toasters

toasters@lists.teaparty.net

13531 discussions

Re: Forcing full duplex on 720
by Paul Albers 13 Jan '99

13 Jan '99

Hello Jamie, Do you have the switch set to autonegotiate? If you do, what will happen is the switch will attempt to autonegotiate with the filer, but since e0 was set manually, we don't communicate back with the switch. By default then the switch will negotiate down to half duplex, no matter what the other side is set to. The best way I think to fix this, is to manually set both switch and filer's interfaces to full duplex. Turning both autosensing and autonegotiation off on the corresponding switch ports. As Andrew mentioned below, setting auto works very well with most switches. There are switches out there that will not autonegotiate well. In these cases, if the filer does not get response back, we drop down to half duplex. This is the protocol I mentioned above. Then it is likely that the switch could be running at full duplex. Either method is fine, just do some testing with the auto setting working with your current switch to make sure all works fine. Regards, -paul Escalations Engineer Network Appliance 408.367.3103 At 08:18 AM 1/13/99 +0000, you wrote: >You haven't given enough info for me to tell what type of card e0 is, but >the quad card will certainly support autosensing. Try setting the >mediatype to "auto" - it should then negotiate with the switch correctly. >If e0 supports autonegotiation it will have only one socket - in which case >you can do the same there. > >If it doesn't work, open a call with NetApp support and supply more >information - at least the switch details, sysconfig -v > >Andrew Bond, >UK Systems Engineer > >At 06:46 pm 1/12/99 -0800, Jamie Lawrence wrote: >>At 06:36 PM 1/12/99 -0800, Jerry Talkington wrote: >>>Jamie Lawrence wrote: >>>> >>>> Hi - >>>> >>>> I have a 720 with a quad card which never negotiates to full >>>> duplex with out 10/100 switch. The Solaris boxes on the same >>>> switch negotiate to FD just fine, and the 720 thinks FD is >>>> enabled, but the switch only ever negotiates to HD. >>>> >>>> Has anyone seen this before? >>> >>>Not sure I understand the question, those all seem to be at full >>>duplex. As a general rule, setting the filer up to autonegotiate is a >>>bad idea, since it will try to come up faster than the interface can >>>negotiate a connection. I always recommend that people manually set >>>both the filer (through the rc file) and the switch to whatever duplex >>>they want to use. >> >>Sorry - That message was poorly written. The filer thinks the interface >>is full duplex. The switch thinks it is half duplex. I am setting the >>interface manually in rc: >> >>ifconfig e0 mediatype 100tx-fd `hostname`-e0 netmask 255.255.255.128 >>ifconfig e2a mediatype 100tx-fd `hostname`-e2a >>ifconfig e2b mediatype 100tx-fd `hostname`-e2b >>ifconfig e2c mediatype 100tx-fd `hostname`-e2c >>ifconfig e2d mediatype 100tx-fd `hostname`-e2d >> >>This is an unmanaged switch, so forcing the speed is not an option, >>but I don't think the switch is faulty because the Sun machines behave >>as expected - both the machines and the switch declare a full duplex >>connection. >> >>-j >> > >

1 0

Fore ATM Drivers on F760
by Des Barden 13 Jan '99

13 Jan '99

We are looking to connect a F760 into a Fore Systems ATM LAN. The LAN has has four ATM switches that are meshed for resilience. The ATM services are configured to use LANE and Distributed LAN Emulation. To make use of the features available in the Fore Systems version five code we would like to connect the F760 via four OC3 ATM cards. What I would like to know if by having multiple cards that we could make use of UNI failover, failover groups and load balancing. Regards, Des Barden Team Leader Unix/LAN IS Regional Services (Harlow) Esn: 742-5856

