toasters

toasters@lists.teaparty.net

13531 discussions

RE: Input/output errors from Linux clients?
by Hubbard, Dan 26 Apr '01

26 Apr '01

Ditto here. Have 5 machines all in production working for 4+ months now..all 2.2 kernels. -----Original Message----- From: Colm Buckley [mailto:colm@tuatha.org] Sent: Thursday, April 26, 2001 10:29 AM To: toasters(a)mathworks.com Subject: Re: Input/output errors from Linux clients? > Is anyone surprised that linux and NFS still do not play nice together? Ooh, *baguette*! Linux's NFS has been "okay" since the mid-2.2 kernels, and I've had no problems whatsoever with it as of 2.4. Yes, things were Bad in the past, but truly, honestly, surely, everything's okay now. Colm (de-bitching lists since 1995...) -- Colm Buckley, NewWorld Commerce, 16 South Cumberland St., Dublin 2, Ireland. Personal: +353 87 2469146 / colm(a)tuatha.org / http://www.tuatha.org/~colm/ Business: +353 1 4334334 / colm(a)nwcgroup.com / http://www.nwcgroup.com/ No conductor in recorded history has ever asked for "less tenor".

1 0

legato ndmp
by neil lehrer 26 Apr '01

26 Apr '01

hi, does anybody know an engineer at legato that really understands ndmp and how it is implemented in networker? we can't get it to work[with the correct version, patch, and enablers] and their customer service/techsupport has been very poor. since we already use legato for other things, management would prefer to stay with it. thanks.

1 0

Defragging / Balancing
by Andrew Siegel 26 Apr '01

26 Apr '01

Here's something that I've been curious about for a while: how does WAFL deal with fragmentation and balancing? For example, when you add a new disk to a volume, does DataONTAP/WAFL do any shifting of data to the new disk in the background in order to balance out disk usage across the volume? Does any kind of defragmentation take place automagically? (Is fragmentation even an issue with WAFL?) I ask because we are having just these issues with another manufacturer's journaling file system, requiring us to spend significant time and effort to stay on top of. -- Andrew Siegel, Head of Systems Blue Sky Studios, Inc. White Plains, NY 914-259-6336 direct abs(a)blueskystudios.com

1 0

More info on Input/Output errors...
by Nick's Lists 26 Apr '01

26 Apr '01

A bit more info I forgot to include in the first place along with responses to some other folks: We're using 2.2.16-21.7.1smp - We're holding off on 2.4 for now, as our vendor, VA Linux doens't feel comfortable with us moving up to 2.4 yet (I know, we are pushing it, but we don't want to go against their recommendations for now). If this problem is resolved in 2.4, then that'll be another thing to push us towards 2.4. Network wise - Transport is UDP This is occuring on clients in multiple sites, against 7 different filers as well. I don't see how it could be a network problem tho, but I'll go over the configuration of the filers. In all cases, the clients and servers are all on the same subnet, and in some cases, on the sames switch. I've verified that we are getting 100 mbit full duplex negotiation to the switch. Andrew - good suggestion on the soft vs hard. I'll go over my logic for that, and try to figur eout why we're doing that. It may be ahold over from when I had a <insert name of compeditors product here> that was heavily loaded and was responding very slowly. The thing is - other OS's aren't experiencing this problem in this environment, just the Linux boxes. That leads me to think its not a common problem like network of filer config. Thanks! - Nick ----- Original Message ----- From: "Traitel, Eyal" <eyal(a)netapp.com> To: "'Nick's Lists'" <mrlist(a)noid.org> Sent: Thursday, April 26, 2001 12:08 AM Subject: RE: Input/output errors from Linux clients? > Might be unrelated, and not exactly hitting your problem, but from all we know and hear and encounter, move up to kernels 2.4 and up for real use of NFS in Linux. > > What's your kernel version then ? > > Eyal. > > ---------------------------------------------------------------------- > eTraitel - I'm the new eBuzzword around !!! > ---------------------------------------------------------------------- > Eyal Traitel - Filer Escalation Engineer > CNA, MCSE, CSA, LCA, NetApp CA > > Network Appliance BV > Holland Office Center > Kruisweg 799b > 2132 NG, Hoofddorp > The Netherlands > Office: +31 23 567 9685 > Cellular: +31 6 5497 2568 > Email: eyal(a)netapp.com > ---------------------------------------------------------------------- > Get answers NOW! - NetApp On the Web - http://now.netapp.com > ---------------------------------------------------------------------- > > > -----Original Message----- > From: Nick's Lists [mailto:mrlist@noid.org] > Sent: Thursday, April 26, 2001 8:18 AM > To: toasters(a)mathworks.com > Subject: Input/output errors from Linux clients? > > > I've been getting some strange errors from my Linux clients... > > cp: Input/output error > and > cat: write Input/output error > > They occur randomly when I try to copy or cat 20 - 40mb sized files to and > from a filer. Solaris (x86 and Sparc) servers copying the same size files > around are not experiencing the same problem. > > The file systems are all mounted with nfs2, soft, bg, intr. > > Has anyone else seen this before? > > TIA! > > - Nick > > >

