toasters

toasters@lists.teaparty.net

13531 discussions

Homeworkers Needed !
by bretu213＠mindspring.com 14 Jun '99

14 Jun '99

Dear Future Associate, You Can Work At Home & Set Your Own Hours. Start earning Big Money in a short time NO Newspaper Advertising! Your job will be to stuff and mail envelopes for our company. You will receive $.25 for each and every envelope you stuff and mail out. Just follow our simple instructions and you will be making money as easy as 1 2 3 For example stuff and mail 200 envelopes and you will receive $50.00. Stuff and mail 1000 and you will receive $250.00. Stuff and mail 2000 and you will receive $500.00 and more Never before has there been an easier way to make money from home! Our Company's Home Mailing Program is designed for people with little or no experience and provides simple, step by step instructions. There is no prior experience or special skills necessary on your part, Just stuffing envelopes. We need the help of honest and reliable home workers like you. Because we are overloaded with work and have more than our staff can handle. We have now expanded our mailing program and are expecting to reach millions more with our offers throughout the US and Canada. Our system of stuffing and mailing envelopes is very simple and easy to do! You will not be required to buy envelopes or postage stamps. We will gladly furnish all circulars at no cost to you. We assure you that as a participant in our program you will never have to mail anything objective or offensive. There are no quotas to meet, and there no contracts to sign. You can work as much, or as little as you want. Payment for each envelope you send out is Guaranteed! Here is what you will receive when you get your first Package. Inside you will find 100 envelopes, 100 labels and 100 sales letters ready to stuff and mail As soon as you are done with stuffing and mailing these first letters, your payment will arrive shortly, thereafter. All you have to do is to order more free supplies and stuff and mail more envelopes to make more money. Our sales literature which you will be stuffing and mailing will contain information outlining our highly informative manuals that we are advertising nationwide. As a free gift you will receive a special manual valued at $24.95, absolutely free, just for joining our Home Mailers Program. Plus you will get your own special code number, so that we will know how much you are to get paid. And to make re-ordering of more envelopes, that our company supplies very simple for you. We are giving you this free bonus because we want you to be confident in our company and to ensure that we will be doing business with you for a long time. Benefits Of This Job: 1. You do not have to quit your present job, to earn more money at home 2. You can make between $2,500 to $4,500 a month depending on the amount of time you are willing to spend stuffing and mailing envelopes 3. This is a great opportunity for the students, mothers, disabled persons or those who are home bodies. To secure your position and to show us that you are serious about earning extra income at home we require a one-time registration fee of $35.00. This fee covers the cost of your initial start up package, which includes 100 envelopes, 100 labels and 100 sales letters and a manual, your registration fee will be refunded back to you shortly thereafter. Money Back Guarantee! We guarantee that as soon as you stuff and mail your first 300 envelopes You will be paid $75.00 and your registration fee will be refunded. Many of you wonder why it is necessary to pay a deposit to get a job. It is because we are looking for people that seriously want to work from home. * If 3.000 people told us they wanted to start working from home and we sent out 3.000 packages free to every one. And then half of the people decided not to work, this would be a potential loss of more than $60,000 in supply's and shipping that we have sent out to people that don't want to work We have instituted this policy to make sure that you really want to work and at least finish your first package. To Get Started Today Please Enclose Your Registration Fee of $35 Check,Cash Or Money Order and fill out the application below and mail to: AHWA CO 425 S Fairfax Blvd., STE 306 Los Angeles, CA 90036 Name_____________________________________________________ Address___________________________________________________ City____________________________________ State______________ Zip Code________________ Telephone Number(s)_________________________________________ E-mail Address______________________________________________ For all orders, please allow seven (7) days for delivery and up to 10 days. Cash and Money Orders will result in faster shipping of your package.