1 0

Re: Gigabit ethernet card placement
by Scott Miller 12 Jan '99

12 Jan '99

On Jan 11, 10:19pm, Roland Lee wrote: > Subject: Re: Gigabit ethernet card placement > Daniel, on the F630, the gigabit card should be in slot 10, What if you have a cluster? The ServerNet card must also go into slot 10. -Skottie > and on the > F520 and the F540, the card must be in slot 3. Those are the only slots > supported and tested, though they may work in other slots. > Performance-wise, you would want the fastest network card to be in slot 10 > on the F630 anyways. > > Roland Lee > NetApp TSE > > On Mon, 11 Jan 1999, Daniel Quinlan wrote: > > > Performance-wise, does it matter what PCI slot you put a Gigabit > > ethernet card? (This goes back to the question about which slots are > > 64bit vs. 32bit.) Specifically, I'm wondering about the F630 and > > maybe the F540. > > > > I wasn't able to find much information about installing Gigabit > > ethernet cards on now.netapp.com. > > > > - Dan > > > >-- End of excerpt from Roland Lee -- ----- Scott Miller | Walt Disney Feature Animation skottie(a)fa.disney.com | Digital Studio Technology

1 0

NDMP fix works!
by Menachem Kaiser (r53417) 12 Jan '99

12 Jan '99

Date: Tue, 12 Jan 1999 16:17:35 +0200 From: "Menachem Kaiser (r53417)" <r53417(a)email.sps.mot.com> Reply-To: "Menachem Kaiser (r53417)" <r53417(a)email.sps.mot.com> Organization: Motorola Semiconductor Israel To: andrew.lockhart(a)veritas.com, "robert.slater(a)veritas.comkahana"@clalsys.clalcomp.co.il, support(a)veritas.com, brian.blake(a)veritas.com, mike.proveder(a)veritas.com, brad.haggert.veritas.com(a)email.sps.mot.com CC: support(a)netapp.com, bobbl(a)netapp.com, pas(a)veritas.com, Nachshon Yeshurun <nachshon(a)msil.sps.mot.com>, Haim Leib <haiml(a)msil.sps.mot.com>, pas(a)veritas.com, "dan(a)netapp.com Eyal Traitel" <eyalt(a)msil.sps.mot.com>, karl(a)corp.webtv.net, toaster(a)mathworks.com, "Nick Janakas (qa4418)" <qa4418(a)email.sps.mot.com>, "Hans-Peter Kiekuth (hkiekut1)" <hkiekut1(a)email.sps.mot.com> Subject: NDMP fix works! Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, We wish to thank all those who responded to my request. We changed the following two variables in the bp.conf file in veritas netbackup software as follows: CLIENT_CONNECT_TIMEOUT = 3600 CLIENT_READ_TIMEOUT = 3600 From our initial test the backup does not crash any more. We will continue checking the backup and restore using NDMP. Regards Manny Date: Mon, 11 Jan 1999 14:51:21 +0200 From: "Menachem Kaiser (r53417)" <r53417(a)email.sps.mot.com> Reply-To: "Menachem Kaiser (r53417)" <r53417(a)email.sps.mot.com> Organization: Motorola Semiconductor Israel X-Mailer: Mozilla 4.06 [en] (WinNT; I) MIME-Version: 1.0 To: andrew.lockhart(a)veritas.com, robert.slater(a)veritas.com, kahana(a)clalsys.clalcomp.co.il, support(a)veritas.com, brian.blake(a)veritas.com, mike.proveder(a)veritas.com, brad.haggert.veritas.com(a)email.sps.mot.com CC: support(a)netapp.com, bobbl(a)netapp.com, pas(a)veritas.com, Nachshon Yeshurun <nachshon(a)msil.sps.mot.com>, Haim Leib <haiml(a)msil.sps.mot.com>, pas(a)veritas.com, dan(a)netapp.com, Eyal Traitel <eyalt(a)msil.sps.mot.com> Subject: NDMP doesn't work ! Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Dear Sir's, We have an open call at Veritas call #110239. On Jan 7th the call was upgraded to severity 1. According to the SLA agreement with Veritas this call should be upgraded to SVP WW operations. We have not yet received any solution. The local Veritas representitive has explained to us that he has been in contact with Veritas UK. and hopes to be in contact with Veritas U.S. this evening. One of the key reasons we chose the combination of Veritas and Network Appliance were the assurances we got that we would be able to restore data to the filer in the event of a disaster within a relatively short period of time. (at a rate of 50 GB/Hr). We have two F760 Netapp filers that we are unable to utilize since we cannot effectively restore the data in the event of a disaster. Although we can backup and restore the data across the net, the severe impact of hours of downtime related to a double disk failure while restoring a filer over the net, are not feasible. Thus we have ceased migrating data to these filers. This is a major setback in the normal operations of the company and has already made a serious impact on our business. We expect to hear from you by the end of the business day with a plan and timetable for solving this problem. Thank-You Regards