2 1

FW: More info on Input/Output errors...
by Mark Baenziger 26 Apr '01

26 Apr '01

We had nearly identical problems with our Linux boxen (various kernels 2.2.5-2.2.14) running off of our filer. We tried dozens of different NFS settings without any luck. The only thing that finally got all of our NFS problems under control was going with a custom kernel with all of trond's various NFS patches in place. RFC compliance is a wonderful thing. Go ahead and compile new kernels with patches in place. Patches for various kernels are at http://www.fys.uio.no/~trondmy/src/. Additional information is (of course) at http://nfs.sourceforge.net/. You do not need to go with a 2.4 kernel for this; these patches can be applied (with varying levels of success) for kernels from 2.2.14 on up. Mark -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com]On Behalf Of Nick's Lists Sent: Thursday, April 26, 2001 8:58 AM To: toasters(a)mathworks.com Subject: More info on Input/Output errors... A bit more info I forgot to include in the first place along with responses to some other folks: We're using 2.2.16-21.7.1smp - We're holding off on 2.4 for now, as our vendor, VA Linux doens't feel comfortable with us moving up to 2.4 yet (I know, we are pushing it, but we don't want to go against their recommendations for now). If this problem is resolved in 2.4, then that'll be another thing to push us towards 2.4. Network wise - Transport is UDP This is occuring on clients in multiple sites, against 7 different filers as well. I don't see how it could be a network problem tho, but I'll go over the configuration of the filers. In all cases, the clients and servers are all on the same subnet, and in some cases, on the sames switch. I've verified that we are getting 100 mbit full duplex negotiation to the switch. Andrew - good suggestion on the soft vs hard. I'll go over my logic for that, and try to figur eout why we're doing that. It may be ahold over from when I had a <insert name of compeditors product here> that was heavily loaded and was responding very slowly. The thing is - other OS's aren't experiencing this problem in this environment, just the Linux boxes. That leads me to think its not a common problem like network of filer config. Thanks! - Nick ----- Original Message ----- From: "Traitel, Eyal" <eyal(a)netapp.com> To: "'Nick's Lists'" <mrlist(a)noid.org> Sent: Thursday, April 26, 2001 12:08 AM Subject: RE: Input/output errors from Linux clients? > Might be unrelated, and not exactly hitting your problem, but from all we know and hear and encounter, move up to kernels 2.4 and up for real use of NFS in Linux. > > What's your kernel version then ? > > Eyal. > > ---------------------------------------------------------------------- > eTraitel - I'm the new eBuzzword around !!! > ---------------------------------------------------------------------- > Eyal Traitel - Filer Escalation Engineer > CNA, MCSE, CSA, LCA, NetApp CA > > Network Appliance BV > Holland Office Center > Kruisweg 799b > 2132 NG, Hoofddorp > The Netherlands > Office: +31 23 567 9685 > Cellular: +31 6 5497 2568 > Email: eyal(a)netapp.com > ---------------------------------------------------------------------- > Get answers NOW! - NetApp On the Web - http://now.netapp.com > ---------------------------------------------------------------------- > > > -----Original Message----- > From: Nick's Lists [mailto:mrlist@noid.org] > Sent: Thursday, April 26, 2001 8:18 AM > To: toasters(a)mathworks.com > Subject: Input/output errors from Linux clients? > > > I've been getting some strange errors from my Linux clients... > > cp: Input/output error > and > cat: write Input/output error > > They occur randomly when I try to copy or cat 20 - 40mb sized files to and > from a filer. Solaris (x86 and Sparc) servers copying the same size files > around are not experiencing the same problem. > > The file systems are all mounted with nfs2, soft, bg, intr. > > Has anyone else seen this before? > > TIA! > > - Nick > > >

1 0

More Info - Re: Input/output errors from Linux clients?
by Nick's Lists 26 Apr '01

26 Apr '01

A bit more info I forgot to include in the first place - We're using 2.2.16-21.7.1smp - We're holding off on 2.4 for now, as our vendor, VA Linux doens't feel comfortable with us moving up to 2.4 yet (I know, we are pushing it, but we don't want to go against their recommendations for now). If this problem is resolved in 2.4, then that'll be another thing to push us towards 2.4. Network wise - Transport is UDP This is occuring on clients in multiple sites, against 7 different filers as well. I don't see how it could be a network problem tho, but I'll go over the configuration of the filers. In all cases, the clients and servers are all on the same subnet, and in some cases, on the sames switch. I've verified that we are getting 100 mbit full duplex negotiation to the switch. Andrew - good suggestion on the soft vs hard. I'll go over my logic for that, and try to figur eout why we're doing that. It may be ahold over from when I had a <insert name of compeditors product here> that was heavily loaded and was responding very slowly. The thing is - other OS's aren't experiencing this problem in this environment, just the Linux boxes. That leads me to think its not a common problem like network of filer config. Thanks! - Nick ----- Original Message ----- From: "Traitel, Eyal" <eyal(a)netapp.com> To: "'Nick's Lists'" <mrlist(a)noid.org> Sent: Thursday, April 26, 2001 12:08 AM Subject: RE: Input/output errors from Linux clients? > Might be unrelated, and not exactly hitting your problem, but from all we know and hear and encounter, move up to kernels 2.4 and up for real use of NFS in Linux. > > What's your kernel version then ? > > Eyal. > > ---------------------------------------------------------------------- > eTraitel - I'm the new eBuzzword around !!! > ---------------------------------------------------------------------- > Eyal Traitel - Filer Escalation Engineer > CNA, MCSE, CSA, LCA, NetApp CA > > Network Appliance BV > Holland Office Center > Kruisweg 799b > 2132 NG, Hoofddorp > The Netherlands > Office: +31 23 567 9685 > Cellular: +31 6 5497 2568 > Email: eyal(a)netapp.com > ---------------------------------------------------------------------- > Get answers NOW! - NetApp On the Web - http://now.netapp.com > ---------------------------------------------------------------------- > > > -----Original Message----- > From: Nick's Lists [mailto:mrlist@noid.org] > Sent: Thursday, April 26, 2001 8:18 AM > To: toasters(a)mathworks.com > Subject: Input/output errors from Linux clients? > > > I've been getting some strange errors from my Linux clients... > > cp: Input/output error > and > cat: write Input/output error > > They occur randomly when I try to copy or cat 20 - 40mb sized files to and > from a filer. Solaris (x86 and Sparc) servers copying the same size files > around are not experiencing the same problem. > > The file systems are all mounted with nfs2, soft, bg, intr. > > Has anyone else seen this before? > > TIA! > > - Nick > > >

