Designs | | UNIX, Network & WWW Consulting | | (310) 993-9188 | | jduino(a)AeroHead.com | | http://www.AeroHead.com | *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*

1 0

RE: antivirus software
by Gifford, Josh 13 Mar '03

13 Mar '03

Hi David, I was on the beta team for the Trend product and I have been very happy with it. The overhead on the filer is minimal and you can use existing NT servers to scan the filer. If you already use Trend ServerProtect, then it will run under the information server that you already have and you can manage it through the console that you are used to using. Currently Trend is working on another version with a lot of new features, but I'm not sure when I'll see the beta on that. If you have any specific questions, please let me know. Josh J. Gifford MCP Network Systems Administrator Siemens PTD 7000 Siemens Rd Wendell, NC 27591 Office 919.365.2806 Fax 919.365.1080 josh.gifford(a)siemens.com -----Original Message----- From: David Papas [mailto:david.papas@newisys.com] Sent: Thursday, March 13, 2003 12:24 PM To: toasters(a)mathworks.com Subject: antivirus software Just curious which antivirus software, if any, folks are using on their filers out there. We're considering either Symantec or Trend's for an F820 and was wondering what people's experience with scanning filers has been like. Thanks all, David Papas

1 0

antivirus software
by David Papas 13 Mar '03

13 Mar '03

Just curious which antivirus software, if any, folks are using on their filers out there. We're considering either Symantec or Trend's for an F820 and was wondering what people's experience with scanning filers has been like. Thanks all, David Papas

2 1

RE: CIFS getattr load
by Sam Schorr 13 Mar '03

13 Mar '03

Todd, Do you have the no_atime_update=on option set on your filer volume(s)? If not, this can generate much activity as the filer updates the access time for each file hit. Automated testing would really whack them. --sam -----Original Message----- From: Todd C. Merrill [mailto:tmerrill@mathworks.com] Sent: Tuesday, March 11, 2003 2:27 PM To: toasters(a)mathworks.com Subject: CIFS getattr load Instead of standing at a crossroads or a fork in the road, I'm standing in an open field, and need somebody to point me in a direction... My filers are showing loads and loads of CIFS getattr operations, on the order of 80% of the CIFS traffic, which is 80-90% of the total NFS and CIFS traffic. Finding out why this is so, or alleviating it or reducing it so the load on the filers goes down, is my goal. The filers are used in a software development environment, in an automated testing environment (generally no interactive usage by people). The CIFS/getattr load is most imbalanced not when a software build is happening but while testing suites are run. On the order of only 10 Windows clients (generally Win2k) can nearly cripple an F840. So, at this point, any WAG's are welcome. Do directories with large numbers of files stress the getattr operations? Has anyone run across any applications that are very "chatty" with filers? Has anyone run into any bugs (with ONTAP or with Windows) that might show such a symptom? Perhaps the application mix we are running is just this way, but something in my gut tells me something is amiss. I have a sniffer available, but before I get to that detail, I was hoping to mine the wisdom of this list, to approach this vague problem armed with some good ideas. Until next time... The MathWorks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

2 1

RE: ClearCase 5.x on NetApp, mixed NFS/CIFS setup issues
by Clawson, Simon 13 Mar '03