1 0

Gigabit ethernet card placement
by Daniel Quinlan 12 Jan '99

12 Jan '99

Performance-wise, does it matter what PCI slot you put a Gigabit ethernet card? (This goes back to the question about which slots are 64bit vs. 32bit.) Specifically, I'm wondering about the F630 and maybe the F540. I wasn't able to find much information about installing Gigabit ethernet cards on now.netapp.com. - Dan

4 3

Veritas Netabackup and NDMP
by Menachem Kaiser (r53417) 10 Jan '99

10 Jan '99

Hi, We have recently purchased two F760 filers as well as Veritas netbackup and StorageTek Tape library's. Thus far we have been unsuccessful in backing up our filers using NDMP. Has anyone out there have a similar configuration or encountered similar problems? Any help would be greatly appreciated. Thanks Manny

2 1

RE: Veritas Netbackup and NDMP
by Melvin, Grant 10 Jan '99

10 Jan '99

Hi Menachem, Have you contacted support at NetApp and/or Veritas? I know of at least two accounts who are running F630s with NetBackup & STK libraries. F760s should also work. I also know of an eval with F760s going on in Europe which had some tape drive unload issues. The library is controlled via ASCLS and the backup via NDMP and I think the problem has been identified as a synchronization issue, i.e. unload the tape via NDMP before using ASCLS to exchange/move the tape. Cheers, Grant -----Original Message----- From: Menachem Kaiser (r53417) [mailto:r53417@email.sps.mot.com] Sent: Sunday, January 10, 1999 5:50 AM To: toasters(a)mathworks.com; Eyal Traitel; Nachshon Yeshurun Subject: Veritas Netabackup and NDMP Hi, We have recently purchased two F760 filers as well as Veritas netbackup and StorageTek Tape library's. Thus far we have been unsuccessful in backing up our filers using NDMP. Has anyone out there have a similar configuration or encountered similar problems? Any help would be greatly appreciated. Thanks Manny

1 0

unsubscribe
by DESCHAMPS, Stephane 10 Jan '99

10 Jan '99

unsubscribe

1 0

Quota problem on F630
by Cheryl Cruddace 10 Jan '99

10 Jan '99

We have a brand new F630 filer that we brought up in the last month. Actually, it was a conversion from an Auspex filesystem and so some of the setup was historical in nature (for instance, we kept the concept of multiple home directories even though the filer is one large filesystem). We did increase each users quota to 500M per user, which was entered individually for ~ 350 users into the quotas file. There are two users that are reported as being over their 500M quota when you run quota report on the NetApp box. So they can not do any work since it bounces them off with an "exceeds quota" type message. If I run du -sk [username] on the Solaris box it shows the guy using 81Kb, far from over 500M?!?!. Quota report also shows him using 9658 files (which is also not true.. if I run "ls -aR | wc -l" I get 4917). I had to add a new user the other day. The process creates 11 dot files (.login, .cshrc, etc.) but quota report shows 139 files!! Does anyone have a clue what is going on? I had to increase this guy's quota to 1G to get around the problem until I can resolve it but I can't keep handing out 1G like candy at Halloween (although I'm sure NetApp would love that). Cheryl Cruddace LLNL

4 3

Re: Accessing a filer from two NT domains which are on two seperate subnets
by Keith Brown 08 Jan '99

08 Jan '99

>My question is : > >Is there some way, using netbios alias or some other alias mechanism, >whereby I could make the filer appear in the second domain's Network >Neighbourhood, so that the users in the second domain could access it >directly through their own port on the network card, instead of having >to go through the first one ? Actually, the particular filer interface that a client would use to communicate with a filer is not something that is directly influenced by the domain that it or the filer have been configured into. Domains are things that sit at a much higher level of abstraction than ip subnets and cables and so forth. On reading your message, it looks like you have configured all the clients on one subnet to be in one domain, and all the clients on another to be in a second domain, so you have got a one-to-one relationship between "clients that are in a X domain" and "clients that are on a particular ip-subnet", but that need not necessarily be the case. You could easily have half the clients on each subnet in one domain, and the other half from each subnet in another domain, but clients from *both* domains on a single subnet will normally communicate with the filer through the filer interface that is attached to that subnet, irrespective of which of the two domains the filer has been put into. Make sense? Keith