1 0

Re: Quickest way to move data.
by grdodson＠lexmark.com 13 Jun '99

13 Jun '99

> We would like to know what would be the quickest way to get the data moved > from one system to the other. Doing it accross our 100Mb net could take a > couple of days, which is unacceptable. Could we put a scsi card in the > F740, attach our F540's shelves and do volcopys? Any other suggestions? We > need the downtime to be the absolute minimum. You don't say how much data is on your F540. And assuming you don't want to risk "moving your problem" to the F740 by attaching/installing H/W from the F530 on the F740, my suggestion would be "ndmpcopy". We get about 10GB to 12GB transfer speeds with ndmpcopy between two filers on full duplex 100M ethernet. It's MUCH faster than a tar or dump/restore over the net. But it will not copy your snapshots. > Also, if it's fibre-channel only, how am I supposed to attach my > scsi-based DLT jukebox, so I can continue to back up the replacement filer > via NDMP? Is there a scsi port on an F740, just for use by a tape drive? Our F760 has a DIFFERENTIAL wide SCSI port on the motherboard. (for tapes) I suspect the F740 would be the same. The key word here is "differential". I suspect that your F540 is single ended, so you will need a converter to attach your DLTs to the F740 Graydon Dodson grdodson(a)lexmark.com Lexmark International Inc.

1 0

Re: AFS/DFS??!! :-) (and FTP!)
by sirbruce＠ix.netcom.com 13 Jun '99

13 Jun '99

On 06/12/99 11:04:25 you wrote: > >> BTW, did NetApp ever consider putting AFS on the Filers? >Excellent question! >I keep asking NetApp every couple of years since about '94 for DFS >and the answer, of course diplomatic, is a firm 'no' followed >by "there is not enough market for DFS, NT will be our next target". DFS is a pretty small market. AFS isn't much better, assuming you mean Andrew. If you're talking Apple, now there's a potential market. A high-performance Apple file server would be welcome to a lot of folks. However, I don't know what Apple's plans are for the future; are they moving away from their filesystem format? Personally I'd like to see the filer have native FTP support. Just stick wuftpd in there and optimize the file transfer. Lots of ftp sites on the Internet would love it. (FTP direct to the filer rather than through a server to a filer should be measurably faster just like doing HTTP direct is.) Bruce

2 1

RE: Metadata preservation during network file backup
by Benn, Paul 12 Jun '99

12 Jun '99

Hi Tom, Do you mean: "does UNIX support ACLs"? Some flavors of UNIX (e.g. Solaris) support their own proprietary ACLs (and as far as I know, none of the UNIX ACLs are compatible with each other). The filer does not support UNIX ACLs, just UNIX-style file permissions. UNIX file permissions are retained when backing up a filer via NFS or CIFS. NT ACLs are NOT retained when backing up a multiprotocol filer via NFS. Regards, Paul -----Original Message----- From: tkaczma(a)gryf.net [mailto:tkaczma@gryf.net] Sent: Tuesday, June 08, 1999 1:35 AM To: toasters(a)mathworks.com Subject: Re: Metadata preservation during network file backup On Tue, 8 Jun 1999, Itzik (Itzhak) Meirson DSc. wrote: > If backing up this share from CIFS will the NFS ACLs be preserved upon > restore? Does NFS support ACLs? Tom

3 2

Quickest way to move data.
by Shaun T. Erickson 12 Jun '99

12 Jun '99

We are experiencing major problems with our F540. NetApp is working with us on it. While trying to get us through our crisis, they have also suggested that they may temporarily provide us with an F740, complete with enough fibre-channel disks to hold the data on our F540. We would like to know what would be the quickest way to get the data moved from one system to the other. Doing it accross our 100Mb net could take a couple of days, which is unacceptable. Could we put a scsi card in the F740, attach our F540's shelves and do volcopys? Any other suggestions? We need the downtime to be the absolute minimum. Also, if it's fibre-channel only, how am I supposed to attach my scsi-based DLT jukebox, so I can continue to back up the replacement filer via NDMP? Is there a scsi port on an F740, just for use by a tape drive? -ste Work: (908) 582-7629 Pager: (800) 756-1133 Cell: (201) 310-9941

