toasters

toasters@lists.teaparty.net

13531 discussions

Re: Samba 2.0 as Windows PDC ?
by Jason D. Kelleher 28 Jan '99

28 Jan '99

In message <199901281940.NAA21281(a)vortex.more.net>, David Drum writes: >Quoth Mark Muhlestein: > >> When LDAP support is available it should help the account synchronization >> problem greatly. We are planning to provide that in the Win2000 timeframe >. > >I don't think I will live that long. > >Regards, > >David K. Drum >david(a)more.net Don't worry, I hear there are cryogenics companies that will freeze your body and revive you when Win2000 ships... :) jason

1 0

Re: Samba 2.0 as Windows PDC ?
by Mark Muhlestein 28 Jan '99

28 Jan '99

When LDAP support is available it should help the account synchronization problem greatly. We are planning to provide that in the Win2000 timeframe. Mark Muhlestein -- mmm(a)netapp.com -----Original Message----- From: Daniel Quinlan <quinlan(a)transmeta.com> To: Keith Brown <keith(a)netapp.com> Cc: quinlan(a)transmeta.com <quinlan(a)transmeta.com>; toasters(a)mathworks.com <toasters(a)mathworks.com> Date: Thursday, January 28, 1999 12:10 AM Subject: Re: Samba 2.0 as Windows PDC ? >Keith Brown <keith(a)netapp.com> writes: > >>> Does anyone have any experience using Samba 2.0 as a PDC? > >> We don't officially support the use of non-Microsoft domain >> controllers at this time. Sorry... > >Ahem, you didn't actually answer my question. > >>> Also, what products (like Samba 2.0), if any, are people using to >>> synchronize logins between Unix (NIS) and Windows (NT PDC), >>> especially in a mixed environment using NetApp filers. > >> We have one of our own, and it's free! :-) >> [...] the "SecureShare Account Migrator" tool. > >It looks like that tool only lets you go from NT to Unix, which >unfortunately is not useful for us. I was hoping for a program that >allows NT to stay synchronized with an NIS domain (or just /etc/passwd >and /etc/group). > >- Dan >

2 1

RE: Tape Robots and "autoloading".
by Genrich, James R 28 Jan '99

28 Jan '99

Steve & Others, There are a few questions to be answered from the previous information. What is the OS being run on the F630? Is the Q2/15 an Autoloader of Robot version. Based upon a few assumptions there are three options. 1. If cost is the major consideration for using the Dump command as opposed to Backup Software there are a couple of options. First, you can use the control panel on a 9730 to mount drives to certain drives ( I am not sure if this is available in the Q2/15). This is not very automated since an individual will have to be present to use the control panel for mounts and unmounts. Secondly, you could write a script to do this, however this will be time consuming and it is reinventing the wheel (ie backup software is already written and tested) 2. To be able to perform simultaneous backup of 2 or more servers a backup software can be used along with STK ACSLS software (This software simply controls the robot and shares it between multiple servers or multiple backup packages). ACSLS needs a stand alone SUN or IBM box to run on (Sparc 10 or Small AIX server) for UNIX and needs Libattach software from STK if NT is involved. 3. Backup software can be installed on one of the server and the additional servers can be backed up over the network simultaneously using multiplexing capabilities common in most backup software. Based upon the current information these are the possible solutions. Jim Genrich MPSE Storage Technology, Inc. ---------- From: grdodson(a)lexmark.com [SMTP:grdodson@lexmark.com] Sent: Wednesday, January 27, 1999 8:29 PM To: toasters(a)mathworks.com Subject: Tape Robots and "autoloading". We are purchasing a F760 and are faced with the usual backup dilemma. To backup our F630 we have a single drive DLT robot (Breece Hill Q2.15) and use local dumps. This takes 15+ hours a day (3 tapes) and obviously doesn't scale well. So we are investigating larger robots with multiple tape drives. Currently the StorageTek 9730 is looking interesting. The problem we have is by using local dumps we need the tape robot to autoload the drives. Moreover with a multi drive robot the autoload function needs to be able to control which slots are loaded into which drives. Anyone know of the StorageTek supports this? Any other recommendations multi drive 28+ slot tape robots are also welcome. FYI. The reason we aren't just stacking two Breece Hill Q2.15's is that we have been underwhelmed with the usability and quality of these boxes. We find the documentation and user interface to be quite poor. (And they never delivered the S/W upgrade they promised) The reason we aren't using an NDMP backup S/W package is that the nature of the data on this box requires a "level 0" backup every night. Most of the features of the NDMP S/W would never be needed (indexing, etc.) and therefore be an unnecessary expense. Graydon Dodson (606) 232-6483 grdodson(a)lexmark.com Lexmark International Inc.