1 0

unsubscribe
by Mendoza, Kathy 08 Jan '99

08 Jan '99

unsubscribe

1 0

Accessing a filer from two NT domains which are on two seperate subnets
by Raymond Brennan 08 Jan '99

08 Jan '99

Hello All, I've got a F520 with a quad card on a 100BaseT network. I use three of the network interfaces and have given them IP addresses on three seperate subnets. My router has also got only the same three subnets enabled on it. I have got two NT domains, which are on two seperate subnets. The filer belongs to one NT domain. If someone in the second NT domain (which is on a seperate subnet) wants to access the filer from an NT box, then they have to go to the first NT domain to see it. This has an effect on network traffic etc. My question is : Is there some way, using netbios alias or some other alias mechanism, whereby I could make the filer appear in the second domain's Network Neighbourhood, so that the users in the second domain could access it directly through their own port on the network card, instead of having to go through the first one ? Yours Confusedly, Raymond Brennan

1 0

Re: Gb-ether and Extreme (Re: Routing Anomaly..fixed (and how))
by Jim McCoy 04 Jan '99

04 Jan '99

>On Thu 24 Dec, 1998, Bruce Walker <bmw(a)visgen.com> wrote: >> mark writes: >> > So, if I was hoping for high performance for a news spool on a filer [...] >> > I'd use Gb-ether into an Extreme switch. [...] >I'd be surprised if the problem was a complete show-stopper, and I too would >love to know if anyone has good or bad experiences using NetApp's with Gb-ether >they would like to share. We routinely run 630s and 760s on gb ethernet to Alteon AceDirector180 switches. No real problems yet, OTOH if there are problems to be found we are bound to run into them at the least convenient moment... jim mccoy Yahoo! Inc.

1 0

Re: 2 questions.
by guy＠netapp.com 29 Dec '98

29 Dec '98

> Ok..what is the maximum file size allowed in later versions. Thats what I > was looking for. 2^44 bytes (2^32 4K file system blocks). (I'm not sure what the maximum *filer* size is, but it may be less than that, in which case files of that size would have to be holey files.)

3 4

RE: Gb-ether and Extreme (Re: Routing Anomaly..fixed (and how))
by Shaw, Kevin 28 Dec '98

28 Dec '98

I direct connect a sun E6500 to a F760 using Gb ethernet. I have an Oracle data base on this filer... No problems. -----Original Message----- From: mds(a)gbnet.net [mailto:mds@gbnet.net] Sent: Monday, December 28, 1998 2:22 PM To: Bruce Walker Cc: sirbruce(a)ix.netcom.com; toasters(a)mathworks.com Subject: Re: Gb-ether and Extreme (Re: Routing Anomaly..fixed (and how)) On Thu 24 Dec, 1998, Bruce Walker <bmw(a)visgen.com> wrote: > mark writes: > > > > So, if I was hoping for high performance for a news spool on a filer [...] > > I'd use Gb-ether into an Extreme switch. > > Have you actually done this? I have a 720 with Gigabit ether No, I've never had my hands on such advanced networking kit, though I know of a site using them happily enough. The principle is sound...but theory doesn't necessarily matter a fig when your working to a deadline, for example. I'd be surprised if the problem was a complete show-stopper, and I too would love to know if anyone has good or bad experiences using NetApp's with Gb-ether they would like to share. >-- End of excerpt from Bruce Walker

1 0

Re: Gb-ether and Extreme (Re: Routing Anomaly..fixed (and how))
by mds＠gbnet.net 28 Dec '98