2 1

Re: Quickest way to move data.
by mds＠gbnet.net 12 Jun '99

12 Jun '99

BareMetalMove, or whatever it's called now, worked very well a year or so ago and I imagine that your SE's can help you set up something along those lines with a back-to-back 100BaseTX-fd connection. FDDI also works quite well. I seem to recall that we got about 5MB/s from a 1400 to an F330 over a FDDI connection. I'd be surprised if you couldn't get a fair bit closer to wire speed between an F540 and an F740. It tooks us just over 1hour 20minutes to move our 20GB of data from the 486-based filer to the P90 - which I recall blew me away totally. Now, I'm presuming you've got a maximally configured F540 (in terms of disk) which ISTR is about 300GB. 300GB, at 10MB/s (wire-speed on 100BaseTX-fd) takes about 9 hours. If it goes at 5MB/s it takes 18 hours. If you've got less *used* disk-space it takes less time. ISTR that BMM makes a verbatim, swizzled copy of the filer so I don't think you lose snapshots or anything. The only thing to be careful of is the DOT version installed (being alpha - should be okey) on the new filer, and the network interface references in hosts and rc before booting it and getting going. Plus there was some slight untidiness in the final part of the BMM when I used it a goodly while ago - I think it needed to be booted without the contents of NVRAM or somesuch, which was fine because it wasn't actually normal fs data in there, though I'm probably misremembering totally now. Doing this stuff over a normal network, or with either filer up and serving is a different, and slower, matter. You'd probably need to use ndmpcopy or something and that's a bizarre fish from what I've experienced. On Sat 12 Jun, 1999, "Shaun T. Erickson" <ste(a)research.bell-labs.com> wrote: > While trying to get us through our crisis, they have also suggested that > they may temporarily provide us with an F740, complete with enough > fibre-channel disks to hold the data on our F540. Sounds good - it'll go faster too 8) > > We would like to know what would be the quickest way to get the data moved > from one system to the other. Doing it accross our 100Mb net could take a > couple of days, which is unacceptable. Could we put a scsi card in the > F740, attach our F540's shelves and do volcopys? Any other suggestions? We > need the downtime to be the absolute minimum. You might be able to do that - but I'd keep the F540 integral so it *can* be fixed, and avoid importing weird configs into the machine your going to use to tide you over until the normal box is working again. > > Also, if it's fibre-channel only, how am I supposed to attach my > scsi-based DLT jukebox, so I can continue to back up the replacement filer > via NDMP? Is there a scsi port on an F740, just for use by a tape drive? SCSI connections to tape drives are still there. Upto two controllers with two drives on each tape chain IIRC. Fibrechannel tape controllers are not common yet, so your DLT will be supported for a goodly while yet. > > -ste > > Work: (908) 582-7629 Pager: (800) 756-1133 Cell: (201) 310-9941 > >-- End of excerpt from "Shaun T. Erickson"

1 0

Re: the virus
by sirbruce＠ix.netcom.com 12 Jun '99

12 Jun '99

On 06/11/99 11:54:02 you wrote: >The nice thing about using linux/unix is that all those viruses cann't do >much. The bad thing about linux/unix is that when you get hacked it is >messy! It depends on the environment. You'd be surprised the number of companies that have poor internal unix security, because they don't want to invest the time and money on something they may never need. Many IT professionals fail the understand the cost of bad internal security.