13 Mar '03

we found that the key things to check are - groups - You must have the same primary NT group as UNIX group - the same group name. clearcase_aldb - This user must exist in both realms. root=<hostname> - This must be set on the nfs share for all unix systems that will have root access to the vobs and views. access=<hostname> - This must be set on the nfs share for all unix systems that will have access to the shares. There is a great white paper on this at http://www.netapp.com/tech_library/311.html?fmt=print Simon -----Original Message----- From: Marion Hakanson [mailto:hakanson@cse.ogi.edu] Sent: 11 March 2003 21:26 To: John Stoffel Cc: toasters(a)mathworks.com Subject: Re: ClearCase 5.x on NetApp, mixed NFS/CIFS setup issues > . . . > mixed security settings. The problem is that running as root on one > of the ClearCase servers, we cannot remove a file that was created by > a user running on a PC with CC client software. > > Our infrastructure is pretty simple, with just one NIS domain and one > NT domain to work with. But because our corporate IT group won't give > us a short group name in the Win2K domain, we're forced to use the > unix username of 'mss-vobadm' with an NT username of > NA05\sa106068ccalbd. We've got a usermap.cfg as shown below. > > Does anyone have any hints on how we can fix this up, or track down > the proper setup for this environment? > > NA05\sa106068ccalbd => mss-vobadm > NA05\sa106068ccalbd <= root Unencumbered by experience with ClearCase (:-), I'll say that I would try the following in usermap.cfg (assuming that NA05\* refers to local users on the filer, or to your single NT domain): NA05\sa106068ccalbd == mss-vobadm NA05\Administrator == root > And we've tried to use the following cifs options: > . . . > cifs.nfs_root_ignore_acl on > . . . With the above, NFS root users (which are now equivalent to Windows admin users) should be able to override any Windows permissions. It _may_ be necessary to "chown root" the problem files/directories before you'll be able to remove them, depending on what Windows ACL's are in effect. Of course, for root to have this type of access, your NFS partition must be exported with "root=" access to the NFS clients which will be trying to remove this stuff. This type of arrangement works for us here, though without the issue of ClearCase involved. It may be overkill for your situation, but it's a decent place to start. And as I've said before, in case you haven't already done so: There's an excellent article on NOW, with a title something like "Security Troubleshooting Guide", which describes all these issues in great detail. Definitely read through this before converting. Oh yes: I think that if a Windows client has a file open, you won't be able to delete it from the Unix side. You also can't delete a directory which has a Windows share pointing at it. Regards, -- Marion Hakanson <hakanson(a)cse.ogi.edu> CSE Computing Facilities

2 1

Re: NetApps used equipment policies - Service Resourcing Int'l - a reseller
by David Esper 13 Mar '03

13 Mar '03

Dear Toasters, My company Service Resourcing International, Inc is a reseller of Network Appliance hardware. I work hard to provide a positive experience for our customers, and I am happy to work with you, your local dealers and/or Network Appliance directly. Please give me a call if you have any Network Appliance Buy/Sell needs. I also have available F760's with CIFS, NFS Licenses. Guaranteed! Best Regards, David Bransky Service Resourcing International, Inc 4958 Corliss Road Lyndhurst, OH 44124 Tel:(216)382-1400 Fax(216)382-2129 Direct:(216)398-0990 http://www.computer-used.com dbransky(a)ameritech.net WE BUY/SELL: AUSPEX, EMC, HP, NetFRAME, NETWORK APPLIANCE, SGI, STORAGETEK _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=features/junkmail

1 0

Re: Storage management
by Alek O. Komarnitsky (N-CSC) 13 Mar '03

13 Mar '03

> From owner-toasters(a)mathworks.com Wed Mar 12 15:35 MST 2003 > From: "Bui, Marcus" <Marcus.Bui(a)aiminvestments.com> > > Hello Toasters, > > I'm looking for a solution to manage the disk space usage on the storage > environment (Windows NT/2000, Netapp Filer, SAN - Sun, Hitachi). Need to > track who is the 'disk hog', duplicate files, keep historical data, etc.. > I'm sure there's more to it. Anyways, have not dealt with any storage > management solution so any suggestion will help. If you are interested in a non-GUI Perl Code solution that will slice-n-dice your storage data in just about any way possible, you might take a look at yadu (Yet Another du) which can be found at: http:/www.komar.org/ -> Misc. Tech Stuff -> yadu alek P.S. If you want to "stretch" your data, you might also take a look at the "Longest Car on the Internet" which is also there ... or simply type "longest car" into Google and go from there! ;-)

1 0

Storage management
by Bui, Marcus 12 Mar '03

12 Mar '03

Hello Toasters, I'm looking for a solution to manage the disk space usage on the storage environment (Windows NT/2000, Netapp Filer, SAN - Sun, Hitachi). Need to track who is the 'disk hog', duplicate files, keep historical data, etc.. I'm sure there's more to it. Anyways, have not dealt with any storage management solution so any suggestion will help. thanks, Marcus Bui

1 0

FAS960/Solaris/Veritas/Oracle
by Brent.Killion＠pinnaclewest.com 12 Mar '03

12 Mar '03

Toasters- Is there anyone out there using FAS960/Solaris/Veritas Foundation Suite/Oracle yet? We are getting ready to begin testing/implementation and I'm looking for some lessons learned specifically on using Snapshots with Oracle backups and LUNs. Thanks, Brent Killion