28 Dec '98

On Thu 24 Dec, 1998, Bruce Walker <bmw(a)visgen.com> wrote: > mark writes: > > > > So, if I was hoping for high performance for a news spool on a filer [...] > > I'd use Gb-ether into an Extreme switch. > > Have you actually done this? I have a 720 with Gigabit ether No, I've never had my hands on such advanced networking kit, though I know of a site using them happily enough. The principle is sound...but theory doesn't necessarily matter a fig when your working to a deadline, for example. I'd be surprised if the problem was a complete show-stopper, and I too would love to know if anyone has good or bad experiences using NetApp's with Gb-ether they would like to share. >-- End of excerpt from Bruce Walker

1 0

2 questions.
by Jaye Mathisen 27 Dec '98

27 Dec '98

Using 4.3.4 on a F540, 256MB RAM, 8MB NVRAM. I'm using the oh-so-secret method of using SQL server on a filer... Initial tests with small databases have been great. Big problem: I can't create a database device larger than 4GB's on the CIFS share. Anything up to 4 is fine, anything over, uh huh. Local drives can handle larger, but that's not where I want to be... 1) Any ideas on whats wrong here? 2) Anybody else using a filer under SQL server?

2 3

Re: 2 questions.
by guy＠netapp.com 27 Dec '98

27 Dec '98

> > If so, the problem is that we don't support files >4GB in releases prior > > to ONTAP 5.1. > --- > > And what about later versions (just because the question is so obviously > un-answered)? If we didn't support it in 5.1 or later versions, I would have said "the problem is that we don't support files >4GB" without adding "prior to ONTAP 5.1" - I assumed, apparently incorrectly, that it was obvious from the way I stated it....

1 0

Re: Routing Anomaly..fixed (and how)
by mds＠gbnet.net 24 Dec '98

24 Dec '98

On Wed 23 Dec, 1998, sirbruce(a)ix.netcom.com wrote: > The only other thing I haven't seen mentioned is number of disks. This can > be a limiting factor in performance, particularly towards the lower end. > Having more disks that you need will not make you appreciably faster, but > having too few will be very noticeable. Oh too true, especially as data on disks is becoming relatively less accessible in high-throughput applications. Brian Wong's book has a lucid explanation for why more spindles of lower capacity can make a hell of a lot more sense than fewer, bigger disks. > > In current filers, I don't know what the right ratio would be. I used to > use about 100-200ops/disk as a rule of thumb, but of course it depends on the > kid of ops. If you have an array of 21-28 disks, then I doubt that's the > limiting factor. If you only have 14, you might consider adding some more > to the filesystem and seeing if that improves your write throughput. Adding > more NVRAM is of course good if possible for your filer. I'd speculate that the new multiple RAID group feature in DOT5 also has an important performance dimension, besides the reliability aspects. For writes, and degraded reads, the fewer disks in a RAID group the higher the performance I'd expect in the RAID calculation. Now, assuming a volume has a number of RAID groups, I'd presume the WAFL filesystem will still consume blocks over all the RAID groups and make attempts to write whole stripes into each RAID group, thereby gaining the fullest bandwidth available to the platters, given the calculations that have to be done on the data. One interesting thing to note is how much the filer needs to read from disk while performing write-intensive work. If the filer isn't reading much from disk at all, then you're laying down fresh data stright from memory and the machine will be running at full tilt. If there's quite a lot of reading going on then the machine is having to work harder to lay the data down on the platters. This is more likely to happen when the volume is on the full side, has been running with some degree of churn for a while and contains relatively small files. Ie old data being removed to be replaced by new data - as in usenet news spools. So, if I was hoping for high performance for a news spool on a filer I'd be thinking of a 760 with many 9GB disks (the smallest available I believe) in an unclustered configuration. I'd also play with the RAID group sizes until I'd struck a rough balance between the costs, the performance and the resilience. I'd use Gb-ether into an Extreme switch. I'd toy with using the no-atime option, perhaps play with both NFS2 and NFS3 to see which gave best performance. And I'd use different hosts for spooling the news, and for serving the news...but that's application level stuff and far less easy to work on and comment on. >-- End of excerpt from sirbruce(a)ix.netcom.com