1 0

More Info - Re: Input/output errors from Linux clients?
by Nick S 26 Apr '01

26 Apr '01

1 0

RE: "Case survey" stuff from NetApp
by Mohler, Jeff 26 Apr '01

26 Apr '01

I would think that it would take more of my time, to load up the WWW browser to work it there, than to insert five numbers between the ()'s. -----Original Message----- From: Sullivan, Justin [mailto:Justin.Sullivan@netapp.com] Sent: Wednesday, April 25, 2001 1:22 PM To: 'Chris Thompson'; toasters(a)mathworks.com Subject: RE: "Case survey" stuff from NetApp Capital idea.. We would certainly like to have more surveys returned - we should like to see EVERY survey returned. You have an excellent point - why not have a web-based survey form? Easy, friendly, quick.. Back in the days when I was a NetApp customer, I admit, I threw them away too - too much trouble to fill out the e-mail form; who has that kind of time anymore? I have passed your idea along to the folks in Customer Satisfaction who can make this happen. Cheers! --- Justin Sullivan Network Appliance, Technical Support Engineer Network Appliance Certified Associate justins(a)netapp.com Get answers NOW! - NetApp On the Web - http://now.netapp.com Tech Support Hotline: 1-888-4NETAPP -----Original Message----- From: Chris Thompson [mailto:cet1@cus.cam.ac.uk] Sent: Wednesday, April 25, 2001 3:04 PM To: toasters(a)mathworks.com Subject: "Case survey" stuff from NetApp What do you do with the the e-mail that arrives from Quality.Manager(a)netapp.com after a problem is closed with NetApp, inviting you to fill in a "5 STARS Survey" form? I admit I throw them in the e-wastebin. It seems to me that providing only a method of returning these (probably only marginally meaningful) answers by e-mail is absolutely pessimal. If they really want to increase the response rate, surely NetApp should provide a method using forms at now.netapp.com? My vague recollection is that at one time a WWW-browser-based method was offered, but this doesn't seem to be the case now. Chris Thompson University of Cambridge Computing Service, Email: cet1(a)ucs.cam.ac.uk New Museums Site, Cambridge CB2 3QG, Phone: +44 1223 334715 United Kingdom.

2 1

RE: Adding a 36gb shelf to F740 with 18gb shelves
by Sullivan, Justin 26 Apr '01

26 Apr '01

So, my apologies.. Sam corrected me. This changed with the new blockmap structure in ONTAP 6.0 and above (all my testing was with 5.3.7R3). I do not need "reprogramming" - I don't live in California. :) I was not saying create a new volume. I was saying create a new RAID group in the existing volume. This is where the 'raidsize' volume option comes in. You can do it manually, of course, but 'raidsize' is set for each volume (not the whole filer), and it's less error-prone to change 'raidsize' to the desired value. You can mix drive sizes within volumes without any problem, but all the drives in a given RAID group in the volume should be the same size. Set 'raidsize' on the volume in question to the size of the existing 18GB drive RAID group, run 'sysconfig -r' for details. When you go to add the 36GB drives - you will have to add at least two if the RAID group is already at 'raidsize' - it will automatically create a new RAID group with your spiffy new 36GB drives. You will lose one drive to parity, but it could avoid performance woes down the road. Lacking ASCII artistic skills, imagine having ten small juice glasses and two big ice tea glasses all lined up in a row. This is where your water (essential for all life, also known as "data") goes. As you fill up the small juice glasses, you can only keep pouring into the two big ice tea glasses, but you won't be able to pour it nearly as fast as you could if you were filling up more glasses. It is still a bad idea to mix drive sizes within a RAID group, because of what can happen to performance as things start to fill up. Creating a new RAID group is still (in my opinion), the way to go, but of course it's your data, your way. If capacity is more important than performance, you don't want to use a precious 36GB drive for parity. - Justin -----Original Message----- From: Jeff Bryer [mailto:bryer@sfu.ca] Sent: Wednesday, April 25, 2001 9:48 PM To: Justin.Sullivan(a)netapp.com Cc: toasters(a)mathworks.com Subject: Re: Adding a 36gb shelf to F740 with 18gb shelves Thanks Justin. So we can't change the parity disk from an 18gb disk to a 36gb disk kind of on the fly. We'd have to create a new volume and do a vol copy or some such thing. Not sure where we got the idea we could 'upgrade the parity disk'. So we either have to buy 18gb shelves or re-architect our layout. (or waste half a disk adding a 36gb to an 18gb set but that's not a great option as our volume is growing quickly). > > Yes, you're going to have a problem.. :( > > If you add 36GB drives to an existing RAID group of 18GB disks, the 36GB disks will be reported and used as 18GB drives. > > Once a RAID group is established with a parity drive of xxGB, any disks added to that RAID group will only be used to the same capacity (xxGB) as the parity drive. > > If you reverse the situation, and add 18GB drives to an existing 36GB drive RAID group, all drives will be used to their full capacity, but you will run into performance troubles as the volume fills up and the system is forced to write only to the larger disks, thereby reducing the number of physical spindles working for you. > > You will definitely want to create a new RAID group (or a new volume) for your 36GB drives to take advantage of the higher capacity. > > --- > Justin Sullivan > Network Appliance, Technical Support Engineer > Network Appliance Certified Associate > justins(a)netapp.com > Get answers NOW! - NetApp On the Web - http://now.netapp.com > Tech Support Hotline: 1-888-4NETAPP > -- Jeff Bryer bryer(a)sfu.ca Systems Administrator (604) 291-4935 Academic Computing, Simon Fraser University

