toasters June 1998

toasters@lists.teaparty.net

36 participants
44 discussions

cvtusers question
by agy 11 Jun '98

11 Jun '98

>Is there a way to force CVTUSER to create the user list export both >alphabetically AND with ascending UID's? This way, as we add users to >the filer, we don't have to check for "free UIDs"? Or if we're not even >USING NFS (which we are not) and all we're using is the "cifs.home_dir" >option and nothing else, couldn't eveyone have the SAME UID??? There is a new version of cvtusers scheduled for July which will do what you want, as well as giving you the ability to edit the stuff before it's written to file. Check the NOW site toward mid-July in the Tools and Utilities section. The name of the new version is SecureShare Account Migrator. It is perfectly legal to give all users the same UID. Doing this results in FAT-style security, i.e. none. If that's what you want, go for it. Actually, since you're an NT shop, let me ask you, do you use a Domain Controller? If you do, the filer will be happy to use that for authentication as well, eliminating the conversion problem altogether. Alan -- --------------------------------------------------------------- Alan G. Yoder agy(a)netapp.com Network Appliance, Inc. 2770 San Tomas Expressway voice 408-367-3031 Santa Clara, CA 95051 fax 408-367-3451 ---------------------------------------------------------------

1 0

CVTUSER Questions...
by Weeks, Thomas 11 Jun '98

11 Jun '98

Hey all.. I was planning on using the CVTUSER utils to dump all my NT users into my FS230's (OT V4.3.4) /etc/passwd file in preparation for using the filer as a pure CIFS based home drive for our NT login users... (no NFS) My Q is... Is there a way to force CVTUSER to create the user list export both alphabetically AND with ascending UID's? This way, as we add users to the filer, we don't have to check for "free UIDs"? Or if we're not even USING NFS (which we are not) and all we're using is the "cifs.home_dir" option and nothing else, couldn't eveyone have the SAME UID??? Anyone been here before? Thanx... Tom

1 0

Re: filer-mrtg question
by Mark D Simmons 11 Jun '98

11 Jun '98

Well, I've got our local install of mrtg running with the non-cifs bits of filer-mrtg (I'm using an SGI for this monitoring task and already had an mrtg setup in place...so filer-mrtg was bit narrow-focus for my taste). Basically I lifted the first three parts of the filer-mrtg config file and run mrtg out of cron with them, for each of the filers I look after. I'm looking at the whole MIB for other useful things but my understanding of SNMP is still maturing. mrtg doesn't seem to be the right tool for some of the things I want to do though, along with the fact that MIB doesn't quite seem as simple as it could be. I know people will want me to justify that last statement, so here goes: There are raw figures for the number of, say, NFSPROC_READ calls since the stats were zeroed, AND there are percentages for the number of NFSPROC_READ calls against total NFS ops since stts began. This seems to be unuseful as it can be worked out from the raw figures, and SNMP was always meant to provide the raw data with the least processing at the agent side... What it practically means to me is that the MIB is large and I'm not very good at reading it. SNMP's umpteen different representations and translations are also causing me grief. If anyone's done more than rudimentary things with SNMP and mrtg, or anyother free snmp stuff, and feels like trumpeting their genius I promise to be suitably impressed when you show us how to do it properly! 8) Last time I counted there are 533 basic data (datums would convey the meaning better, but isn't a real word ;) in the MIB and some are only interesting when you perform some arithmetic on them and graph the results, and mrtg is confusing me as to how to get there from here. Probably my fault. Hey ho, back to the grindstone.

1 0

filer-mrtg question
by Alexei Rodriguez 09 Jun '98

09 Jun '98

Has anybody put this in place and have the nfsops code working? I see it is not there, yet graphs get generated for it. I just don't want to spin my wheels if others have it working already. Thanks. Alex

1 0

System RAM for F330 (or used NetApp stuff)
by Brian Marcotte 09 Jun '98

09 Jun '98

Lets try this one more time... NetApp tells me I need the following RAM for my F330: Buffered CAS DRAM, SIMM, 72 Pin, 8Mx36 60NS fast page mode, 4CAS 2K refresh, gold contacts This is not quite PC RAM, so I've not managed to locate a source. Any help locating this RAM would be greatly appreciated! Also, if anyone knows a source for used NetApp equipment, that would be a great help also. Specifically, I'd like parts suitable for a F330, or entire lower-end filers (F330, F210 ,F220, etc). I know that NetApp occasionally has refurbs, but I'd like more sources. Thanks! Note: I already know the dangers of using third party equipment in my filer. -- - Brian

1 0

subscribe toasters
by Madhu Chandra 09 Jun '98

09 Jun '98

subscripe toasters: can you subscribe me. Also can you tell me how to access and use it Madhu Chandra Sys Admin. Numeric Investors

1 0

5.0.1 or later?
by Daniel Quinlan 09 Jun '98

09 Jun '98

Anyone out there using 5.0.1 (or later, especially 5.0.1D1) with NDMP to directly attached SCSI tape drives? If you are, I'd like to hear whether or not you've had problems, or if any problems from earlier releases were resolved for you. Thanks.a - Dan