3 2

Re: Routing Anomaly..fixed (and how)
by sirbruce＠ix.netcom.com 24 Dec '98

24 Dec '98

The only other thing I haven't seen mentioned is number of disks. This can be a limiting factor in performance, particularly towards the lower end. Having more disks that you need will not make you appreciably faster, but having too few will be very noticeable. In current filers, I don't know what the right ratio would be. I used to use about 100-200ops/disk as a rule of thumb, but of course it depends on the kid of ops. If you have an array of 21-28 disks, then I doubt that's the limiting factor. If you only have 14, you might consider adding some more to the filesystem and seeing if that improves your write throughput. Adding more NVRAM is of course good if possible for your filer. I'd also do the standard systems reviews, making sure that your interfaces are running at the right speeds, that the routing on the filer is set up correctly (although for ethernet response packets I don't think those are routed - then again, that may have changed; it's complicated depending on the actual media type), and so on. Still, 20MB/sec with your network configuration doesn't sound bad to me. I'm not sure you'll be able to do too much better than that. If you want more, try slapping a gigabit ethernet card in the puppy. :) Bruce

1 0

Re: Routing Anomaly..fixed (and how)
by Keith Brown 23 Dec '98

23 Dec '98

>Yes, I have 32Mb of RAM, but at 50% or 10 seconds, the NVRAM flushes. I >only buffer 16Mb before it takes a data dump *grin* if I understand >correctly. You do understand correctly, but while the "flush" is occuring (actually, it's called a consistency point, or CP for short), the other 16 MB of NVRAM is still being put to good use journaling new file system modifications being made by clients. In other words, you and your filer are enjoying a full 32 MB of "benefit" from the NVRAM. In other words, it is *all* being used for something. There is no waste here! :-) >In a clustered environment, I just use that other half of NVRAM >to buffer to the other box instead of mirroring myself in case of a write >failure (power, etc). In an unclustered environment, you are using that "other half" of NVRAM now. When you cluster two systems together though, well... we have a white paper that tells you what happens much better than I can here: http://www.netapp.com/technology/level3/1004.html >Two clients, both on separate 100Mb interfaces, creating spool files for my >news services using 'dd if/dev/zero of=<filename> count=<x>'. Hmmm. Benchmarks don't get much more write intensive than this, do they. :-) So large sequential writes then. Fair enough. >When I was seeing about 1300 NFSops/Sec and 18-20MegaBytes of input traffic, >the CPU on the F760 would peg. Yes. Writing is a more CPU intensive function than reading. Lots of shunting things into NVRAM and performing of RAID calculations etc... None of that needs to be done on reads. Anyway, a *sustained* agregate of ~20 MBytes/Sec write performance sounds excellent to me. You only have ~25 MBytes/Sec worth of pipes coming into the filer (2 x 100Mbits/Sec), and the thing is stashing things away onto disks, with RAID, at 80% of the theoretical maximum rate at which data can arrive at its doorstep from the network. Nope... nothing to be ashamed of there I'm afraid. The F760 is a beast and a half! ;-) Keith

1 0

Re: Routing Anomaly..fixed (and how)
by Keith Brown 22 Dec '98

22 Dec '98

Jeff... I'm glad your routing anomilies got sorted out. Regarding your subsequent mail... >Also, Ive noticed a max input rate of about 20Mb/sec into the filter, then >the CPU tops out on my F760. > >I understand that the NVRAM can only process 16Mb at a time before the >system really gums up, but I did expect a little more input performance. > >Does anyone have any official or un-official comments on this? If your F760 is still of the unclustered variety (which I'm sure it is, as we haven't quite yet shipped the F700 series cluster release of Data ONTAP) then you will currently be enjoying the full 32 MB capacity of your NVRAM. In a clustered pair of filers however, each systems NVRAM is mirrored to the other, meaning that each filer's individual *effective* NVRAM capacity for the file system(s) that they are individually journaling is halved to 16 MB. They each still physically have 32 MB of NVRAM of course. On the performance front, I'm afraid you would need to better stipulate precisely what you mean by a "20 Mb/sec input rate" for me or anyone else at NetApp to be able to make a useful comment on whether it looks about right or not. How are you determining this number? Is this MBytes/Sec or MBits/Sec? How many clients are you using, and through what networks are they talking to the filer? What software are you running? What is that software really doing on the I/O front? How are the timings being made? Etc... Keith