1 0

Re: multiple CIFS HOME directories?
by Mark Muhlestein 28 Jan '99

28 Jan '99

Graham writes: >> % cd /filer/vol0/home >> % ln -s /vol/vol5/home/joe ./joe >> % ln -s /vol/vol5/home/bill ./bill > >We do this and it works very well. One thing that has always confused me >though - the cifs.homedir works - but if i set up /filer/vol0/home as a share and >browse to it all i see are the unresolved links. How come it >can do one and not the other? It'd be nice to be able to browse >to one spot on the filer and see all the home dirs.... For now that won't work because a symlink can't be followed out of a share (for security reasons). We're working on a way to get around that restriction. Mark Muhlestein -- mmm(a)netapp.com

1 0

quotas not working on clients
by George Kahler 28 Jan '99

28 Jan '99

Quota (quota [ -v ] [ username ] on all of our client machines stopped reporting user's disk quota. The problem was reported to NetApp but the answer I got back was that they were able to reproduce the 'problem' and that it works as designed. :-( Here is the background. At our university we have two toasters. A F520 and a F330 (Data ONTAP 4.3.3) that are part of a complex of machines used for providing electronic mail for staff and students (nfs). There are about 40,000+ accounts. The F330 used to contain the spool and the F520 contained two home directories (quota trees), one for staff and the other for students. Quota for spool (F330) were none. Quota for the individual home directories were set. The /etc/quotas file looked something like this /home_sta tree 39G /home_stu tree 20G # /home_sta/j/joe user 40M .. /home_stu/a/stud1 user 5M /home_stu/z/stud99 user 6M .. For performance reasons the spool was migrated from the F330 to the F520 and the /home_stu directory was migrated from the F520 to the F330. The F330 has only the one quota tree, /home_stu, where quotas are still defined as shown above. The F520 has two quota trees, /spool and /home_sta, where we don't want any individual quotas for the spool, however, we do need individual quotas on the /home_sta directories. The new /etc/quotas file looks like this /spool tree 25G /home_sta tree 39G # /home_sta/j/joe user@home_sta 45M /home_sta/x/fred user@home_sta 90M ... Now, all of the client machines (SunOS, Solaris 2.6) stopped reporting any user quotas. The quotas ARE still being enforced by the toaster, however, when a user queries his/her quota they get back just the headings, % id uid=101(dummy) gid=277(employee) % pwd /home/sta/x/dummy % quota -v Disk quotas for dummy (uid 101): Filesystem usage quota limit timeleft files quota limit timeleft % If, on the other hand, your home directory is on the F330 which contains only the one quota tree, the quotas do get reported correctly. % id uid=195(dummy1) gid=202(mailuser) % pwd /home/stu/t/dummy1 % quota -v Disk quotas for dummy1 (uid 195): Filesystem usage quota limit timeleft files quota limit timeleft /home/stu 72 5120 5120 32 10240 10240 % Any help would be greatly appreciated and I apologize for the long and winded description of the problem. George ------------------------------------------------------------------------------- George Kahler e-mail: george(a)yorku.ca UNIX Systems Administrator humans: (416) 736-5257 x.22699 Computing Services, York University machines: (416) 736-5830 Toronto, Ontario, Canada, M3J-1P3