1 0

RE: Weired Linux error
by Tom Felbinger 12 Mar '03

12 Mar '03

Thanks!!! -----Original Message----- From: Kade, Mike [mailto:Mike.Kade@netapp.com] Sent: Wednesday, March 12, 2003 11:06 AM To: tom.felbinger(a)synopsys.com Subject: RE: Weired Linux error Oops... Here it is. http://www.fys.uio.no/~trondmy/src/ Mike ---------------------------------------------------------------------------- ------- Serving Data By Design - Network Appliance, Inc Michael Kade Technology Evangelist (480) 510-8002 mkade(a)netapp.com -----Original Message----- From: Tom Felbinger [mailto:tom.felbinger@synopsys.com] Sent: Wednesday, March 12, 2003 8:48 AM To: Kade, Mike Subject: RE: Weired Linux error Mike, Thanks, but I'm not finding the link you mentioned. Just to be clear, all my nfs mounts are thru the automounter, both autofs and amd. autofs uses udp by default, but amd used tcp by default. I modified the /etc/amd.net file to use udp instead. If memory serves me correctly, I did interactively mount the filer using tcp and still encountered the data corruption. I've very interested in resolving this and look forward to the link. Thanks to all for their input. Tom -----Original Message----- From: Kade, Mike [mailto:Mike.Kade@netapp.com] Sent: Tuesday, March 11, 2003 3:18 PM To: tom.felbinger(a)synopsys.com Subject: RE: Weired Linux error Tom, The following link is for the approved patches for NFS (this is the home website for the developer/maintainer of Linux NFS). A lot of these patches were actually done by NetApp. NetApp has spent quite a bit of time looking at Linux issues with NFS over the last year. A lot of this work shows up in these patches. Regards, Mike ---------------------------------------------------------------------------- ------- Serving Data By Design - Network Appliance, Inc Michael Kade Technology Evangelist (480) 510-8002 mkade(a)netapp.com -----Original Message----- From: Tom Felbinger [mailto:tom.felbinger@synopsys.com] Sent: Tuesday, March 11, 2003 7:04 AM To: toasters(a)mathworks.com Subject: RE: Weired Linux error A couple folks mentioned running NFS over TCP for linux. However, I've encountered data corruption (i.e. running a filer hosted binary will core dump) with RH 7.1 and 7.2 with NFS over TCP. Is there some patch/kernel rev/something I need to apply??? I stopped pursuing this because RedHat and NetApp said this was a known issue. Sounds like some people have it working though. Tom -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com]On Behalf Of Steve Losen Sent: Monday, March 10, 2003 6:04 PM To: toasters(a)mathworks.com Subject: Re: Weired Linux error > A few RedHat 7.2 nfs clients mounting a F840 volume via automount error out with > the following. I have looked into everyting including NIS timeouts, filer > sysstat, network performance and everything looks to be fine. Intrestingly all > the other non linux Clients accessing the same auto-mount are working perfectly. > > > Mar 9 06:35:12 linuxclient kernel: nfs: server filer OK > Mar 9 06:35:13 linuxclient last message repeated 52 times > Mar 9 06:35:31 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:35:41 linuxclient kernel: nfs: server filer OK > Mar 9 06:35:54 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:36:34 linuxclient kernel: nfs: server filer OK > Mar 9 06:36:46 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:38:36 linuxclient kernel: nfs: task 41929 can't get a request slot > Mar 9 06:38:41 linuxclient kernel: nfs: task 43328 can't get a request slot Be sure you are mounting with the "tcp" flag. We have found that a Linux NFS v3 client using udp has this problem, but it goes away if you use tcp. I think Linux defaults to udp. You may need to enable tcp on the filer with options nfs.tcp.enable on