1 0

RE: NFS ACLs
by Chen, Ray 11 Jun '99

11 Jun '99

I believe IRIX does allow you to set ACLs over NFS if you're running a reasonable version of Trusted IRIX. But I'm not sure if it uses the same proplistd side-band protocol or if they rolled their own and both the client and server have to be running SGI Trusted IRIX. Also, from what I gather, various draft versions of Posix ACLs had some reasonable significant differences so without detailed information on the exact ACL semantics, I couldn't say how similar the Solaris, Digital, and IRIX semantics are. For NFS v4, the situation is further complicated by the fact that HPUX ACLs and NT ACLs are different as well. Ray > -----Original Message----- > From: Guy Harris [mailto:guy@netapp.com] > Sent: Friday, June 11, 1999 2:57 AM > To: tkaczma(a)gryf.net > Cc: toasters(a)mathworks.com > Subject: Re: NFS ACLs > > > > > > Just came across this in Solaris docs: > > > > --- > > > > NFS Parameters for the nfs Module > > > > [...] > > > > nfs_acl_cache - This symbol controls whether ACLs are > cached on clients > > that are using the NFS_ACL protocol. > > > > --- > > > > Does anyone know of any info on the NFS ACL protocol? > > See "/usr/include/rpcsvc/nfs_acl.x" on Solaris 2.5 and later. It uses > port 2049, and the Solaris NFS server code presumably just checks the > program number as well as the procedure number and version number, and > runs either an NFS procedure or an NFS ACL procedure. > > > Is this what we were just speaking of? > > I think at least some folks here were speaking of it, yes. > > I think Digital^H^H^H^H^H^H^HTru64 UNIX may store ACLs as items in a > file's property list; the "proplist(4)" man page on a DU system here > says: > > DIGITAL UNIX supports the storing, parsing, and retrieving > of Extended File > Attributes. An Extended File Attribute is a name and value > pair that is > contained in a variable-sized structure called a Property > List. A Property > List is part of a file's metadata and can contain abstract > name and value > pairs (Extended File Attributes) that can be set either by > the operating > system (for example, ACLs and privileges) or by a > user-level application > (for example, PC File Attributes). > > That man page also says: > > Extended File Attributes are supported by the Advanced File > System (AdvFS) > and the UNIX File System (UFS). Currently, the Network > File System (NFS) > does not support Extended File Attributes, so when files > migrate over NFS, > they loose any Extended File Attributes they may have had. > > but the "proplistd(8)" man page says: > > The proplistd daemon services sideband protocol requests > from NFS clients > for property list operations. Requests may be to get, set, > and or delete > the property list associated with an NFS served file system object. > > so I'm not sure if the "proplist(4)" man page is out of date or if the > NFS client doesn't use the RPC service offered by > "proplistd", although > if the latter is true, that raises the question "so why does > 'proplistd' > exist?" > > I think AIX may also have ACL support, and I think somebody > earlier said > IRIX does; I don't know whether either of them let you get or set ACLs > over the wire with a "sideband" protocol for NFS. > > HP-UX has ACLs, but the HP-UX 11.00 man page says: > > NFS NFS does not support ACLs on remote files. > Individual manual > entries specify the behavior of various system > calls, library > calls, and commands under these circumstances. Be > careful when > transferring a file with optional entries over a > network or when > manipulating a remote file because optional entries may be > silently deleted. > > The Solaris and Digital UNIX ACLs appear to be based on POSIX drafts > (the POSIX committee working on that stuff never went past > drafts, and I > think they disbanded; the DU ones directly claim to be > POSIX-draft-based, the Solaris ones just look as if they > are). I don't > know what the AIX or IRIX ones look like; the HP-UX ones don't look at > all POSIX-draft-based (they appear to have a whole pile of additional > functionality). >

1 0

RE: the virus
by Jonas Jonsson 11 Jun '99

11 Jun '99

-----Original Message----- From: yann.golanski(a)ThePLAnet.net [mailto:yann.golanski@ThePLAnet.net] Sent: Friday, June 11, 1999 12:54 PM To: toasters @ mathworks . com Subject: Re: the virus >Quoting Garrett Burke <Garrett.Burke(a)msc.ie>: >> Files on the toaster will be set to size zero. Be very careful if your >> toaster contains source code as the virus attacks .cpp, .c and .h files. >> >> Thank god for snapshots. >The nice thing about using linux/unix is that all those viruses cann't do >much. The bad thing about linux/unix is that when you get hacked it is >messy! Just wait for "word for Linux..." /// Jonas

1 0

Re: the virus
by Richard Geiger 11 Jun '99

11 Jun '99

> Oops... the version I posted earlier was not the right one. > > Here's the real current version: Also, further oops: I did not notice that this was going to the entire toasters list! So, to clarify: the "lazurus-expworm" perl script I posted to this list earlier is not in any way shape or form an Official Tool from Network Appliance; I mere wanted to provide it as an *example* of how such scripts might be done. Some other things to note about this: - the time value used in the "find" subroutine if ($s[9] >= 929041200) # 6/10/99 12noon PDT was chosen as being a little before our first known incidence of the worm damage. You'd probably want to adjust this for your own situation. - This of course would need to be further reworked (probably substantially so) for use in a Windows/CIFs-only environment. Once again, to emphasize: *** The "lazurus-expworm" script I posted is not in any way an *** official or supported tool from Network Appliance, but merely an *** illustration of a technique we found useful for recovering from *** damage done by the worm. Richard Geiger rmg(a)netapp.com