2 1

RE: Samba 2.0 as Windows PDC ?
by Allison, Bridget 28 Jan '99

28 Jan '99

>Does anyone have any experience using Samba 2.0 as a PDC? I don't know of any filer customer using Samba's PDC functionality successfully with the filer. I do know that based on my periodic checks with the Samba team on just when it makes sense for us to sink our engineering time into making sure we work with a Samba PDC, it's too early to start that work right now. The PDC support is not production quality - that's the word from the Samba team rather than mine. So, we do intend to work with a Samba PDC, it's part of our plan and we're trying to make sure the Samba side is fully cooked before we start our integration work. In the meantime, if there are customers using it successfully they haven't told me. They certainly couldn't use it with the NTFS style security today - Samba 2.0 doesn't have any SIDS to return to the filer. Since that's the case today, I'm presuming your question is more around the user authentication per se - just one aspect of PDC functionality but the primary one Samba does right now. >Also, what products (like Samba 2.0), if any, are people using to >synchronize logins between Unix (NIS) and Windows (NT PDC), especially >in a mixed environment using NetApp filers. Keith suggested our one-way tool, the MS Services for Unix is another one-way tool. I've seen some discussion on this topic on one of the Samba mailing lists - I think it was the samba-ntdom list - just this past week or so. You might want to check out the archives http://us1.samba.org/samba/archives.html If you find something you'd recommend then please let me know - we'd be interested in finding something to recommend to customers. Bridget ----------------------- Network Appliance Inc. 408 367 3419

1 0

multiple CIFS HOME directories?
by Daniel Quinlan 28 Jan '99

28 Jan '99

Is it possible to have multiple default shares for CIFS homes? I am running 5.1.2P2. For example, let's say I have two quota trees on a NetApp. One is /vol/vol0/home_eng and the other is /vol/vol0/home_finance. Let's say home_eng is exported via NFS (for engineers) and home_finance is not (because it contains sensitive information such as employee salaries). I want to allow people to access both via CIFS. It is assumed that there will not be any username collisions between the two hierarchies. So if there is a /home_eng/bob directory, there won't be a /home_finance/bob directory (and vice versa). Can what I am trying to accomplish be done? (People like being able to map \\filer\myname to a drive letter. They don't like having to do stuff like \\filer\something -> W:\ and then having to go to W:\myname.) Thanks. - Dan

2 1

Re: Samba 2.0 as Windows PDC ?
by Keith Brown 28 Jan '99

28 Jan '99

>Does anyone have any experience using Samba 2.0 as a PDC? We don't officially support the use of non-Microsoft domain controllers at this time. Sorry... >Also, what products (like Samba 2.0), if any, are people using to >synchronize logins between Unix (NIS) and Windows (NT PDC), especially >in a mixed environment using NetApp filers. We have one of our own, and it's free! :-) Check out... http://now.netapp.com/tools/ and select the "SecureShare Account Migrator" tool. Keith

3 2

Re: multiple CIFS HOME directories?
by Keith Brown 28 Jan '99

28 Jan '99

>Is it possible to have multiple default shares for CIFS homes? Not at the moment, although some enhancements are currently being made in this area. For people who have UNIX though, there is a workaround. Here's the gig... The idea is that you create a single home directories directory and populate it with symbolic links to the real home directories which can be anywhere on the filer. So, if: options cifs.home_dir {/vol/vol0}/home and joe and bill have home directories in /vol/vol5/home, then from a UNIX client you can: % cd /filer/vol0/home % ln -s /vol/vol5/home/joe ./joe % ln -s /vol/vol5/home/bill ./bill and bill and joe will get to play on the autoshare world even though their home directories are not in the cifs.home_dir directory. You might be forgiven for thinking "What about the not-following-symlinks-out-of-a-share rule?". Well, this doesn't apply here because the symlink *is* the share. Also beware: % cd /filer/vol0/home % ln -s /filer/vol5/home/joe ./joe % ln -s /filer/vol5/home/bill ./bill won't do the job, because if you do this, you set the symlinks to match the NFS *clients* view of where the directories are, not the filer's view. You would then have to engage the /etc/symlink.translations mechanism to translate the symlinks betwixt views, and you probably don't want to get into that. Keith