1 0

RE: Weired Linux error
by Tom Felbinger 12 Mar '03

12 Mar '03

From: Tom Felbinger [mailto:tom.felbinger@synopsys.com] Sent: Wednesday, March 12, 2003 10:48 AM To: Kade, Mike Subject: RE: Weired Linux error Mike, Thanks, but I'm not finding the link you mentioned. Just to be clear, all my nfs mounts are thru the automounter, both autofs and amd. autofs uses udp by default, but amd used tcp by default. I modified the /etc/amd.net file to use udp instead. If memory serves me correctly, I did interactively mount the filer using tcp and still encountered the data corruption. I've very interested in resolving this and look forward to the link. Thanks to all for their input. Tom -----Original Message----- From: Kade, Mike [mailto:Mike.Kade@netapp.com] Sent: Tuesday, March 11, 2003 3:18 PM To: tom.felbinger(a)synopsys.com Subject: RE: Weired Linux error Tom, The following link is for the approved patches for NFS (this is the home website for the developer/maintainer of Linux NFS). A lot of these patches were actually done by NetApp. NetApp has spent quite a bit of time looking at Linux issues with NFS over the last year. A lot of this work shows up in these patches. Regards, Mike ---------------------------------------------------------------------------- ------- Serving Data By Design - Network Appliance, Inc Michael Kade Technology Evangelist (480) 510-8002 mkade(a)netapp.com -----Original Message----- From: Tom Felbinger [mailto:tom.felbinger@synopsys.com] Sent: Tuesday, March 11, 2003 7:04 AM To: toasters(a)mathworks.com Subject: RE: Weired Linux error A couple folks mentioned running NFS over TCP for linux. However, I've encountered data corruption (i.e. running a filer hosted binary will core dump) with RH 7.1 and 7.2 with NFS over TCP. Is there some patch/kernel rev/something I need to apply??? I stopped pursuing this because RedHat and NetApp said this was a known issue. Sounds like some people have it working though. Tom -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com]On Behalf Of Steve Losen Sent: Monday, March 10, 2003 6:04 PM To: toasters(a)mathworks.com Subject: Re: Weired Linux error > A few RedHat 7.2 nfs clients mounting a F840 volume via automount error out with > the following. I have looked into everyting including NIS timeouts, filer > sysstat, network performance and everything looks to be fine. Intrestingly all > the other non linux Clients accessing the same auto-mount are working perfectly. > > > Mar 9 06:35:12 linuxclient kernel: nfs: server filer OK > Mar 9 06:35:13 linuxclient last message repeated 52 times > Mar 9 06:35:31 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:35:41 linuxclient kernel: nfs: server filer OK > Mar 9 06:35:54 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:36:34 linuxclient kernel: nfs: server filer OK > Mar 9 06:36:46 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:38:36 linuxclient kernel: nfs: task 41929 can't get a request slot > Mar 9 06:38:41 linuxclient kernel: nfs: task 43328 can't get a request slot Be sure you are mounting with the "tcp" flag. We have found that a Linux NFS v3 client using udp has this problem, but it goes away if you use tcp. I think Linux defaults to udp. You may need to enable tcp on the filer with options nfs.tcp.enable on

1 0

CIFS getattr load
by Todd C. Merrill 11 Mar '03

11 Mar '03

Instead of standing at a crossroads or a fork in the road, I'm standing in an open field, and need somebody to point me in a direction... My filers are showing loads and loads of CIFS getattr operations, on the order of 80% of the CIFS traffic, which is 80-90% of the total NFS and CIFS traffic. Finding out why this is so, or alleviating it or reducing it so the load on the filers goes down, is my goal. The filers are used in a software development environment, in an automated testing environment (generally no interactive usage by people). The CIFS/getattr load is most imbalanced not when a software build is happening but while testing suites are run. On the order of only 10 Windows clients (generally Win2k) can nearly cripple an F840. So, at this point, any WAG's are welcome. Do directories with large numbers of files stress the getattr operations? Has anyone run across any applications that are very "chatty" with filers? Has anyone run into any bugs (with ONTAP or with Windows) that might show such a symptom? Perhaps the application mix we are running is just this way, but something in my gut tells me something is amiss. I have a sniffer available, but before I get to that detail, I was hoping to mine the wisdom of this list, to approach this vague problem armed with some good ideas. Until next time... The MathWorks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