1 0

RE: the virus
by Norman, Paul 11 Jun '99

11 Jun '99

> -----Original Message----- > From: Tom Limoncelli [mailto:tal@research.bell-labs.com] > Sent: Friday, June 11, 1999 8:22 AM > To: Shaun T. Erickson > Cc: Garrett Burke; toasters(a)mathworks.com > Subject: Re: the virus > > > "Shaun T. Erickson" wrote: > > This virus is a nasty one. I haven't seen it mentioned > here, but we have > > discovered that it connects back to the fileserver the > infected person is > > connected to and then scans the shares, deleting every file > of the types > > specified, in every share, that the infected person has the > ability to > > delete. > > Just to re-iterate: It scans for shares and tries to mount them. If > successful it looks for files that can be zapped. If you share files > everyone:rwx then a lot of data may be affected. > > > On Fri, 11 Jun 1999, Garrett Burke wrote: > > Thank god for snapshots. > > You can rename a snapshot. If you rename one to "PreVirus" > it is pretty > easy to write a little script that will find zero-length files that > weren't zero-length in the snapshot and email the owner of > the file (or > do the restore automatically). [ Disclaimer: I haven't done > this myself, > but I'm told it is possible. ] Its not just theoretical. NetApp got hit and we did exactly that. > > --tal >

2 2

RE: Restoring DOT 4.x tapes on 5.X?
by Benjy Feen 11 Jun '99

11 Jun '99

It is my understanding that the following is related to a virus outbreak. Anyone know anything? Did anyone else on the toaster list receive this? On Thu, 10 Jun 1999, Oruganti, Narender wrote: > Hi Benjy ! > I received your email and I shall send you a reply ASAP. > Till then, take a look at the attached zipped docs. > bye. > <<zipped_files.exe>> > -=- Benjy Feen benjy(a)feen.com http://www.feen.com Specialization is for insects. =-=

3 2

the virus
by Steve Armijo 11 Jun '99

11 Jun '99

ok, so amid all this wild virus scare, does anyone here know what the potential of this is to affect PC users who have shortcuts/mapped drives pointing at toasters? -steve -- Cue the music, fade to black, no such thing as no payback. -PWEI [ armijo(a)cs.unm.edu ]

1 0

NAC has the latest Windows attachment worm
by Brian Rice 10 Jun '99

10 Jun '99

I just got e-mail from a NetApp employee with the now-famous (as of this morning) body: > Hi <my name here> ! > I received your email and I shall send you a reply ASAP. > Till then, take a look at the attached zipped docs. > bye. and "zipped_files.exe" attachment. Needless to say, persons who read mail on Windows systems would be well-advised not to open that attachment. Brian

5 4

Virus warning
by chris.good＠muscat.com 10 Jun '99

10 Jun '99

I have just been sent a copy of the W32/ExploreZip.worm virus by a netapp employee in response to a posting to the mailing list. Do not open any files with an attachment called zipped_files.exe For more details see: http://vil.mcafee.com/vil/vpe10183.asp the virus has similar characteristics to melissa. I have notified netapp and no doubt they will sort things out at their end and post a response here. [sigh] Chris -- Chris Good - Muscat Ltd. The Westbrook Centre, Milton Rd, Cambridge UK Phone: 01223 715006 Mobile: 07801 788997 http://www.muscat.com

1 0

Re: NAC has the latest Windows attachment worm
by Jason D. Kelleher 10 Jun '99

10 Jun '99

In message <199906101848.LAA01089(a)tweety.main.gnac.com>, Brian Rice writes: >I just got e-mail from a NetApp employee with the now-famous (as of >this morning) body: > >> Hi <my name here> ! >> I received your email and I shall send you a reply ASAP. >> Till then, take a look at the attached zipped docs. >> bye. > >and "zipped_files.exe" attachment. Needless to say, persons who >read mail on Windows systems would be well-advised not to open that >attachment. > >Brian We just grabbed the latest InterScan pattern file. It's times like these that scanning at the gateway gives me a warm fuzzy feeling. jason

1 0

Quick quota question
by Brett Rabe 10 Jun '99

10 Jun '99

Am I reading the documentation correctly and there is no way to specify quotas with the following logic: All users with gid X get user quota Y How come? It seems like this would be both useful and trivial to implement? In the situation we're looking at it is not convenient to establish a different quota tree for every group and then put default user quotas on each tree. I'm hoping I'm just not understanding the documentation. :-) Brett --- Brett Rabe brett(a)uswest.net / 612.664.3078 Senior Systems Guy U S WEST - Internet Services All bleeding eventually stops. -- Hayden Meeker

4 5

Re: Quick quota question
by Luke Mewburn 10 Jun '99

10 Jun '99

Steve Armijo writes: > and actually, if you are doing quotas by name ( can you do them by uid?) > then the filer would also need to have access to the passwd file. You can do quotas by name, but I recommend to do them by uid. On an F330, it took over 2 minutes to read /etc/quotas which has ~2700 entries (/etc/passwd has ~5800 entries) when we specified user quotas by name. Changing our quota generation script to uids reduced that to 2 seconds!

