toasters

toasters@lists.teaparty.net

13513 discussions

don't run an F630 with the case open
by Tom Yates 21 May '98

21 May '98

d'oh of the day contest entry, just fyi: you can run an F330 with the case open, or at least ajar. i often do after i've done work, before i seal the case down tight, to make sure that it boots and i'm not going to have to undo everything to fix it. well, don't try the same trick with an F630; it doesn't work. thanks to rich ciammaichella at netapp for not roasting me when we discovered that this was a major part of my F630 booting problem. for those who haven't looked closely inside an F630, there are no longer cables going from the board to the front-of-case displays; there are now fixed connectors that seat when the board tray is slid fully home. this may be true for other high-end models, too - i don't know. Tom Yates - Unix Chap - The Mathworks, Inc. - +1 (508) 647 7561 MAG#65061 DoD#0135 AMA#461546 1024/CFDFDE39 0C E7 46 60 BB 96 87 05 04 BD FB F8 BB 20 C1 8C

1 0

NTP or rdate.....
by gordon＠nabaus.com.au 20 May '98

20 May '98

Yes, put me on the voting list too. Netapp folk have been misteriously quiet regarding this thread. We know nearly ALL of the netapp dudes read from this list, so how about a response from Netapp? What is Netapps official position regarding NTP? That's my 2cents worth. The check is in the mail.... .... Gordon. Gordon Maudsley, Unix and Network dude, National Australia Bank.

1 0

Re: NetApp File Server (FS230) mounted directly on an NT server
by Keith Brown 19 May '98

19 May '98

Thomas... Taking your various issues one at a time... >It looks like while yes >there is indeed NT server authentication happening now, no matter WHO I >want to access the CIFS shares, they STILL have to MANUALLY PLACED into >the /etc/passwd file!!!! Right. That is the case in current Data ONTAP releases, and the root reason for this requirement is that the security model used by the WAFL file system today is a UNIX security model. In order to assess what a Windows user can and can not do to the files and directories resident on its file system, a filer has to associate each Windows connection (user) with a UNIX identity (a UID and GID(s)) so it can control the users access to data using the UNIX inode based security that it stores (read, write and execute for owning UID, owning GID and other). The /etc/passwd isn't really storing Windows users accounts per se, it's storing mappings between account names and UIDs/GIDs. >What is the POINT of full blown NT server >(domain) authentication if I STILL have to input >ALL of the users into the /etc/passwd file! The Windows clients still get to authenticate themselves in the normal way with the domain infrastructure on startup. The full Windows security/authentication protocols are enforced on the wire, and the users also wind up being subject to all the correct limitations and restrictions that the Windows administrator may have assigned to their account (times of day they are allowed to use the network and so forth...). >We have TONS of >UNIX users across campus who are on various NIS+ domains >and as a result many have overlapping UIDs! So called "overlapping UIDs" situations are not handled by the above mechanism today. If by overlapping UIDs you mean multiple users posessing the same UID, I thought this was generally regarded as a pretty bad idea even to start out with? Unless the accounts with identical UIDs have different lexigraphical user names, that's that. If they do, you could actually get to "break up" the different users with the same UIDs by allocating them different UIDs in the filers /etc/passwd file. However, if you configure the filer into your existing NIS infrastructure as a client, then the problem will remain. As an FYI, we have a utility available on our NOW web site that many of our customers have found useful. Check out CVTUSERS in the "Tools & Utilities" section of NOW. It will automatically generate passwd and group databases from a Windows NT domain infrastructure for you. >Is the ONLY way to get around this is by going to 5.x? Not a current 5.X (5.0) release, no. The resolution to this "problem" as described above is actually quite a bit larger than you might imagine, and is embodied in a major enhancement to the Data ONTAP microkernel and WAFL file system that is already announced and about to be unleashed shortly. Specifically, the solution lies in enhancing WAFL to maintain true Windows NT style security attributes (NTFS Security Descriptors) such that Windows client users can be treated as exactly what they are, Windows client users who might not even have UNIX accounts at all! :-) The new capability is briefly mentioned in the announcement of our previous Windows Network Software enhancement (WNS 2.0) at: http://www.netapp.com/news/level3b/news_rel_971103.html The new software is currently in external test, and has even been briefly reviewed by Windows NT Systems magazine in their article at: http://www.netapp.com/news/level3b/news_rel_971103.html (See the "Sidebar: Midyear Release" box toward the end of the article). I'll save all the details for another time, but suffice to say that what you have coming your way is not just a "resolution", it's a whole new solution in its own right. :-) >Is there any way to MOUNT a NetApp file server directly to an NT file >system and access it like unix does with a /mnt directory? Well, yes. But apparently you already know about it because you write: >I know about the menthod of mounting to NT via a drive letter... but >you can't re-share that as a resource under that the NT file system CAN >you? You mean "re-sharing" a network drive or the directories thereon? No, you can't do that with current versions of Windows NT, at least not without resorting to some undocumented tricks which nobody has ever been able to confirm to me *really* exist. It's like re-exporting NFS mounted file systems on a UNIX machine, arguably not a good idea and definately not allowed. >You know.. like this: > given that > NetApp box="\\barium\C$" > > and > NT Server="\\ntbox" > >net use \\ntbox\shares\share \\barium\share >net use \\ntbox\shares\winapps \\barium\winapps What? Now there's a "net use" syntax that I've never seen before. >Anyone wanna tell me exactly why things can't be done like this? Because it would be a syntax error? :-) Keith