1 0

ClearCase 5.x on NetApp, mixed NFS/CIFS setup issues
by John Stoffel 11 Mar '03

11 Mar '03

Hi all, We're starting to roll out a new ClearCase instance here, and we're taking advantage of the support between Rational and Network Appliance to put all the data onto our F760 running 6.3.1 with both NFS and CIFS licenses. The VOB and View servers just run the lock managers and other processes. The clients (NT and Linux) just access the storage directly from the NetApp after getting the info they need from the VOB/View servers. We've setup a seperate volume for ClearCase (CC) VOBs and VIEWs with mixed security settings. The problem is that running as root on one of the ClearCase servers, we cannot remove a file that was created by a user running on a PC with CC client software. Our infrastructure is pretty simple, with just one NIS domain and one NT domain to work with. But because our corporate IT group won't give us a short group name in the Win2K domain, we're forced to use the unix username of 'mss-vobadm' with an NT username of NA05\sa106068ccalbd. We've got a usermap.cfg as shown below. Does anyone have any hints on how we can fix this up, or track down the proper setup for this environment? NA05\sa106068ccalbd => mss-vobadm NA05\sa106068ccalbd <= root And we've tried to use the following cifs options: > options cifs cifs.audit.enable off cifs.audit.file_access_events.enable on cifs.audit.logon_events.enable on cifs.audit.logsize 524288 cifs.audit.saveas /etc/log/adtlog.evt cifs.bypass_traverse_checking on cifs.comment cifs.guest_account cifs.home_dir cifs.home_dir_namestyle cifs.idle_timeout 1800 cifs.max_mpx 50 cifs.netbios_aliases cifs.netbios_over_tcp.enable on cifs.nfs_root_ignore_acl on cifs.oplocks.enable on cifs.oplocks.opendelta 8 cifs.per_client_stats.enable off cifs.perm_check_ro_del_ok on cifs.perm_check_use_gid on cifs.restrict_anonymous.enable off cifs.save_case on cifs.scopeid cifs.search_domains cifs.show_snapshot off cifs.shutdown_msg_level 2 cifs.sidcache.enable on cifs.sidcache.lifetime 1440 cifs.snapshot_file_folding.enable off cifs.symlinks.cycleguard on cifs.symlinks.enable on cifs.trace_login off cifs.wins_servers X,Y,Z Along with the following for share security: tswtoast> cifs shares Name Mount Point Description ---- ----------- ----------- vobstg /vol/ccvol/vobstg everyone / Full Control mss-ccase{g} / Full Control mss-eng{g} / Full Control NA05\sa106068ccalbd / Full Control viewstg /vol/ccvol/viewstg everyone / Full Control mss-eng{g} / Full Control NA05\sa106068ccalbd / Full Control clearcase /vol/vol0/clearcase everyone / Full Control Any hints or help would be appreciated. Thanks, John John Stoffel - Senior Unix Systems Administrator - Lucent Technologies stoffel(a)lucent.com - http://www.lucent.com - 978-399-0479

2 1

RE: Snmp traps
by Yee, Steven 11 Mar '03

11 Mar '03

netapp.mib describes the Network Appliance custom MIB which includes the built-in traps that a NetApp appliance may send (assuming you have snmp configured correctly) traps.dat is an *aid* that provides a short name to numeric OID mapping for most of the netapp.mib variables (its not so good for table entries) that can be used if you wish to create custom, user-defined snmp traps via the 'snmp traps' command. steve. -----Original Message----- From: Rune Bakken [mailto:rb@telenor.net] Sent: Tuesday, March 11, 2003 6:41 AM To: Asbjørn Bjørkelund Cc: toasters(a)mathworks.com Subject: Re: Snmp traps Asbjørn Bjørkelund <asbjorn.bjorkelund(a)proact.no> writes: > Hello Toasters > > We are looking for some help for defining snmp traps, or to be more correct > is there anybody who has already created > lists with traps. i'm looking for 6-7 traps for example for Clusters: this > filer is dead, Clusters: CFO this filer cannot takeover, CFO partner may be > down, > disk failed, power supply failed. Have you checked the files under filer:/etc/mib/ netapp.min traps.dat ?? >>>>>>.rune -- Rune Bakken Senior System Administrator Telenor Business Solutions (Nextra), ManagedServices/*/IP-Services http://ips.telenor.net/

1 0

CIFS shares and auditing
by Geoff Hardin 11 Mar '03

11 Mar '03

Toasters, Let me start by stating that I am a UNIX admin and I administer my filers strictly by command line or scripts we've written. Most of our data is shared via NFS with a few CIFS shares setup on only a couple filers. That said, a user want a log of who accesses his share and which files were viewed/changed. Reading the man pages and the admin guide for ONTAP 6.1, it appears that the best/only solution is the cifs.audit.enable option. I have enabled this per the instructions in the System Administrators Guide (http://now.netapp.com/NOW/knowledge/docs/ontap/rel61r1/html/sag/cifs36.htm#…). According to this document, you can turn auditing on in five simple steps. My problem is with step number 4: "Activate access auditing for individual files and directories according to your Windows documentation." Since I'm a UNIX admin, I don't understand the Windows aspect of filer administration. I tried getting our NT admins to take a look at it, but they say they can't set the file and directory auditing on the filer using the User Manager tool (the menu items under the Policies menu are unselectable). I have tried creating a non-root, administrative account using the useradmin console command, but that left them with fewer rights on the filer than as the domain admin. Any help or pointers would be greatly appreciated. Thanks, Geoff Hardin geoff.hardin(a)dalsemi.com This space is for rent in order to increase company revenue.