4 3

ndmpcopy for linux?
by Gil Smith 22 Dec '98

22 Dec '98

Hi, Has anyone compiled and used the ndmpcopy utility on Linux? If so can you pass on any required changes directly to me. --Thanks, Gil

2 1

RE: ClearCase VOBs on NetApp
by Clarke, Bruce 21 Dec '98

21 Dec '98

Network Appliance is working closely with Rational to fully test the procedures and risks for putting the VOB database on the filer. As has been mentioned, putting the various storage pools has always worked and many customers use this to great advantage. Until the move of Network Appliance into the database space, putting the VOB database on the filer was considered risky. Now that both Oracle and Sybase have certified NetApp, it is just a matter of testing, documentation, and time before Rational provides similar certification since theirs is just a modified Raima database anyway. Some discussion on this topic has already taken place on the Clearcase Users Group discussion groups (CCIUG). If you can, I would suggest accessing those archives for the last 2 months to see where this discussion has gone on that forum and get firsthand Rational's response and plans. Bruce Clarke Network Appliance, Inc. -----Original Message----- From: Nick Thompson [mailto:nickthompson@lucent.com] Sent: Friday, December 18, 1998 1:51 AM To: Toasters(a)mathworks.com Subject: ClearCase VOBs on NetApp Hi, We are just starting to deploy ClearCase... I have been speaking to engineers at Network Applications who assured me that it is possible to create the WHOLE of a ClearCase VOB (including the db) on a NetApp filer (and any other NFS filer for that matter). Since Rational told me this was not possible, I got curious and tried it anyway. It worked! Has anybody else here tried this configuration? Had any problems? According to the ClearCase docs, if you can snapshot a complete VOB (as on a Toaster) then it is not necessary to lock it prior to backup (though some disk may be lost to empty containers if you restore from it). Anybody tried/trust this or would you advise against it? Potentially this could make our backup scheme MUCH simpler. Thanks, -- _ _______ __ __ Nick Thompson | Lucent Technologies / \/ / / __\/ // /\ Lucent Technologies | (Bell Labs Innovations) / / / /__/ _ / / Microelectronics House | nickthompson(a)lucent.com /_/\_/_/____/_/\\_\/ Kings Ride, Ascot | Tel: +44 1344 865951 \_\/\\_\____\_\/ \_\ Berkshire SL5 8AD | Fax: +44 1344 865959

1 0

Re: 5.x Restore cannot read 4.x dumps
by Scott Miller 18 Dec '98

18 Dec '98

On Dec 18, 1:01pm, Stephen Manley wrote: > Subject: Re: 5.x Restore cannot read 4.x dumps > > Below is a dialog with our F630 running 5.1.2P1 trying to read a tape written > > with 4.x. We were able to recover data from this tape using a filer running > > 4.3R4D2. > > > > > > range> restore tvf rst0l > > RESTORE: Tape Block size should be 10240, not 64512. > > Fri Dec 18 13:00:12 EST [isp_main]: Tape block was shorter than byte count. Requested = 4096, Read = -50176 > > RESTORE: Could not initialize media. > > range> restore tvfb rst0l 64512 > > RESTORE: No buffers available. > ... > We figured people would get bitten by this change, so we added the error > message (which is obviously not effective) > RESTORE: Tape Block size should be x, not y. > > Now, how do I know the error message didn't work? This message didn't work because it is not obvious. I'd recommend changing the message to somthing like: RESTORE: restore defaulted to a tape block size of 63K RESTORE: your dump was written with a tape block size of 10K RESTORE: re-trying the restore with the block size of 10K ... and then doing the restore with the blocksize that is on the tape. Since you could print the message, you have all the information needed to do The Right Thing (tm) on the users behalf. If I'm doing a restore, it's usually to recover from a disaster. When I'm recovering from a disaster, people are anxious for their data, I'm tired, frustrated and not in the mood to deal with balky restore software. -Skottie -- ----- Scott Miller | Walt Disney Feature Animation skottie(a)fa.disney.com | Digital Studio Technology

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