1 0

"Case survey" stuff from NetApp
by Chris Thompson 26 Apr '01

26 Apr '01

What do you do with the the e-mail that arrives from Quality.Manager(a)netapp.com after a problem is closed with NetApp, inviting you to fill in a "5 STARS Survey" form? I admit I throw them in the e-wastebin. It seems to me that providing only a method of returning these (probably only marginally meaningful) answers by e-mail is absolutely pessimal. If they really want to increase the response rate, surely NetApp should provide a method using forms at now.netapp.com? My vague recollection is that at one time a WWW-browser-based method was offered, but this doesn't seem to be the case now. Chris Thompson University of Cambridge Computing Service, Email: cet1(a)ucs.cam.ac.uk New Museums Site, Cambridge CB2 3QG, Phone: +44 1223 334715 United Kingdom.

3 2

Client Pak Restore Problems
by David C Baumgardner 26 Apr '01

26 Apr '01

Is anyone using Legato's Client Pak backing up to a windows system? I have been having on going problems getting restores from tape for the last four months. Each release they issue seems to work fine for a while then quits working(as far as the restore is concerned). The only recovery that works now is the SSID recover with the "-a" option for single file or directory. That can take up to 6 to 7 hours by the time it runs through the save set. I would like to hear from anyone having troubles with Client Pak . David Baumgardner CAD/CAM Administrator Container Engineering Owens Brockway Glass Containers Voice 419-247-8274 Fax 419-247-7608 Email David.Baumgardner(a)owens-ill.com

1 0

RE: Adding a 36gb shelf to F740 with 18gb shelves
by Sullivan, Justin 26 Apr '01

26 Apr '01

Yes, you're going to have a problem.. :( If you add 36GB drives to an existing RAID group of 18GB disks, the 36GB disks will be reported and used as 18GB drives. Once a RAID group is established with a parity drive of xxGB, any disks added to that RAID group will only be used to the same capacity (xxGB) as the parity drive. If you reverse the situation, and add 18GB drives to an existing 36GB drive RAID group, all drives will be used to their full capacity, but you will run into performance troubles as the volume fills up and the system is forced to write only to the larger disks, thereby reducing the number of physical spindles working for you. You will definitely want to create a new RAID group (or a new volume) for your 36GB drives to take advantage of the higher capacity. --- Justin Sullivan Network Appliance, Technical Support Engineer Network Appliance Certified Associate justins(a)netapp.com Get answers NOW! - NetApp On the Web - http://now.netapp.com Tech Support Hotline: 1-888-4NETAPP -----Original Message----- From: Jeff Bryer [mailto:bryer@sfu.ca] Sent: Wednesday, April 25, 2001 6:26 PM To: toasters(a)mathworks.com Subject: Adding a 36gb shelf to F740 with 18gb shelves We have an F740 with 3 shelves of 18GB disks. It's config'd as two volumes (two raid sets). We've almost outgrown the capacity (we're down to one hot spare and a free disk). We want to add a new disk shelf to the system, but we're wondering if we're going to lose anything by going to a shelf of 36gb disks. We don't have any intention of adding new volumes (or new raid sets), just growing the existing volumes. Can we add a shelf of 36gb disks and then switch our 2 existing 18gb parity disks over to 36gb disks? We'd keep a 36gb disk as a hot spare and use all of the 18gb disks in the volumes (including the 2 freed up parity disks and the freed up hot spare). Are we going to run into problems with this? Or do we need to keep an 18gb disk around as a hot spare as well (ie when an 18gb disks dies and uses a 36gb disks as a hot spare, can we force it to go back to the 18gb disk once the faulty 18gb disk has been replaced?). Are the 36gb disks going to be used at full capacity in this situation (when mixed with the 18gb)? Or is there a better way to do this? Thanks, -- Jeff Bryer bryer(a)sfu.ca Systems Administrator (604) 291-4935 Academic Computing, Simon Fraser University

5 5

RE: Input/output errors from Linux clients?
by Bond, Andrew 26 Apr '01

26 Apr '01

Soft mounts are not recommended if you are writing to the filesystem. Use hard mounts. The errors are likely caused by network problems, which could be a simple misconfiguration. You should open a call with NetApp Customer Support. Regards, Andrew > -----Original Message----- > From: Nick's Lists [mailto:mrlist@noid.org] > Sent: 26 April 2001 07:18 > To: toasters(a)mathworks.com > Subject: Input/output errors from Linux clients? > > > I've been getting some strange errors from my Linux clients... > > cp: Input/output error > and > cat: write Input/output error > > They occur randomly when I try to copy or cat 20 - 40mb sized > files to and > from a filer. Solaris (x86 and Sparc) servers copying the > same size files > around are not experiencing the same problem. > > The file systems are all mounted with nfs2, soft, bg, intr. > > Has anyone else seen this before? > > TIA! > > - Nick > > > > >