1 0

Weired Linux error
by Premanshu Jain 11 Mar '03

11 Mar '03

A few RedHat 7.2 nfs clients mounting a F840 volume via automount error out with the following. I have looked into everyting including NIS timeouts, filer sysstat, network performance and everything looks to be fine. Intrestingly all the other non linux Clients accessing the same auto-mount are working perfectly. Mar 9 06:35:12 linuxclient kernel: nfs: server filer OK Mar 9 06:35:13 linuxclient last message repeated 52 times Mar 9 06:35:31 linuxclient kernel: nfs: server filer not responding, still trying Mar 9 06:35:41 linuxclient kernel: nfs: server filer OK Mar 9 06:35:54 linuxclient kernel: nfs: server filer not responding, still trying Mar 9 06:36:34 linuxclient kernel: nfs: server filer OK Mar 9 06:36:46 linuxclient kernel: nfs: server filer not responding, still trying Mar 9 06:38:36 linuxclient kernel: nfs: task 41929 can't get a request slot Mar 9 06:38:41 linuxclient kernel: nfs: task 43328 can't get a request slot

6 7

Re: Snmp traps
by Rune Bakken 11 Mar '03

11 Mar '03

Asbjørn Bjørkelund <asbjorn.bjorkelund(a)proact.no> writes: > Hello Toasters > > We are looking for some help for defining snmp traps, or to be more correct > is there anybody who has already created > lists with traps. i'm looking for 6-7 traps for example for Clusters: this > filer is dead, Clusters: CFO this filer cannot takeover, CFO partner may be > down, > disk failed, power supply failed. Have you checked the files under filer:/etc/mib/ netapp.min traps.dat ?? >>>>>>.rune -- Rune Bakken Senior System Administrator Telenor Business Solutions (Nextra), ManagedServices/*/IP-Services http://ips.telenor.net/

1 0

RE: Weired Linux error
by Touretsky, Gregory 11 Mar '03

11 Mar '03

What are your mount options? We run NFS over TCP on all our machines, including hundreds of RH7.1 Linux boxes. No data corruption so far (or at least, we hope so ;-) -- Gregory -----Original Message----- From: Tom Felbinger [mailto:tom.felbinger@synopsys.com] Sent: Tuesday, March 11, 2003 4:04 PM To: toasters(a)mathworks.com Subject: RE: Weired Linux error A couple folks mentioned running NFS over TCP for linux. However, I've encountered data corruption (i.e. running a filer hosted binary will core dump) with RH 7.1 and 7.2 with NFS over TCP. Is there some patch/kernel rev/something I need to apply??? I stopped pursuing this because RedHat and NetApp said this was a known issue. Sounds like some people have it working though. Tom -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com]On Behalf Of Steve Losen Sent: Monday, March 10, 2003 6:04 PM To: toasters(a)mathworks.com Subject: Re: Weired Linux error > A few RedHat 7.2 nfs clients mounting a F840 volume via automount error out with > the following. I have looked into everyting including NIS timeouts, filer > sysstat, network performance and everything looks to be fine. Intrestingly all > the other non linux Clients accessing the same auto-mount are working perfectly. > > > Mar 9 06:35:12 linuxclient kernel: nfs: server filer OK > Mar 9 06:35:13 linuxclient last message repeated 52 times > Mar 9 06:35:31 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:35:41 linuxclient kernel: nfs: server filer OK > Mar 9 06:35:54 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:36:34 linuxclient kernel: nfs: server filer OK > Mar 9 06:36:46 linuxclient kernel: nfs: server filer not responding, still > trying > Mar 9 06:38:36 linuxclient kernel: nfs: task 41929 can't get a request slot > Mar 9 06:38:41 linuxclient kernel: nfs: task 43328 can't get a request slot Be sure you are mounting with the "tcp" flag. We have found that a Linux NFS v3 client using udp has this problem, but it goes away if you use tcp. I think Linux defaults to udp. You may need to enable tcp on the filer with options nfs.tcp.enable on

1 0

Jump to page:

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