1 0

Re: command reference guide discontinued in 5.x
by applix!jimsparc!jim＠uunet.uu.net 19 May '98

19 May '98

>command reference guide has been discontinued in 5.x? you're kidding, right? You have to be kidding! -- Jim Morton, Applix Inc., Westboro, MA jim(a)applix.com

1 0

Re: Support for Network Time Protocol?
by mer＠world.evansville.net 19 May '98

19 May '98

Another vote for NTP support. It doesn't need to be full-fledged -- client-only is fine. The rsh/rdate thing is a kludge. -- Marc Rouleau VP and Chief Technology Officer Voice: (812) 479-1700 Fax: (812) 479-3439 Network WCS http://www.networkwcs.com

5 4

As long as we're voting
by Amy Chused 19 May '98

19 May '98

I'd vote for NTP too. Amy Chused Server Administration GTE Internworking

1 0

Re: Support for Network Time Protocol?
by Chris March : TTSG : 781-393-3236 19 May '98

19 May '98

I'll add my vote to get support for NTP. Chris March Thomson Technology Services Group

1 0

Re: Support for Network Time Protocol?
by Don Glascock 19 May '98

19 May '98

> Another vote for NTP support. ... And another: we too have asked for NTP support, but have not heard when this will occur. Our toaster has fulfilled its charter well ("a box that encapsulates file service for NFS & CIFS clients") except in this regard: a file's modification time can be critical in certain situations, and therefore must be accurate. Don Glascock -- Donald S. Glascock Mailstop GU-1016 Special Purpose Processor Development Group Mayo Foundation Rochester, MN 55905 Glascock.Donald(a)Mayo.EDU +1.507.284.1850 +1.507.284.9171 (fax) http://www.mayo.edu/sppdg/sppdg_home_page.html "No matter where you go, there you are."

1 0

Re: syslog and /etc/messages format
by Mark D Simmons 19 May '98

19 May '98

(after with much rude elision) Jason wrote: > We currently use logsurfer and swatch to parse syslog messages > generated by our F330. We also use procmail to forward "critical" > email from the filer to alpha-pagers. We use swatch too. > There isn't an appendix which contains a _complete_ > list of all error messages and sample _syntax_ in the filer > documentation. Tech Support couldn't even get me a list of all the > possible email subject headers the filer might send in case of error! You know that's something that IBM and Digital used to do religiously, and it's in most good books on the subject of Technical Documentation. One wee book or greppable ASCII file on-line that contains the full gamut of error messages (even if what the message means is sanitised to 'phone NetApp' for some of the really weird ones) would be great. > So, I'm all for more specific messages -- as long as they are > documented. Please. Ditto.

1 0

Re: Support for Network Time Protocol?
by Mark D Simmons 19 May '98

19 May '98

We also asked for NTP support a while back. I've currently got all 9 main filers rdating to an NTP'd Unix box every hour and so far everything is okey, but proper NTP support would reduce the niggly tasks on the admin server and my confidence in the time on the filers. 2p

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