1 0

Fwd: Re: OT - swag (fwd)
by Jeff Innis 26 Apr '01

26 Apr '01

3 2

RE: "Case survey" stuff from NetApp
by Mike Sphar 25 Apr '01

25 Apr '01

It depends on how you open the browser. If it's just a pre-defined URL that takes you right to the survey form where you just type your answers and submit it, that would likely be faster for me. But it depends on your email client and your work habits, how long your machine takes to bring up a browser, etc. Ideally, if it were up to me, both options would work. There would be a URL at the top of the message and also the ability to reply with your answers. -- Mike Sphar - Sr Systems Administrator - Engineering Support Services - Remedy Corporation BOFH, GWP, MCP, MCP+I, MCSE, BFD -----Original Message----- From: Mohler, Jeff [mailto:jeff.mohler@netapp.com] Sent: Wednesday, April 25, 2001 3:44 PM To: Sullivan, Justin; 'Chris Thompson'; toasters(a)mathworks.com Subject: RE: "Case survey" stuff from NetApp I would think that it would take more of my time, to load up the WWW browser to work it there, than to insert five numbers between the ()'s. -----Original Message----- From: Sullivan, Justin [mailto:Justin.Sullivan@netapp.com] Sent: Wednesday, April 25, 2001 1:22 PM To: 'Chris Thompson'; toasters(a)mathworks.com Subject: RE: "Case survey" stuff from NetApp Capital idea.. We would certainly like to have more surveys returned - we should like to see EVERY survey returned. You have an excellent point - why not have a web-based survey form? Easy, friendly, quick.. Back in the days when I was a NetApp customer, I admit, I threw them away too - too much trouble to fill out the e-mail form; who has that kind of time anymore? I have passed your idea along to the folks in Customer Satisfaction who can make this happen. Cheers! --- Justin Sullivan Network Appliance, Technical Support Engineer Network Appliance Certified Associate justins(a)netapp.com Get answers NOW! - NetApp On the Web - http://now.netapp.com Tech Support Hotline: 1-888-4NETAPP -----Original Message----- From: Chris Thompson [mailto:cet1@cus.cam.ac.uk] Sent: Wednesday, April 25, 2001 3:04 PM To: toasters(a)mathworks.com Subject: "Case survey" stuff from NetApp What do you do with the the e-mail that arrives from Quality.Manager(a)netapp.com after a problem is closed with NetApp, inviting you to fill in a "5 STARS Survey" form? I admit I throw them in the e-wastebin. It seems to me that providing only a method of returning these (probably only marginally meaningful) answers by e-mail is absolutely pessimal. If they really want to increase the response rate, surely NetApp should provide a method using forms at now.netapp.com? My vague recollection is that at one time a WWW-browser-based method was offered, but this doesn't seem to be the case now. Chris Thompson University of Cambridge Computing Service, Email: cet1(a)ucs.cam.ac.uk New Museums Site, Cambridge CB2 3QG, Phone: +44 1223 334715 United Kingdom.

1 0

Seeking ONTAP version information
by John Jetmore 25 Apr '01

25 Apr '01

My company is potentially needing to upgrade the OS on one of our Filers (760). I believe that 6.0.1R3 is the latest in the 6.* family. Is it fairly stable? We are running 5.3.6R1 right now. Does anyone see any potential problems with that upgrade path? Any information or shared experiences would be appreciated. Thanks, --John Jetmore

2 1

Re: snapshot backup
by Rick Hulsey 25 Apr '01

25 Apr '01

It is different in that the snapshots represent read only copies of the files in the filesystem. When you use NFS to mount the snapshot and copy the files to tape, Legato will still look at the time stamps on the files and backup the files that have changed since the time stamp in the Legato client file index. This means that Legato will backup files that have changed since the last backup if you specify an incremental backup. However, that is different than the block level changes that are always ongoing in the WAFL filesystem, and it is that characteristic that enable snapshots to work. All of the stuff related to WAFL (internal filesystem stuff) is, and has to be transparent to the backup software and even the operating system. Based on all of that stuff it looks to me like Legato can not do what you are looking for. >>> "E.J. Jantz" <ej_jantz(a)bswintl.com> 04/25/01 08:55AM >>> just a question: If the 'snapshots are "copy on writes' doesn't that mean that a backup of .snapshot is equal to and functions as a full backup? If not, can Legato using a NFS mount on a unix server capture just the info the .snapshot uses? If so, how? thanks... EJ Jantz, IST Dept. BSW International One West Third, Ste 100 Tulsa, OK 74103-3505 918 582-8771 ej_jantz(a)bswintl.com 918 295-4166 918 587-3594 fax >>> "Clawson, Simon" <simon_clawson(a)mentorg.com> 4/25/01 4:32:08 AM >>> Is it possible to backup just the .snapshot and then perform a restore from this data? I am looking for ways to speed up the backups and this seems to be a possible option. It was suggested to me by a colleague as a method of backing up Clearcase VOBS, but would it be viable for the whole filer. Of course I would need to run a full backup at tleast once a month, but this would help me speed up most of my weekly runs. Any ideas/suggestions/views? Simon Simon Clawson HDS Team Systems Administrator Mentor Graphics Uk Rivergate London Road Newbury Berkshire RG14 2QB <<Clawson, Simon.vcf>>