2 1

Samba 2.0 as Windows PDC ?
by Daniel Quinlan 28 Jan '99

28 Jan '99

Does anyone have any experience using Samba 2.0 as a PDC? Also, what products (like Samba 2.0), if any, are people using to synchronize logins between Unix (NIS) and Windows (NT PDC), especially in a mixed environment using NetApp filers. Thanks for any info or suggestions. - Dan

1 0

Tape Robots and "autoloading".
by grdodson＠lexmark.com 28 Jan '99

28 Jan '99

We are purchasing a F760 and are faced with the usual backup dilemma. To backup our F630 we have a single drive DLT robot (Breece Hill Q2.15) and use local dumps. This takes 15+ hours a day (3 tapes) and obviously doesn't scale well. So we are investigating larger robots with multiple tape drives. Currently the StorageTek 9730 is looking interesting. The problem we have is by using local dumps we need the tape robot to autoload the drives. Moreover with a multi drive robot the autoload function needs to be able to control which slots are loaded into which drives. Anyone know of the StorageTek supports this? Any other recommendations multi drive 28+ slot tape robots are also welcome. FYI. The reason we aren't just stacking two Breece Hill Q2.15's is that we have been underwhelmed with the usability and quality of these boxes. We find the documentation and user interface to be quite poor. (And they never delivered the S/W upgrade they promised) The reason we aren't using an NDMP backup S/W package is that the nature of the data on this box requires a "level 0" backup every night. Most of the features of the NDMP S/W would never be needed (indexing, etc.) and therefore be an unnecessary expense. Graydon Dodson (606) 232-6483 grdodson(a)lexmark.com Lexmark International Inc.

1 0

DEC TZ887 tape drive
by Nigel Metheringham 27 Jan '99

27 Jan '99

Anyone got one of these going on a NetApp (a 630 running 5.2.1)?? Its supposed to be just the same as a Quantum 4700, but initial tests are failing to write on it. The same drive was previously in use successfully on some other system (not NetApp). Do I need to hack the drive type ID - I thought all that was not needed with the latest releases. However 5.2.1 does not appear to be looking at the cloned_tapes in the same way as 5.1 (it didn't work right then either). Using the diag stuff on the mt command, a dump gives the following output:- DUMP: creating "/vol/web/.snapshot/snapshot_for_backup.2" snapshot. DUMP: Using Full Volume Dump DUMP: Date of this level 0 dump: Wed Jan 27 17:54:18 1999. DUMP: Date of last level 0 dump: the epoch. DUMP: Dumping /vol/web/ to rst0a DUMP: mapping (Pass I)[regular files] DUMP: mapping (Pass II)[directories] Wed Jan 27 17:55:30 GMT [netapp2: isp_main]: Tape (4.5):cmdblk 0xfffffc0000ae2ef0 op 0xa block 0 illegal request (5 30, 0) DUMP: Tape write failed. Wed Jan 27 17:55:30 GMT [netapp2: isp_main]: Tape (4.5) unrecognized error 0x5. DUMP: DUMP IS ABORTED DUMP: Deleting "/vol/web/.snapshot/snapshot_for_backup.2" snapshot. All clues gratefully received :-) Nigel. -- [ Nigel.Metheringham(a)theplanet.net - Systems Software Engineer ] [ Tel : +44 113 207 6112 Fax : +44 113 234 6065 ] [ Real life is but a pale imitation of a Dilbert strip ] [ We're recruiting http://www.theplanet.net/profile/recruit.htm ]

1 0

DRP documents...
by eyal traitel 27 Jan '99

27 Jan '99

Has anyone wrote a Disaster Recovery Plan documents for a filer? Can someone share his documents (or at least the technical parts...) ? Yours, Eyal Traitel

1 0

DATA OnTap and WINS...
by eyal traitel 27 Jan '99

27 Jan '99

Hi! Does anyone has experience with NetApp's WINS support and Samba's WINS? I get all the time "No WINS servers are responding; still trying"... Yours, Eyal Traitel

1 0

list tweak
by fubar＠mathworks.com 26 Jan '99

26 Jan '99

I've just tweaked the toasters list, making it a bona fide majordomo list. Hopefully (if I've got it right ;-), we'll stop seeing those darned "sub-scribe" messages to the list... The change may effect any filtering you have. From now on, list mail should appear to be from "owner-toasters(a)mathworks.com", rather than me. If it breaks your setup, sorry. If there are any real problems, feel free to shout at me. ;-) K.