1 0

RE: Quota's, Qtrees, Old_WAFL, NewWAFL, etc...
by Weeks, Thomas 08 Jun '98

08 Jun '98

Yes... this sounds nice, but we do not use the filers tape drives (at this time) for backups... we are using NFS based Networker (Sun based Legatto product). I imagine after we get away from NFS and go to pure CIFS, we'll be moving our backups to full NT based (to keep any/all CIFS ACLs' intact with out files). Also, I imagine that your implimentation of dump does not allow for single file (or single directory) restorations does it? I realize that there are a couple products out there to that handle ACLS/UFS permissions backup & restoration via NPDM, but we are already running three backup variants right now and adding yet another is just not feasible. Thanx anyway... Thomas Weeks GTE SYSTEMS ENGINEER Testing-Integration Center "The Pit" Brooks AFB, DSN 240-5444 > -----Original Message----- > From: grant(a)netapp.com [SMTP:grant@netapp.com] > Sent: Friday, June 05, 1998 7:06 PM > To: toasters(a)mathworks.com > Subject: Re: Quota's, Qtrees, Old_WAFL, NewWAFL, etc... > > > Hi Thomas, > > Since you mention quota trees, I'd like to jump in and > discuss the benefits of quota trees within 5.1 (shipping > soon) and backup: > > My recommendation is that you should arrange your data in > quota trees whenever possible when running 5.1 for these > reasons: > > 1. You can have different security models running on a per > quota tree basis (ntfs, mixed (multi-protocol) or unix). > > 2. Pass I & II of dump has been optimized for full quota > tree (as well as full volume) backups. > > 3. dump has a new Q option that allows all non-quota tree > data to be backed up. This allows you to frequently > backup quota trees, but perhaps backup /etc and other > miscellaneous data on an infrequent basis and you don't > need to backup the whole volume to do so. > > Cheers, > Grant

2 1

Re: Quota's, Qtrees, Old_WAFL, NewWAFL, etc...
by grant＠netapp.com 06 Jun '98

06 Jun '98

Hi Thomas, Since you mention quota trees, I'd like to jump in and discuss the benefits of quota trees within 5.1 (shipping soon) and backup: My recommendation is that you should arrange your data in quota trees whenever possible when running 5.1 for these reasons: 1. You can have different security models running on a per quota tree basis (ntfs, mixed (multi-protocol) or unix). 2. Pass I & II of dump has been optimized for full quota tree (as well as full volume) backups. 3. dump has a new Q option that allows all non-quota tree data to be backed up. This allows you to frequently backup quota trees, but perhaps backup /etc and other miscellaneous data on an infrequent basis and you don't need to backup the whole volume to do so. Cheers, Grant

1 0

Re: Quota's, Qtrees, Old_WAFL, NewWAFL, etc...
by Keith Brown 05 Jun '98

05 Jun '98

> >I see that you can only use qtree's in the root... >??? - What is the real difference between quotas and qtrees? Filer qtrees are the entities on which you can slap tree quota restrictions in the /etc/quotas file. There are three types of quotas supported; User, group and tree. The former two are largely independent of qtree's, although they can be used in conjunction with them. The three can combine in relatively complicated ways and I'd recommend reading chapter 10 of the Systems Administration Guide for a full explanation. >??? - with 4.3.4 I can do non-root quota's for pure CIFS users right? If you mean quotas on individual Windows/CIFS users that are not associated with qtrees, then yes. Effectively, you would just define a regular user quota for each client. You could use group quotas and default quotas too (default quotas are just special cases of user quotas). >example: > >/cifs_home/<NT_username1> user 100M 20k >/cifs_home/<NT_username2> user 100M 20k >/cifs_home/<NT_username3> user 50M 10k Right. This would be fine. Just remember that the first field, the quota target field for a user quota is ultimately there to specify the UID of the user you are putting a quota on. If you are going to use a directory pathname in this field, the UID for the user quota comes from the owner of that directory. For this to work, the directory /cifs_home/<NT_username1> would have to have an ownership of whatever UID <NT_username1> was equated to via <NT_username1>'s entry in /etc/passwd (or NIS). >??? - These are just quotas on user directories right? No. User quotas are *not* specifically associated with the users home directory. User quotas expressed this way limit the users consumption of file and space resources on the root volume by default. If you want to limit the users space on another volume, you'd use the: /cifs_home/<NT_username1> user@/vol/foo 100M 20K syntax. >??? - I don't need quota trees for this type of individual CIFS user >quotas do I? Not if the above description is what you were looking for. > >The manual says that the user must have an entry in /etc/passwd. >??? - Is this true in a pure NT/CIFS file space? Yes. >??? - If so, Why? (no UID/GID?!) Because the user and group quota mechanism needs the files and directories created on the file systems to be owned by the UID's and GID's it's trying to impose its limits on for things to work. >??? - Will this change with 5.x? Not initially in 5.1 no. Sorry... >Lastly... >??? - Should I wait to impliment quotas until AFTER I go with 5.x or >will my 4.3.4 quotas and qtrees transition over ok to 5.x? Whatever system you come up with in 4.3.4 will transition just fine to 5.X as long as you keep doing things largely the same way. If you start creating multiple volumes in 5.X, then obviously some quota configurations may be effected by that. If you stop routinely mapping Windows/CIFS clients to valid UID's and GID's through /etc/passwd (which becomes perfectly feasible if you are using the NTFS compatible file system security in 5.1), then you will lose user and group quota-ing capability for those users, at least until we sort out some quota enhancements that will let you place quotas on users who are entirely defined by a Windows NT SID. However, tree quotas should keep on truckin' quite nicely, limiting even these kinds of users (the unmapped kind). Keith

1 0

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 June 1998