1 0

Adding a 36gb shelf to F740 with 18gb shelves
by Jeff Bryer 25 Apr '01

25 Apr '01

We have an F740 with 3 shelves of 18GB disks. It's config'd as two volumes (two raid sets). We've almost outgrown the capacity (we're down to one hot spare and a free disk). We want to add a new disk shelf to the system, but we're wondering if we're going to lose anything by going to a shelf of 36gb disks. We don't have any intention of adding new volumes (or new raid sets), just growing the existing volumes. Can we add a shelf of 36gb disks and then switch our 2 existing 18gb parity disks over to 36gb disks? We'd keep a 36gb disk as a hot spare and use all of the 18gb disks in the volumes (including the 2 freed up parity disks and the freed up hot spare). Are we going to run into problems with this? Or do we need to keep an 18gb disk around as a hot spare as well (ie when an 18gb disks dies and uses a 36gb disks as a hot spare, can we force it to go back to the 18gb disk once the faulty 18gb disk has been replaced?). Are the 36gb disks going to be used at full capacity in this situation (when mixed with the 18gb)? Or is there a better way to do this? Thanks, -- Jeff Bryer bryer(a)sfu.ca Systems Administrator (604) 291-4935 Academic Computing, Simon Fraser University

1 0

RE: snapshot backup
by Melvin, Grant 25 Apr '01

25 Apr '01

Hi Igor, > There's one potential problem - if your backup does not finish in time > between 2 snapshots, they will rotate (say, hourly.0 will become > hourly.1, and the new hourly.0 has started) , so in the end you're > backing up not what you started backing up. For scheduled backups one > should be cautious to, for example, do a full backup from weekly.0 and > incremental one from nightly.0, but if you're doing a on-the-fly > dump, always keep your snapshot schedule in mind. The snapshots used to rotate and dump would then crash the filer, since all its knowledge about a static filesystem just changed! So now we lock the snapshot that dump is backing up and the rotation of snapshots would fail which is also not good, but preserves data availability. You can create (or rename) snapshots with your own names, such as vob_snap, but once you do this you (the user) are managing the snapshot, i.e. dump creates & deletes its snapshots, snapmirror creates & deletes its snapshots, snap sched creates & ... well you get the picture. Cheers, Grant

1 0

Re: snapshot backup
by jharrington＠millermartin.com 25 Apr '01

25 Apr '01

We are using Backup Exec and to get around the time constraints of the snapshots came up with the following option. Backup exec will execute a command before the backup to create a new snapshot called backup-snap "rsh filer1 snap create vol1 backup-snap" Once that is done it will run the job we created to backup the newly created backup-snap snapshot and once it has finished it will delete the snapshot "rsh filer1 snap delete vol1 backup-snap" Igor Schein <igor(a)txc.com> on 04/25/2001 05:06:20 PM Please respond to igor(a)txc.com To: "Traitel, Eyal" <eyal(a)netapp.com> cc: "'E.J. Jantz'" <ej_jantz(a)bswintl.com>, toasters(a)mathworks.com, simon_clawson(a)mentorg.com (bcc: Joe Harrington/M&M) Subject: Re: snapshot backup On Wed, Apr 25, 2001 at 08:28:39AM -0700, Traitel, Eyal wrote: > Our snapshots are to be more exact "copy on new write" as opposed to "copy on old write", with the advantage of no penalty at all in terms of performance of taking the snapshot, and performance of filesystem after taking it (think of "copy on old write" - filesystem needs to read old block, write it in the backup location, then write the new block in old place - that's 3 I/Os. For us it's only 1). > We do not have to go back to specific locations on the disk, and that's a real advantage in terms of performance. > > In regards to full compared to snapshot - with NetApp, it's probably always better to backup snapshots, as they are point-in-time - when you've started your backup at 19:00, all of it will be consistent to 19:00 sharp, unlike traditional backup, which data changes while you backup... So - if you get me right - you can even do full backup from a snapshot... There's one potential problem - if your backup does not finish in time between 2 snapshots, they will rotate (say, hourly.0 will become hourly.1, and the new hourly.0 has started) , so in the end you're backing up not what you started backing up. For scheduled backups one should be cautious to, for example, do a full backup from weekly.0 and incremental one from nightly.0, but if you're doing a on-the-fly dump, always keep your snapshot schedule in mind. Igor

1 0

RE: fibre channel hubs
by Aiello, Tony 25 Apr '01

25 Apr '01

Neil, Hubs will create a problem for you in this application. I imagine you'd configure the unmanaged hub with two or more filers and the library all attached to a port on the hub. When things init, the filers will become aware of each other because they will be present in the map of devices on the Fibre Channel loop. They will then proceed to attempt to login with each other. As their both initiators, this won't go well. The other problem will be that resets propagate. Action by one filer could easily reset your tape drive causing the backup or restore to abort. It was for exactly these reasons we went with the managed hub solution. By partitioning or zoning properly each filer sees only the tape drives and propagation of resets or loop initializations is filtered. You might be interested in checking NOW for information on our real fabric implementation. It has some very nice features. You can connect many filers into a Brocade switch and share multiple tape drives. We've also added a feature called tape aliasing. Tape aliasing works in conjunction with these shared resource to insure that there's a consistent mapping of tape devices to logical tape names. It insures that from filer to filer if you want to use rst0a it will always be a tape drive at either a certain switch,port,id location or will always find a device with a particular world-wide name regardless of where its attached. You can choose. Check it out on NOW at: http://now.netapp.com/NOW/knowledge/docs/olio/SAN_GOLD/FC2/index.shtml Tony ------------------------------------------- Tony Aiello, Mgr. Storage SW mailto:taiello@netapp.com Ph:(408)822-6515 > -----Original Message----- > From: neil lehrer [mailto:nlehrer@ibb.gov] > Sent: Wednesday, April 25, 2001 12:46 PM > To: toasters > Subject: fibre channel hubs > > > has netapp qualified any of the vixel hubs for tape backup? > the hubs are much > cheaper than the vixel 8100 and i don't think the lesser > performance will be an > issue in my case - two filers. > > thanks >