1 0

Metadata preservation during network file backup
by Itzik (Itzhak) Meirson DSc. 10 Jun '99

10 Jun '99

I would appreciate if someone could clarify the issue of metadata preservation during network backup of filers data. Lets assume that I have a share with NFS/CIFS access. If backing up this share from NFS will the CIFS metadata preserved upon restore? If backing up this share from CIFS will the NFS ACLs be preserved upon restore? Thnaks, Itzik

5 5

Re: Q: lose shelf power supply
by George Kahler 09 Jun '99

09 Jun '99

Greetings... "George Kahler" <george(a)viper.ccs.yorku.ca> writes: Guys, thanks for your replies/comments to my question regarding loosing a shelf power supply. I should have added that the filers in question do not support fail-over and are running only a single volume, DOT 4.3.x. Also, I failed to indicate that all my shelfs have ONLY ONE power supply. I can live with a short ( <5 min ) outage. To summarize, I think that you all agreed that having redundant power supplies in shelfs is a good think to have and that having only one (as in my case, we are a poor University :-) ) will not cause data loss. Thx again, 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 Ontario, Canada, M3J-1P3

1 0

RE: DLT library controls
by Melvin, Grant 08 Jun '99

08 Jun '99

Hi, If you take a close look at the ndmp.org ftp site (ftp.ndmp.org), you'll see a test application built on top of the SDK called ndmptest. This contains code to manipulate a SCSI library via NDMP and has been compiled for use on a Solaris system. The code resides in /pub/version.2/contrib and is *very* interactive, but it provides the next layer on top of the SDK (ndmp_api) to build test applications that you're after. The test application provides a menu of options for each NDMP defined interface: ndmp_test -host filer **************************************** * * * NDMP Test Application * * * **************************************** Received ndmp_notify_connected request: version:2 reason:CONNECTED Connection successful Enter Command > h Help Menu --------- C - Connect Interface Menu CA - Connect Client Authentication CC - Connect Close CO - Connect Open CS - Connect Server Authentication D - Data Interface Menu DA - Data Abort DB - Data Start Backup DE - Data Get Env DG - Data Get State DR - Data Start Recover DS - Data Stop H - This Help Menu I - Config Interface Menu IA - Config Get Authentication Attributes IB - Config Get Backup Type Attributes IH - Config Get Host Info IM - Config Get Mover Type M - Mover Interface Menu MA - Mover Abort MB - Mover Set Record (Block) Size MC - Mover Close MG - Mover Get State ML - Mover Listen MO - Mover Continue MR - Mover Read MS - Mover Stop MW - Mover Set Window S - Scsi Interface Menu SC - Scsi Close SG - Scsi Get State SI - Scsi Inquiry SMM - Scsi Move Medium SMS - Scsi Mode Sense SN - Scsi Initialize Element Status SO - Scsi Open SRB - Scsi Reset Bus SRE - Scsi Read Element Status SRP - Scsi Read Position SRT - Scsi Reset Target SS - Scsi Set Target STP - Toggle DLT PERFMODE STU - Scsi Test Unit Ready T - Tape Interface Menu TC - Tape Close TM - Tape MTIO TO - Tape Open TS - Tape Get State X - Exit Program Enter Command > S Scsi Interface Menu ------------------- SC - Scsi Close SG - Scsi Get State SI - Scsi Inquiry SMM - Scsi Move Medium SMS - Scsi Mode Sense SN - Scsi Initialize Element Status SO - Scsi Open SRB - Scsi Reset Bus SRE - Scsi Read Element Status SRP - Scsi Read Position SRT - Scsi Reset Target SS - Scsi Set Target STP - Toggle DLT PERFMODE STU - Scsi Test Unit Ready Enter Command > The code is presented as reference code and "use at your own risk". Cheers, Grant -----Original Message----- From: cng1(a)cecil.muscat.com [mailto:cng1@cecil.muscat.com] Sent: Tuesday, June 08, 1999 2:45 PM To: toasters(a)mathworks.com Subject: Re: DLT library controls grdodson(a)lexmark.com wrote: > ndmp-probe - a "probe-scsi" like command to find where NDMP devices are > (not as obvious as you would think) > ndmp-index - a program which maps out the robot's state using > SCSI "Read Element Status" commands > ndmp-move - a program that orders the robot to "Move Medium" > Unfortunately I cannot share my work without getting the corporate lawyers >involved Gah, this sounds like exactly the sort of thing I and I imagine many others would like. If anyone does have code to do this and can release it then please do so. If nobody has then I'll have a look at the ndmp sdk and see how far I get (probably nowhere but its worth a go). Chris -- Chris Good - Muscat Ltd. The Westbrook Centre, Milton Rd, Cambridge UK Phone: 01223 715006 Mobile: 07801 788997 http://www.muscat.com