1 0

Fwd: Re: duplicate inode numbers - correct URL
by Andrew Bond 25 Jan '99

25 Jan '99

1 0

duplicate inode numbers
by Tom Reingold 25 Jan '99

25 Jan '99

Hey! My user could not copy a snapshot to the original place because "cp" claimed the files are identical. I think cp thinks so because the inode numbers are the same. The files do not have identical contents, so it seems like a serious bug that the filer issued duplicate inode numbers. Am I wrong? The output belog is wide. Please view it with a wide window. # ls -il myfile .snapshot/*/myfile 2436244 -rw-rw---- 1 pete pete 85933 Jan 24 17:50 .snapshot/hourly.0/myfile 2436244 -rw-rw---- 1 pete pete 85211 Jan 24 15:50 .snapshot/hourly.1/myfile 2436244 -rw-rw---- 1 pete pete 84223 Jan 24 11:56 .snapshot/hourly.2/myfile 971294 -rw-rw---- 1 pete pete 83436 Jan 24 00:47 .snapshot/hourly.3/myfile 971294 -rw-rw---- 1 pete pete 79270 Jan 23 12:57 .snapshot/hourly.4/myfile 971294 -rw-rw---- 1 pete pete 79270 Jan 23 12:57 .snapshot/hourly.5/myfile 971294 -rw-rw---- 1 pete pete 75104 Jan 23 02:42 .snapshot/hourly.6/myfile 971294 -rw-rw---- 1 pete pete 75104 Jan 23 02:42 .snapshot/hourly.7/myfile 971294 -rw-rw---- 1 pete pete 79270 Jan 23 12:57 .snapshot/nightly.0/myfile 971294 -rw-rw---- 1 pete pete 70938 Jan 22 19:41 .snapshot/nightly.1/myfile 971294 -rw-rw---- 1 pete pete 62606 Jan 21 22:20 .snapshot/nightly.2/myfile 2436244 -rw-rw---- 1 pete pete 86541 Jan 24 22:24 myfile Tom Reingold tommy(a)bell-labs.com http://www.bell-labs.com/~tommy Bell Labs, the Research and Development unit of Lucent Technologies Murray Hill, NJ, 07974-0636 US

2 1

Microsoft exchange
by Jason Middlebrooks 24 Jan '99

24 Jan '99

Is Microsoft exchange supported on the netapp filers? Jason Middlebrooks Systems Engineer Datalink Phone: 770-933-9327x260 2264 Northwest Parkway, Suite I Pager: 888-502-1625 Marietta, GA 30067 Fax: 770-933-9272

10 11

Recoverable error rates
by Aiello, Tony 24 Jan '99

24 Jan '99

Hello, There's been some questions of late regarding the meaning of soft or recovered errors. A recovered error is defined as the disk needing to take some action other than the initial access to the media in order to satisfy an i/o request. The methods used may be to retry the request, reseeking to the sector, recreating the data from error correction information or all of the above. Recovered errors do not indicate pending disk failure. Drive manufacturer's specify a bit error rate of so many recoverable errors per bits of access. Note that bits of access includes sector overhead. A 512 byte sector is bracketed by header and trailer information used to keep track of the location as well as error recovery information. While recovered errors do not in themselves indicate pending drive failure, an excessive rate of more than 10 a day on a filer under average load would indicate problems in media management. This condition must be examined more closely to really quantify the error rate. A single sector written with a bit flip will always return a recovered event when read. Naturally. The ECC does not match the stored data so the error correction is invoked to reproduce the proper image. So when examining the recovered error log look at the block address. If each entry references the same block, that is ONE recovered error, regardless of how many entries are made against it. If each one indicates a different address then the amount of traffic must be considered. Using sysstat to look at the megabyte per second rating of the system. Using that determine the megabyte per hour rating. Multiply by 8 to get bits. A reasonable number of for the recovered error rate is 10 per 10^12 bits of access. So consider a disk giving an average response of 20 ms per 4K i/o. It's serving 50 transactions per second or .2 MB/sec. In a 24 hour period then it serves about 1.4 * 10^11 bits. Not far from the recovered bit error rate. So a recovered error every day would not even be excessive. Naturally a filer serving more transactions per second would have a higher rate of expected recovered errors. Realistically, disks don't perform that poorly and may go for long periods of time without reporting a single recovery event. Our primary motivation in logging these events is more for performance. Since error recovery takes time, a disk having a large number of recovered errors could adversely effect performance. So we log these as a means to aid in diagnosing performance problems. The short story is: drives are expected to have these recovered errors. Recovered errors don't indicate a pending failure. Removing a disk because of soft errors wastes time and materials. Futher it exposes you to real damage should some other device experience an unrecoverable error. Tony -------------------------- mailto: taiello(a)netapp.com Ph:(408) 367-3251 Fax: (408) 367-3151