1 0

RE: snapshot backup
by Traitel, Eyal 25 Apr '01

25 Apr '01

Our snapshots are to be more exact "copy on new write" as opposed to "copy on old write", with the advantage of no penalty at all in terms of performance of taking the snapshot, and performance of filesystem after taking it (think of "copy on old write" - filesystem needs to read old block, write it in the backup location, then write the new block in old place - that's 3 I/Os. For us it's only 1). We do not have to go back to specific locations on the disk, and that's a real advantage in terms of performance. In regards to full compared to snapshot - with NetApp, it's probably always better to backup snapshots, as they are point-in-time - when you've started your backup at 19:00, all of it will be consistent to 19:00 sharp, unlike traditional backup, which data changes while you backup... So - if you get me right - you can even do full backup from a snapshot... On Legato - I think you can just mention .snapshot/hourly.0 for example in your path of backup. If you use NDMP, you anyway backup a snapshot, as the filer's dump creates one automatically for you. Hope it somewhat helps, Eyal. ---------------------------------------------------------------------- eTraitel - I'm the new eBuzzword around !!! ---------------------------------------------------------------------- Eyal Traitel - Filer Escalation Engineer CNA, MCSE, CSA, LCA, NetApp CA Network Appliance BV Holland Office Center Kruisweg 799b 2132 NG, Hoofddorp The Netherlands Office: +31 23 567 9685 Cellular: +31 6 5497 2568 Email: eyal(a)netapp.com ---------------------------------------------------------------------- Get answers NOW! - NetApp On the Web - http://now.netapp.com ---------------------------------------------------------------------- -----Original Message----- From: E.J. Jantz [mailto:ej_jantz@bswintl.com] Sent: Wednesday, April 25, 2001 3:55 PM To: toasters(a)mathworks.com; simon_clawson(a)mentorg.com Subject: Re: snapshot backup just a question: If the 'snapshots are "copy on writes' doesn't that mean that a backup of .snapshot is equal to and functions as a full backup? If not, can Legato using a NFS mount on a unix server capture just the info the .snapshot uses? If so, how? thanks... EJ Jantz, IST Dept. BSW International One West Third, Ste 100 Tulsa, OK 74103-3505 918 582-8771 ej_jantz(a)bswintl.com 918 295-4166 918 587-3594 fax >>> "Clawson, Simon" <simon_clawson(a)mentorg.com> 4/25/01 4:32:08 AM >>> Is it possible to backup just the .snapshot and then perform a restore from this data? I am looking for ways to speed up the backups and this seems to be a possible option. It was suggested to me by a colleague as a method of backing up Clearcase VOBS, but would it be viable for the whole filer. Of course I would need to run a full backup at tleast once a month, but this would help me speed up most of my weekly runs. Any ideas/suggestions/views? Simon Simon Clawson HDS Team Systems Administrator Mentor Graphics Uk Rivergate London Road Newbury Berkshire RG14 2QB <<Clawson, Simon.vcf>>

2 1

RE: snapshot backup
by Melvin, Grant 25 Apr '01

25 Apr '01

Hi Simon, > Is it possible to backup just the .snapshot and then perform a restore > from this data? Yes, when restore looks at the backup tape it doesn't see the .snapshot/weekly.0 preamble so the backup to NetApp looks the same as if you hadn't specified a snapshot. As Steve pointed out some of the NDMP backup applications remember this preamble and need to be redirected on restores. > I am looking for ways to speed up the backups and this seems to be a > possible option. As our capacities grew snapshot create & delete times started to increase (pre 6.0 releases). With Data Ontap 6.0 we re-engineered our snapshot implementation and re-introduced "instantaneous" snapshot creation. Dump either via rsh, console or NDMP only ever backups from a snapshot either a named snapshot or one it creates (& manages) itself. If you are running a pre-6.0 release then I can believe you might see less impact on backups by using a named snapshot either from the snap sched set (we don't normally recommend you use these snapshots for dump because they get updated on the snap sched schedule and dump locks the snapshot its working on, e.g. the two schedules can conflict meaning you miss a valuable snap sched update) or created by the user using rsh snap create. In the past we've seen customers create a snpshot rather than let dump create it since they are going to perform multiple qtree or sub directory backups and just want one snapshot to be created. In Data Ontap 6.0 or beyond, there should be no noticable difference using dump on a named snapshot or when it needs to create its own snapshot. /etc/dumpdates is a database that dump uses to remember the time the last backup was done on the path used, so on an incremental dump we simply create a new snapshot, pick up the time of the last dump and then read out all files & directories in the new snapshot that have been modified since that timestamp. But maybe the time you are saving is actually on quiescing Clearcase and the VOB data and not on the snapshot creation. I wasn't clear. Cheers, Grant =========== grant(a)netapp.com ========== Grant Melvin === === Software Development Manager === === Data Availability & Management === |\ | __ ___ /\ __ __ === Network Appliance === | \ | |__ | /__\ |__| |__| === 475 East Java Drive === | \| |__ | / \ | | (R) === Sunnyvale === === California, 94089 === === Tel:(408)822-6761 =========== Network Appliance ========== Fax:(408)822-4578 -----Original Message----- From: Clawson, Simon [mailto:simon_clawson@mentorg.com] Sent: Wednesday, April 25, 2001 8:39 AM To: 'Traitel, Eyal'; 'E.J. Jantz'; toasters(a)mathworks.com; Clawson, Simon Subject: RE: snapshot backup That sounds like exactly what I meant. We currently backup / and everything below that. What I was hoping I could do was backup /.snapshot/weekly0 and then use this as my full backup. I user Veritas NetBackup to fire off the NDMP job on the filer, so I guess I would need to modify the backup path to /.snapshot/Weekly0? Simon > -----Original Message----- > From: Traitel, Eyal [mailto:eyal@netapp.com] > Sent: Wednesday April 2001 16:29 > To: 'E.J. Jantz'; toasters(a)mathworks.com; simon_clawson > Subject: RE: snapshot backup > > > Our snapshots are to be more exact "copy on new write" as > opposed to "copy on old write", with the advantage of no > penalty at all in terms of performance of taking the > snapshot, and performance of filesystem after taking it > (think of "copy on old write" - filesystem needs to read old > block, write it in the backup location, then write the new > block in old place - that's 3 I/Os. For us it's only 1). > We do not have to go back to specific locations on the disk, > and that's a real advantage in terms of performance. > > In regards to full compared to snapshot - with NetApp, it's > probably always better to backup snapshots, as they are > point-in-time - when you've started your backup at 19:00, all > of it will be consistent to 19:00 sharp, unlike traditional > backup, which data changes while you backup... So - if you > get me right - you can even do full backup from a snapshot... > > On Legato - I think you can just mention .snapshot/hourly.0 > for example in your path of backup. If you use NDMP, you > anyway backup a snapshot, as the filer's dump creates one > automatically for you. > > Hope it somewhat helps, > Eyal. > > > ---------------------------------------------------------------------- > eTraitel - I'm the new eBuzzword around !!! > ---------------------------------------------------------------------- > Eyal Traitel - Filer Escalation Engineer > CNA, MCSE, CSA, LCA, NetApp CA > > Network Appliance BV > Holland Office Center > Kruisweg 799b > 2132 NG, Hoofddorp > The Netherlands > Office: +31 23 567 9685 > Cellular: +31 6 5497 2568 > Email: eyal(a)netapp.com > ---------------------------------------------------------------------- > Get answers NOW! - NetApp On the Web - http://now.netapp.com > ---------------------------------------------------------------------- > > > -----Original Message----- > From: E.J. Jantz [mailto:ej_jantz@bswintl.com] > Sent: Wednesday, April 25, 2001 3:55 PM > To: toasters(a)mathworks.com; simon_clawson(a)mentorg.com > Subject: Re: snapshot backup > > > just a question: If the 'snapshots are "copy on writes' doesn't that > mean that a backup of .snapshot is equal to and functions as a full > backup? If not, can Legato using a NFS mount on a unix server capture > just the info the .snapshot uses? If so, how? > > thanks... > > EJ Jantz, IST Dept. > BSW International > One West Third, Ste 100 > Tulsa, OK 74103-3505 > 918 582-8771 > ej_jantz(a)bswintl.com > 918 295-4166 > 918 587-3594 fax > > > >>> "Clawson, Simon" <simon_clawson(a)mentorg.com> 4/25/01 4:32:08 AM > >>> > Is it possible to backup just the .snapshot and then perform a restore > from > this data? > > I am looking for ways to speed up the backups and this seems to be a > possible option. It was suggested to me by a colleague as a method of > backing up Clearcase VOBS, but would it be viable for the whole filer. > Of > course I would need to run a full backup at tleast once a month, but > this > would help me speed up most of my weekly runs. > > Any ideas/suggestions/views? > > Simon > > Simon Clawson > HDS Team Systems Administrator > Mentor Graphics Uk > Rivergate > London Road > Newbury > Berkshire > RG14 2QB > <<Clawson, Simon.vcf>> >

2 1

fibre channel hubs
by neil lehrer 25 Apr '01

25 Apr '01

has netapp qualified any of the vixel hubs for tape backup? the hubs are much cheaper than the vixel 8100 and i don't think the lesser performance will be an issue in my case - two filers. thanks

1 0

RE: "Case survey" stuff from NetApp
by Maglinger, Paul 25 Apr '01

25 Apr '01

I reply to them. If they don't know what's going on, how can they fix it. Anyone filling it out should be just as quick to praise their performance as they are to condemn it. I have only put in one "bad" report having to do with response time. That prompted a phone call from them and they seemed truly concerned. All in all, I've been pleased with the service (except for not getting a leather jacket.... oops, there goes that inane banter again). -----Original Message----- From: Chris Thompson [mailto:cet1@cus.cam.ac.uk] Sent: Wednesday, April 25, 2001 14:04 To: toasters(a)mathworks.com Subject: "Case survey" stuff from NetApp What do you do with the the e-mail that arrives from Quality.Manager(a)netapp.com after a problem is closed with NetApp, inviting you to fill in a "5 STARS Survey" form? I admit I throw them in the e-wastebin. It seems to me that providing only a method of returning these (probably only marginally meaningful) answers by e-mail is absolutely pessimal. If they really want to increase the response rate, surely NetApp should provide a method using forms at now.netapp.com? My vague recollection is that at one time a WWW-browser-based method was offered, but this doesn't seem to be the case now. Chris Thompson University of Cambridge Computing Service, Email: cet1(a)ucs.cam.ac.uk New Museums Site, Cambridge CB2 3QG, Phone: +44 1223 334715 United Kingdom.

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