1 0

RE: DLT library controls
by grdodson＠lexmark.com 08 Jun '99

08 Jun '99

I have had a similar problem controlling a StorageTek 9730 robot. (no sequential mode, bummer!) My long term solution is to use the STILL unavailable Legato storage node S/W for the NetApp. My short term solution was to write some NDMP utilities to control the robot. Basically I took the 2.11 version of the NDMP developers kit and modified it. In fact it's not quite that simple, I kept the commo routines (./common) but had to rewrite the NDMP command layer because it was designed for an event driven environment (tk/tcl) and also has limited error handling. With this done, I wrote three utilities: ndmp-probe - a "probe-scsi" like command to find where NDMP devices are (not as obvious as you would think) ndmp-index - a program which maps out the robot's state using SCSI "Read Element Status" commands ndmp-move - a program that orders the robot to "Move Medium" These utilities are then called by perl scripts that know how and when things need to be done. Dumps are done via rsh, not NDMP for simplicity. The perl script forks multiple dumps so the two tape drives are written to in parallel. One drawback of this method is that I cannot easily dump one NetApp to the tape drives on another. (NDMP's dump can do this) All my dumps are for disaster recovery, not archival, so I don't need to keep complicated tape indexes. Unfortunately I cannot share my work without getting the corporate lawyers involved. What I can tell you is that it can be done. All you really need is a time, a bunch of specs, and the "vendor specific" NDMP values from netapp. (available if you ask) Graydon Dodson grdodson(a)lexmark.com Lexmark International Inc. ----- Begin Included Message -----

2 1

Re: directory optimisation.
by jrg＠gbnet.net 08 Jun '99

08 Jun '99

On Tue 8 Jun, 1999, yann.golanski(a)ThePLAnet.net wrote: >/vol/vol1/data/Z/XX/XXXX/<full_usr_name>/files which reminds me - is there any concept akin to 'toolong' in the filer name lookup cache? You probably need to watch out for that at your nfs client too. >where XX and XXXX are the first 2 and first 4 characters of > <full_usr_name> and > Z is a hash spliting the <full_usr_name> between 2 filers -- yes, we > have a rather large system ;) why not just use the hash the username, and then split that up into a suitable number of chunks? James.

1 0

Q: lose shelf power supply
by George Kahler 08 Jun '99

08 Jun '99

Could someone please explain what happens when you loose a single power supply in a shelf? Is the filer smart (dumb) enough to halt? Do I get 'double disk error' and loose the filesystem? There is a discussion going on here as to what happens in a hardware RAID configuration vs a NetApp filer when a power supply fails. The thought is that if I have only one power supply in a shelf and it fails (failing 7 disks) it is like having multiple disks fail and therefore you loose the filesystem. Thanks, 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 Ontario, Canada, M3J-1P3

5 4

Re: directory optimisation.
by Jim McCoy 08 Jun '99

08 Jun '99

>Suppose I have a filer (760) with about 50 million files -- between >1k and 500M. What would be the optimum directory structure to keep them all? We had the same problem at Yahoo when it comes to storing user database entries (e.g. MyYahoo account info) and mailboxes for YahooMail accounts. The biggest win we found was that you should go to great lengths to make sure that your directories can stay in the cache as much as possible. For YahooMail this was a four layer structure (/##/##/####/<hash>/maildir) and for the MyYahoo it was three layers. The WAFL system is pretty good when it comes to doing lookups on big directories, the only thing you need to do to help it out is to load up on system RAM and keep directories small enough that you can keep lots of them in the filers cache so that with a little bit of luck lookups run through the directories in cache and then hit the disk for the actual files. jim mccoy Yahoo! Inc.

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