3 2

Re: NFS mail spool [ Was Re: Microsoft exchange ]
by Keith Brown 22 Jan '99

22 Jan '99

Use of NetApp filer technology to act as the storage portion of large scale e-mail installations is a relatively common application of our products. Almost "routine" in fact! :-) A very good technical paper has been written by one of our customers on this subject, which can be perused at: http://www.earthlink.com/about/papers/mailarch.html A similar paper describing the net news application of filers is also available from: http://www.earthlink.com/about/papers/scalenews.html A recently published article on the subject of "Hosting E-Mail for Heavyweights", involving our storage solutions can be viewed at: http://www.internetworld.com/print/1998/10/12/infrastructure/19981012-hosti… tml And finally, one of our own papers on the subject of e-mail performance benchmarking is available from: http://www.netapp.com/technology/level3/3022.html All just FYI... Keith

1 0

RE: Microsoft exchange
by Mohler, Jeff 22 Jan '99

22 Jan '99

While we're on the subject of mail, I was thinking of moving our sendmail spool area from a drive local to the Sun mail server to one of our NetApp boxes. Anyone have any pros/cons regarding NFS mounting the mail spool to the mail server? -- Two words. Dot locking. (procmail)

1 0

Re: Microsoft exchange
by Jason D. Kelleher 22 Jan '99

22 Jan '99

In message <199901221550.KAA20198(a)prospero.dma.isg.mot.com>, Gordon Keegan w rites: >> On Fri, Jan 22, 1999 at 09:47:48AM -0500, Jason Middlebrooks wrote: >> > Is Microsoft exchange supported on the netapp filers? >> >> Actually, I think exchange requires that all data be stored on local >> drives, just like SQL server does. >> >> -- > >While we're on the subject of mail, I was thinking of moving our >sendmail spool area from a drive local to the Sun mail server to >one of our NetApp boxes. Anyone have any pros/cons regarding NFS >mounting the mail spool to the mail server? Stale NFS file locks caused us some problems initially (procmail processes would clog the server waiting on locks), but that was bug #6118 which was fixed in 5.x. Other than that, no problems. And it's nice to be able to pull accidentally deleted messages out of snapshots. jason

1 0

Sybase not SQL
by Robert Black 22 Jan '99

22 Jan '99

Opps! Sorry all my brain was thinking SQL my eyes were seen Sybase. Been that kind of week. Bob ----------------------------------------------------------------- Robert Black Email: bobbl(a)netapp.com Network Appliance, Inc. Voice: (800) 550-6827 Systems Engineer - Pacific NW Pager: (800) 465-7149 -----------------------------------------------------------------

1 0

RE: Microsoft exchange
by Allison, Bridget 22 Jan '99

22 Jan '99

> Actually, I think exchange requires that all data be stored on local > drives, just like SQL server does. That's true today but NetApp is working with Microsoft to change that. Check out http://now.netapp.com/solutions/index.shtml (requires NOW login and password) for more info. If you need more info after that then contact me directly at bridget(a)netapp.com Bridget ============================= Bridget Allison Network Appliance Inc. office: (408) 367 3419 cell: (408) 823 1126

1 0

unsubscribe
by Daniel Long 22 Jan '99

22 Jan '99

unsubscribe

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters