toasters

toasters@lists.teaparty.net

13531 discussions

RE: EMC Celerra vs NetApp Filer
by Sam Schorr 10 Apr '00

10 Apr '00

Homestead.com has 8 F760 filers. We have just installed a Celerra with 7 datamovers attached to a Symmetrix 3930 (3 raw terabytes, 1 TB effective). While both are NAS solutions, the approaches are very different. How they are evaluated must be done from the perspective of the specific environment they are used in. In ours, both perform the same duty - we have roughly 70 HTTP servers (MS IIS 4.0) on Compaq Proliant servers (NT 4.0 SP6a), a few Linux and FreeBSD servers and some non-HTTP servers using either CIFS or NFS. The mix is about 65-70% CIFS and the rest NFS. Each "client" server uses UNC pathing to find the specific volume\directory\file, or to create the same. Performance - too early to compare, but initial results show parity of response. I have no hard data other than the individual read/write tests we have done that show little significant variance. Administration - big difference. The netapp is very easy to setup and install and monitor. the combined Celerra/Symmetrix environment is not. Let's just say that I am not the most technical guy in the world and I setup 2 F760's, cross-connected for Snapmirror, with all the RAID definitions, etc., by myself in one afternoon. I would not even know where to begin doing this in the EMC environment. On the other had, for the price difference, I expect EMC to get that done and get it done correctly. Support - here again the 2 approaches differ greatly. The EMC platform comes with premium service - you get no choice, you only get top-of-line, phone-home, on site in 2 hours. netapp, even though we have premium support, does not have the same level of system-to-support automatic ties (autosupport really is not the same). being located very close to netapp has mitigated any issue this might cause and we have nothing but great support from them. Why did Homestead.com choose to buy EMC at this point? The main reason is that Network Appliance's next generation is not yet ready, so the ability to scale based on current products was not optimal for us. Also, we wanted a RAID 1+ solution - WAFL is RAID 4 (and very good, just not as tolerant of disk failures). Lastly, I could use parts of the Symmetrix for direct attach (SAN) while having the other for NAS. Sam Schorr Homestead.com ph: (650) 549-3152 fax: (650) 364-7329 sschorr(a)homestead-inc.com Homestead: It's a new world. Make your place. -----Original Message----- From: Gilles TOURPE [mailto:gilles@bnpcn.com] Sent: Monday, April 03, 2000 5:28 AM To: toasters@mathworks. com Subject: EMC Celerra vs NetApp Filer Hello, Did anybody get in the same site, an EMC Celerra system and a NETAPP F7xx ( at least 740 ) Filer ? I would be interrested in Performances, administration, supports issues ... I'm very curious about the Celerra solution. On the paper it looks smart and really really NAS-oriented ( quite funny for SAN-guys :-) ). Some tests we did we pretty concluant also ( and complex also ... ). Any thinking, remarks ???? Thanx Gilles ------------------------------------------------------------ Gilles TOURPE - System and Network Administrator APS BNP ARBITRAGE - Salle des marches ACTIONS - PARIS - France Phone : (33)-1-40-14-24-03 e-mail : gilles(a)bnpcn.com

12 27

Huge one-time data transfer from UNIX to NetApp
by Rainchik, Aleksandr (MED, Non GE) 10 Apr '00

10 Apr '00

Hi! I have a question: what should be the best way to transfer 20-25Gb of data (lot's of small files) from UNIX to NetApp? Can I do ufsdump on UNIX, pipe it trough rsh to NetApp and do restore there? Thank you.

10 13

RE: Huge one-time data transfer from UNIX to NetApp
by Mohler, Jeff 07 Apr '00

07 Apr '00

Well, depends. Somewhere around 2.5 - 3M files in a copy..nmdpcopy blows up. :-----Original Message----- :From: Mark D Fowle [mailto:Fowle_Mark_D@CAT.com] :Sent: Friday, April 07, 2000 5:00 AM :To: Aleksandr.Rainchik :Cc: toasters :Subject: Re: Huge one-time data transfer from UNIX to NetApp : : :You could try using NDMPCopy. : :Mark Fowle :Caterpillar/BCP : : : : : :Aleksandr.Rainchik@amermsx.med.ge.com on 04/07/2000 07:53:00 AM : :To: toasters@mathworks.com@INTERNET :cc: (bcc: Mark D. Fowle/0V/Caterpillar) :Subject: Huge one-time data transfer from UNIX to NetApp :Retain Until: 05/07/2000 Retention Category: G90 - :Information and : Reports : :Hi! : :I have a question: what should be the best way to transfer 20-25Gb of :data :(lot's of small files) from UNIX to NetApp? : :Can I do ufsdump on UNIX, pipe it trough rsh to NetApp and do restore :there? : :Thank you. : : :

1 0

Re: Linux & NetApp
by Mohan Srinivasan 07 Apr '00

07 Apr '00

You might want to pick up the latest Linux NFS client patches. I had problems (with both stability and performance) with the NFS client support that came bundled in with 2.2.12, but both stability and performance have improved greatly after picking up the patches. The patches include NFSv3 support as well. The latest patches only apply to 2.2.14. You'll have to do some significant merging if you want to stick with older releases. Look in nfs.sourceforge.net, or talk to the VA Linux folks. We've been using NFSv3/UDP+32K xfer sizes on a farm of Linux 2.2.13 clients with several F740's for a couple of months now with no problems. mohan --- Fabio Pietrosanti <fabio(a)telemail.it> wrote: > Hi, i've a Server farms of VaLinux based on kernel 2.2.7-1.23 that use a > F740 NetApp . > Now, NetApp's engineer tell us that linux it's not the better choice for > using NetApp . > We mount share with this parameter : > nfs rsize=8192,wsize=8192,timeo=14,intr,soft > > sometimes( every 2 or 3 week), on some server(different every time) > the NetApp does not respond to server, and i see in the log : > kernel: nfs: server netapp-2-1 not responding, timed out > The load average remain very high until i don't reboot the machine. > > does someone had the same problem? > > p.s.: the logs on the NetApp are normal . > > > > > __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com

1 0

Re: Huge one-time data transfer from UNIX to NetApp
by Mark D Fowle 07 Apr '00

07 Apr '00

You could try using NDMPCopy. Mark Fowle Caterpillar/BCP Aleksandr.Rainchik(a)amermsx.med.ge.com on 04/07/2000 07:53:00 AM To: toasters@mathworks.com@INTERNET cc: (bcc: Mark D. Fowle/0V/Caterpillar) Subject: Huge one-time data transfer from UNIX to NetApp Retain Until: 05/07/2000 Retention Category: G90 - Information and Reports Hi! I have a question: what should be the best way to transfer 20-25Gb of data (lot's of small files) from UNIX to NetApp? Can I do ufsdump on UNIX, pipe it trough rsh to NetApp and do restore there? Thank you.

1 0

RE: ADIC FastStor on F720
by Bond, Andrew 07 Apr '00

07 Apr '00

How was the cloned_tapes file created? You should really let the filer create it for you. If you just edited it yourself, then try putting a comment line at the top - I haven't tested this but it may work - the format of the automatically-created file has commented-out first few lines. Otherwise, delete the file and reboot (there will be an error message at boot telling you the tape drive is unsupported), then follow this procedure: Do 'mt -f rst0a status' - this will create the file /etc/cloned_tapes Edit the cloned_tapes file to map the product ID, ie. add a line: DLT8000 EMULATES DLT7000 Do 'mt -f rst0a status' again. Now 'sysconfig -t' should list the appropriate device files. Note that there will still be an error message about an unsupported tape drive at boot, but it will all continue to work. You might try omitting the reboot - I'm not sure if that will work, but if you follow the procedure above you should be fine. There is a TIP on NOW but I thought sending you the URL would be a bit mean since you said you haven't got a working account presently! Regards, Andrew > -----Original Message----- > From: owner-dl-toasters(a)netapp.com > [mailto:owner-dl-toasters@netapp.com]On Behalf Of bjf > Sent: 04 April 2000 22:49 > To: toasters(a)mathworks.com > Cc: wild(a)rebel.com > Subject: ADIC FastStor on F720 > > > Howdy, > > At my day job, I've got a F720 running 5.3.4R2. I'm trying > to attach a new > ADIC FastStor 7 tape library to it. > > The filer sees the changer and the drive: > > slot 0: SCSI Host Adapter 0b (Qlogic ISP 1040B) > Fast/Wide, Differential > Firmware Version 4.65 Clock Rate 60MHz. > 1: ADIC FastStor DLT 0115 > 3: QUANTUM DLT8000 010F > > But it won't qualify the drive, even if I try using the > /etc/cloned_tapes > file. > > disk0> sysconfig -t > > Tape drive (0b.3) DLT8000 (Non-qualified tape drive) > > /etc/cloned_tapes: > DLT8000 EMULATES DLT7000 > > (also tried QUANTUM DLT8000 EMULATES QUANTUM DLT7000). > > Any ideas? Hopefully my registration on now.netapp.com will come thru > sometime soon... > > Thanks, > > Bryan > > -- > Bryan Fullerton http://bryanfullerton.com/ > Core Competency > Samurai Consulting > Can you feel the Ohmu call? >

6 5

回复: FreeBSD client NFS problems
by William ding 07 Apr '00

07 Apr '00

Hi Mike : We use NFS V3 UDP and set NMBCLUSTER=40960 at kernel config , but always shutdown,could you give us some advice ? Thanks a lot -----Original Message----- 发件人: Mike Horwath <drechsau(a)tiny.net> 收件人: William ding <ding(a)staff.163.com> 抄送: toasters(a)mathworks.com <toasters(a)mathworks.com> 日期: 2000年4月6日 7:14 主题: Re: FreeBSD client NFS problems >On Thu, Apr 06, 2000 at 07:45:53PM -0700, William ding wrote: >> I have two FreeBSD clients NFS mounting a volume from a Filer. >> The versions are 3.4 and 4.0. >> >> When NFS is being used the memory usage of both machines >> climbs steadily and a reboot occurs (about 2-3 times per >> day). >> >> Is anyone aware of this problem and know of any patches >> or new versions that may fix this problem ? > >Are you mounting V2 or V3 and over UDP or TCP? > >I bet things would be dandy at V2 UDP. > >-- >Mike Horwath Admin & Manager @ VISI.com WORK: drechsau(a)visi.com >IRC: Drechsau http://www.visi.com/ HOME: drechsau(a)geeks.org >The only Minnesota ISP with public statistics: http://noc.visi.com/ > Garbage In -- Gospel Out. - berkeley fortune(6) >

1 0

Re: snapshots dates
by Fred Ab 06 Apr '00

06 Apr '00

Looks fine on my machine (Ultra5) and Netapp is a n old 520 and old version of OS: NetApp Release 5.1.2R3: Wed Nov 25 17:20:58 PST 1998 ultrax05--ab[52] pwd /user/abc/.snapshot ultrax05--ab[53] ls -l total 520 drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.0/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.1/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.2/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.3/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.4/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.5/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.6/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 hourly.7/ drwxrwxrwx 101 abc nobody 20480 Apr 5 18:33 nightly.0/ drwxrwxrwx 101 abc nobody 20480 Apr 4 14:47 nightly.1/ drwxrwxrwx 100 abc nobody 20480 Apr 3 16:26 nightly.2/ drwxrwxrwx 100 abc nobody 20480 Mar 31 16:19 weekly.0/ drwxrwxrwx 100 abc nobody 20480 Mar 22 12:30 weekly.1/ ===}Date: Thu, 06 Apr 2000 14:33:45 -0700 ===}From: Frank Gilmurray <frank(a)erg.sri.com> ===}X-Accept-Language: en ===}Mime-Version: 1.0 ===}To: toasters(a)mathworks.com ===}Cc: bryan.anderson(a)netapp.com ===}Subject: snapshots dates ===}Content-Transfer-Encoding: 7bit ===} ===}>From a Solaris 7 system we're seeing the same dates on all our snapshots ===}(but 'snap list' on the filer gives the correct dates). The filer date ===}is correct. Has anyone else seen this? Is there a fix or workaround? ===} ===}Thanks much, Frank ===} ===}PS Is there an archive of the toaster messages? ===} ===} ===}quetzalcoatl:/vol1# ls -l .snapshot ===}total 88 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 hourly.0 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 hourly.1 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 hourly.2 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 hourly.3 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 hourly.4 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 hourly.5 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 nightly.0 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 nightly.1 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 nightly.2 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 weekly.0 ===}drwxr-xr-x 4 root root 4096 Mar 22 11:12 weekly.1 ===}quetzalcoatl:/vol1#

1 0

RE: Legato NetApp client
by Hackett, Roger 06 Apr '00

06 Apr '00

Thanks for the reply Arthur. The NetApp client is a very different beast from any other Legato client I've worked with. Though the Legato ClientPak for NetApp README references an na_install executable that is supposed to install nsrexecd on the NetApp, there is no such process running on the NetApp (determined by issuing rc_toggle_basic followed by ps on the filer console while a backup was in progress). All that seems to be running on the filer is a java thread(s) related to the backup. This makes sense as the ClientPak documentation also says the it interfaces to java engine that runs on the filer. The na_recover command actually executes on the administrative host (apparently communicating to the java engine on the filer) rather than the filer itself. There is no interactive mode available. later... Roger -----Original Message----- From: Arthur Darren Dunham [mailto:add@netcom.com] Sent: Thursday, April 06, 2000 11:42 AM To: Roger.Hackett(a)mms.gov Subject: Re: Legato NetApp client > > We have a NetApp F760 running ONTAP 5.3.5 and the Legato NetWorker ClientPak > for NetApp, release 5.5 that we are configuring for deployment. In my > initial tests I have been able to setup the NetApp as a Legato client, > schedule backups from our Legato server, do manual backups from the admin > host's mount point of specific NetApp volumes and restore entire save sets. > > My problem/question concerns individual file restores. How does one > determine if a specific file is in a specific save set? The Legato NetApp > client docs state the you can use the mminfo command on the Legato host and > specify file names (mminfo -c filer_name -N /vol/volx/file_name) but both > empirical tests and the mminfo manpage show that the -N option is for the > save set name (such as /vol/vol0 or /vol/vol1) and no individual file > information is returned. The results are essentially the same as using the > Indexes button of nwadmin, you get a list of save set names and save set > ids. Correct. mminfo peruses the media database (not the file database) and has no information about individual files. I have been able to restore a single file (na_recover -s nsrhost -S > ssid -a /vol/vol0/qtree_name/filename) by listing each different ssid until > the request succeeds. Hm.. I don't have the netapp client, so I'm not sure how this differs... The normal unix client is interactive if you don't specify a file. After cd'ing into a directory, you can 'versions filename'. It will display the differnet versions of that file that are on different tapes. You could also do 'recover -s nsrhost -t <date>' and get back the last version of the file that was prior to that date. -- Darren Dunham ddunham(a)taos.com Unix System Administrator Taos - The SysAdmin Company Got some Dr Pepper? Santa Clara, CA < Please move on, ...nothing to see here, please disperse >

1 0

回复: FreeBSD client NFS problems
by William ding 06 Apr '00

06 Apr '00

We use "nfsiod -n 20" when starup... but after 5 hours, it will crash! -----Original Message----- 发件人: Nick Hilliard <nick(a)iol.ie> 收件人: ding(a)staff.163.com <ding(a)staff.163.com> 抄送: toasters(a)mathworks.com <toasters(a)mathworks.com> 日期: 2000年4月6日 7:01 主题: Re: FreeBSD client NFS problems >> I have two FreeBSD clients NFS mounting a volume from a Filer. >> The versions are 3.4 and 4.0. >> >> When NFS is being used the memory usage of both machines >> climbs steadily and a reboot occurs (about 2-3 times per >> day). > >Sounds odd. Are you sure that this has got to do with NFS usage? I >haven't used 4.0 in production, but 3.x was pretty solid, and the only times >that these boxes are rebooted is for kernel upgrades. > >How do you define "out of memory", and what sort of things are these boxes >doing? > >Nick >

2 1

improving filer performance
by Todd C. Merrill 06 Apr '00

06 Apr '00

I am trying to (what else?) improve the performance of the filers that I have. Here are the configs and the environment in which they run. My specific questions follow: I have two F740's with 14x 18GB FC-AL disks and a trunked 4x 100 Mb/s ethernet card. These filers serve software development sandboxes with a few million files (median file size about 4 kB), not more than a few thousand in the largest directories. These filers are used for both CIFS and NFS client builds; NFS traffic is limited to UDP only (`options nfs.tcp.enable off`) to keep the network overhead down on that protocol. Mix is about 60-75% NFS; the rest, CIFS. Running ONTAP 5.3.4R3P2. CPU load is more often than not above 50%, often 75-90% or more; usually 3,000 - 5,000 ops/; cache age is rarely above 3--most often 1 or 0. Consistency points are usually at the maximum of once every 10 seconds (as measured by disk writes with `sysstat`). I was poking around on the SPEC pages, and in the footnotes to the NetApp entries are these parameters: options nosnap 1 # to disable periodic snapshot creation, for reproducibility options nosnapdir 1 # to avoid inserting .snapshot entry when reading directories options raid.scrub.enable off # to disable periodic RAID scrubs, for reproducibility options minra 1 # to minimize file read-ahead options udp_lg_dgram.xmit_cksum.offload 1 # to offload checksum computations onto the Gigabit NIC Openboot settings on the F740: setenv java? false # to disable the Java Virtual Machine http://www.spec.org/osg/sfs97/results/res98q3/sfs97-980805-00008.html http://www.spec.org/osg/sfs97/results/res99q2/sfs97-19990416-00045.html I have a few questions: 1a. Regarding snapshots, many of my development sandboxes are "disposable" so I don't care about snapshots. The first two options might help me. In addition, if I set `snap reserve volX 0`, will that actually turn off snapshots, and reduce overhead on the filer? The options only seem to turn off *automatic* snapshots and the *display* of the ".snapshot" directory. 1b. Would disabling these two options *prevent* me from `cd`ing into the .snapshot directory? Even though it is visible by default at the root of the mount point, you can `cd` into the directory ".snapshot" at any point: 37 machine:/devel/scratch>ls -a . .. .snapshot toddc 38 machine:/devel/scratch>cd toddc 39 machine:/devel/scratch/toddc>ls -a . Testing.doc devtools .. dump.image.gz 40 machine:/devel/scratch/toddc>cd .snapshot 41 machine:/devel/scratch/toddc/.snapshot>pwd /devel/scratch/toddc/.snapshot 42 machine:/devel/scratch/toddc/.snapshot>ls backup hourly.1 nightly.0 hourly.0 hourly.2 1c. In addition, I'm thinking of breaking up my 14-disk filer from a RAID 13d+1p configuration into, say, two RAID groups and two volumes as 3d+1p and 9d+1p. (Multiple RAID groups are allowed per volume; I assume multiple volumes per RAID group is still disallowed?) The former would have the stuff I need snapshotted and the latter the stuff I don't. I have a churn rate of about 30-40%. Thoughts? 2. What does "options minra 1" buy me? What is the number when minra=off, the default? (If it's just a boolean, nevermind!) Given I have a development environment with lots of small files, it seems as if turning this option on will benefit me a lot. However, some more information on this would help. 3. Does the udp_lg_dgram.xmit_cksum.offload option also apply to quad-ethernet cards? Will it help if the ports are trunked with a virtual interface? 4. Why is my filer running Java? If I disable that, would we see any administration impact? (These filers are administered solely via the CLI.) I do see the messages file logs [Java Thread]: TimeDaemon: timed: adjusting time messages all the time. Would I loose the ability to keep the filer in time sync? What else might I loose? To prevent a knee-jerk reaction, for the purposes of this discussion, upgrading my filer to an F760 is not an option. ;) I want to leverage what I have now, as best I can. I thought about opening a tech support call, and still may, but I thought some real life input from the trenches would be useful as a start. Until next time... The Mathworks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

4 6

Legato NetApp client
by Hackett, Roger 06 Apr '00

06 Apr '00

We have a NetApp F760 running ONTAP 5.3.5 and the Legato NetWorker ClientPak for NetApp, release 5.5 that we are configuring for deployment. In my initial tests I have been able to setup the NetApp as a Legato client, schedule backups from our Legato server, do manual backups from the admin host's mount point of specific NetApp volumes and restore entire save sets. My problem/question concerns individual file restores. How does one determine if a specific file is in a specific save set? The Legato NetApp client docs state the you can use the mminfo command on the Legato host and specify file names (mminfo -c filer_name -N /vol/volx/file_name) but both empirical tests and the mminfo manpage show that the -N option is for the save set name (such as /vol/vol0 or /vol/vol1) and no individual file information is returned. The results are essentially the same as using the Indexes button of nwadmin, you get a list of save set names and save set ids. I have been able to restore a single file (na_recover -s nsrhost -S ssid -a /vol/vol0/qtree_name/filename) by listing each different ssid until the request succeeds. It looks like we will depend on snapshots for individual file restores and treat the Legato backups as "disaster recovery" only. -- Roger Hackett ACS-GSG Minerals Management Service - Gulf of Mexico Region U.S. Department of the Interior roger.hackett(a)mms.gov

1 0

FreeBSD client NFS problems
by William ding 06 Apr '00

06 Apr '00

I have two FreeBSD clients NFS mounting a volume from a Filer. The versions are 3.4 and 4.0. When NFS is being used the memory usage of both machines climbs steadily and a reboot occurs (about 2-3 times per day). Is anyone aware of this problem and know of any patches or new versions that may fix this problem ? William ding

4 3

RE: Can VIFs work with 3Com switches?
by Bond, Andrew 06 Apr '00

06 Apr '00

Hi Chris, Yes, the 3300 works with the filer. 3300 firmware must be rev 2.0 or above. Regards, Andrew > -----Original Message----- > From: owner-dl-toasters(a)netapp.com > [mailto:owner-dl-toasters@netapp.com]On Behalf Of Chris Thompson > Sent: 05 April 2000 21:18 > To: toasters(a)mathworks.com > Subject: Can VIFs work with 3Com switches? > > > This is thinking a few months into the future for us, but it might be > useful to know the answer now... > > Can one use VIFs in ONTAP together with the port trunking > facilities of > a 3Com SuperStack 3300 switch to treat two or more 100BaseT > connections > as a single interface, with load balancing at both ends? Has > anyone any > experience with such a configuration? > > 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

Can VIFs work with 3Com switches?
by Chris Thompson 06 Apr '00

06 Apr '00

This is thinking a few months into the future for us, but it might be useful to know the answer now... Can one use VIFs in ONTAP together with the port trunking facilities of a 3Com SuperStack 3300 switch to treat two or more 100BaseT connections as a single interface, with load balancing at both ends? Has anyone any experience with such a configuration? 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: EMC Celerra vs NetApp Filer
by Sam Schorr 04 Apr '00

04 Apr '00

Interesting. I got a different answer from my EMC engineering people. The following is a quote: "Stripes can exist either within the Sym ("meta-volume") or outside the Sym using OS (or volume management) tools. Within the Sym, if a stripe is built across Mirrored volumes, the stripe remains available provided that one does not lose both halves of the same mirror. Example, four mirrored pairs (A/a, B/b, C/c, and D/d) with a stripe built across the four. Up to four disks could be lost as long as one of each pair is still good. We will not build a meta-volume such that 2 or more parts reside on the same physical - this violates our configuration rules. Outside the Sym the same holds true provided that one does not build stripes with multiple pieces coming from the same physical. The net is - as long as the underlying parity (mirror or otherwise) maintains the availability of the parts of a stripe, the whole stripe will remain available." --srs -----Original Message----- From: Steve Gremban [mailto:gremban@msp.sc.ti.com] Sent: Tuesday, April 04, 2000 11:04 AM To: toasters@mathworks. com Subject: Re: EMC Celerra vs NetApp Filer On an EMC if a disk goes out on one side of a mirror (we'll call it side A), side A will be offline (all of the disks in side A, not just the bad drive). Therefore, any new writes will be seen on side B but not on the drives of side A and you can't use side A drives to recover from subsequent side B drive failures. My EMC SE said that only if both sides of the mirror have simultaneous drive failures taking them both down at the same time is there a chance of recovering and that it would take a lot of work. (this assumes that the bad drives weren't mirrors of each other) Brian Tao wrote: > On Mon, 3 Apr 2000, Bruce Sterling Woodcock wrote: > > > > I don't see how? When one mirror loses a disk, the whole thing is > > lost; you switch over to the other mirror, which has n-1 disks as > > the RAID 4 array. So your initial chance of failure is 2n-1 and > > then n-1 for a second failure whereas RAID 4 is just n followed by > > n-1 for the second failure To be somewhat anal about it, the initial chance of failure of mirrored partitions is 2(n-1). :) But you're right, you have a higher probability of being put at risk using a mirror (just needs 1 disk going bad out of 2n-2 disks) than with RAID4 (1 disk bad out of n disks). Once that 1st disk is lost the probabilities of a second failure are the same for both (n-1). If EMC had actually implemented their mirroring like Brian Tao mentions below then their mirroring would be much more reliable than RAID4. > > > I think you're talking about RAID 0+1 (taking a RAID-0 set and > mirroring it on another RAID-0 set, which is just silly). RAID 1+0 > does the mirroring first, then the striping/concatenation: > > +----+----+ +--------------+ +------------+ > | 1A | 1B | | RAID-1 set 1 | | | > +----+----+ +--------------+ | | > | 2A | 2B | | RAID-1 set 2 | | | > +----+----+ +--------------+ | | > | 3A | 3B | | RAID-1 set 3 | | | > +----+----+ +--------------+ | | > | 4A | 4B | --> | RAID-1 set 4 | --> | RAID 0 | > +----+----+ +--------------+ |(7 "drives")| > | 5A | 5B | | RAID-1 set 5 | | | > +----+----+ +--------------+ | | > | 6A | 6B | | RAID-1 set 6 | | | > +----+----+ +--------------+ | | > | 7A | 7B | | RAID-1 set 7 | | | > +----+----+ +--------------+ +------------+ > [etc...] > > The A's and B's are drives of a mirrored pair. You could lose, > say, drives 1A, 2A, 4B, 5B and 7A and still have a functional RAID 0, > because no single mirror is completely broken. You could lose the > entire shelf containing the A drives and not have an outage (something > from which today's Netapps cannot automatically recover, clustering or > no clustering). Having mirrored pairs in RAID 4 (i.e., having > mirrored data and parity drives) and the ability to concurrently > rebuild multiple broken mirrors to hot spares would really give Netapp > a leg up on other NAS vendors, IMHO. > > > Oh, are the drives mirror images of each-other too? I didn't > > realize that. So if Drive 1 in Mirror A fails, breaking the mirror, > > and then Drive 2 in Mirror B fails, Mirror B will be smart enough to > > switch over to Drive 2 in Mirror A? > > As far as the RAID 0 is concerned, single drive failures within > each mirrored pair does not result in that stripe being down, because > the remaining half of the mirror is still online. If you lose, say, > drives 4A and 4B in a RAID 1+0, then you're toast. In RAID 1+4 (or > 1+5), you would still have the parity drive. > > > -- > Brian Tao (BT300, taob(a)risc.org) > "Though this be madness, yet there is method in't"

1 0

ADIC FastStor on F720
by bjf 04 Apr '00

04 Apr '00

Howdy, At my day job, I've got a F720 running 5.3.4R2. I'm trying to attach a new ADIC FastStor 7 tape library to it. The filer sees the changer and the drive: slot 0: SCSI Host Adapter 0b (Qlogic ISP 1040B) Fast/Wide, Differential Firmware Version 4.65 Clock Rate 60MHz. 1: ADIC FastStor DLT 0115 3: QUANTUM DLT8000 010F But it won't qualify the drive, even if I try using the /etc/cloned_tapes file. disk0> sysconfig -t Tape drive (0b.3) DLT8000 (Non-qualified tape drive) /etc/cloned_tapes: DLT8000 EMULATES DLT7000 (also tried QUANTUM DLT8000 EMULATES QUANTUM DLT7000). Any ideas? Hopefully my registration on now.netapp.com will come thru sometime soon... Thanks, Bryan -- Bryan Fullerton http://bryanfullerton.com/ Core Competency Samurai Consulting Can you feel the Ohmu call?

1 0

Summary: backup recommendations
by Jeff Bryer 04 Apr '00

04 Apr '00

I had asked about software/hardware recommendations to replace our existing Budtool/DLT4000 equipment. This is for backing up NetApp filers (NDMP support a requirement), SUN servers, and a Linux cluster primarily. Ability to scale to include Netware, and NT is also important, as well as the potential to scale to cover campus-wide backups of Win9x/NT/2000, Mac and other clients. Summary of responses: Software: 1) Quick Restore from Workstation Solutions was recommended as the choice most closely resembling Budtool (licensing model, look and feel) and was thought of as having the best NDMP support today. 2) Networker from Legato, good client support but the NDMP support came last (out of the three choices). It was thought that v6 will have much better NDMP support. Plus the NDMP client pack is expected to be a free upgrade to Budtool owners (or maybe restricted to those on maintenance). 3) Netbackup from Veritas. NDMP support rated number 2, good client support. Quite a few people who had used both Legato and Veritas said they had migrated to Veritas (NDMP support being one reason). Hardware: 1) DLT, by far most people recommended using DLT[78]000s. It's established, people trust the reliability. 2) AIT, on paper specs great. Of those who've used it, some have had good results with direct attach to NetApps, others only once the library was configured in DLT emulation mode. The DLT drives may have had more recommendations but that's very much skewed as most people who responded hadn't used AIT drives. I would like to hear from more people who've actually used AIT drivs. Our current thoughts: Software choice: Veritas. The NDMP support today in Legato is not sufficient for our needs. Quick Restore doesn't have support for enough different clients, a smaller user install base, disaster recovery seemed to be more a focus for the upcoming v3 release. Hardware choice: AIT. We've had to replace our DLT7000 drive three times in the last two years, so we don't feel overlly confident about them (our 4000 stackers have been fine). The specs and road map for AIT look good (on paper). This being said, I still would like to hear from AIT users to feel better about the reliability of the media/drives. Thanks to everyone who responded. -- Jeff Bryer bryer(a)sfu.ca Systems Administrator (604) 291-4935 Academic Computing, Simon Fraser University

1 0

Budtool, NetApp filer & Tape library
by Rainchik, Aleksandr (MED, Non GE) 04 Apr '00

04 Apr '00

Hello, We have NetApp F740. For backups we are using Budtool 4.6.1 with two media servers (Sun's) and tape library connected to them. Right now, when we are doing backups, all data bits are flying over the network from the NetApp to the media server and finally to the tape. I'm curious, if we can connect one SCSI channel from the tape library to the NetApp? And a) use it as NDMP media server b) eliminate data transfer over the network while doing NetApp backups. Any hidden traps? So, we are basically want to replace one media server with a NetApp. And my main concerns is how well NetApp is going to work this way. Can I use all those jbmgr commands to load and unload tapes there? Will index tables be updated correctly on main Budtool server? Thank you.

2 1

Re: Messages from cron_admin about daylight saving time
by Chris Thompson 04 Apr '00

04 Apr '00

Back on Tuesday 21 March I wrote: > > We run a default raid scrub regime (01:00 on Sundays). Last Sunday we > got a message > > Sun Mar 19 01:00:00 GMT [cron_admin]: > Rescheduling raid_scrub_admin to run at Sun Mar 19 01:00:00 GMT 2000 > because of Day Light Saving time change > > [Raid scrubbing itself ran at this time with no problem.] > > Moreover, every time timed tweaks the clock (we use ntp synchronisation) > it again mutters. e.g. > > Tue Mar 21 16:04:33 GMT [cron_admin]: > Rescheduling raid_scrub_admin to run at Tue Mar 21 16:04:33 GMT 2000 > because of Day Light Saving time change > > Now all this is no doubt because next Sunday is clock-changing day > in our timezone (GB), with 2000-03-26 01:00:00 GMT -> 02:00:00 BST. > [American friends will no doubt be seeing this effect next week!] Well, they obviously didn't, and I think I understand why. Contrary to my initial impression, it seems that the USA switches at 02:00:00->03:00:00 (local time), rather than 01:00:00->02:00:00. The messages apparently only happen if the (local) time that the raid scrub is scheduled for does not logically exist. This happened in the UK and in the Republic of Ireland (I got a confirmation from Garret Flynn) and presumably in Portugal (any NetApp users there?). Further east in Europe the clock change is later in local time (because it is at the same real time, 01:00:00 UTC). For those interested: the raid scrub *did* run at the expected time on clock-changing night, and the messages then disappeared. Garret tells me that this is a known bug, 13649. 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

RE:
by Hawley, Rob 04 Apr '00

04 Apr '00

The filer only supports Microsoft NT domain controllers. There are a number of issues not just that samba does not support UNICODE in the same way NT does. -----Original Message----- From: Igor Schein [mailto:igor@txc.com] Sent: Tuesday, April 04, 2000 10:30 AM To: toasters(a)mathworks.com Subject: Hi, My goal is to to get CIFS on a 740 filer to join to PDC running on Samba. Last time I checked there was a unicode issue which was preventing people from using Samba as PDC for netapp filers. I was wondering if the status has changed and any1 on the list has been successfull in doing so. THanks Igor

1 0

None
by Igor Schein 04 Apr '00

04 Apr '00

Hi, My goal is to to get CIFS on a 740 filer to join to PDC running on Samba. Last time I checked there was a unicode issue which was preventing people from using Samba as PDC for netapp filers. I was wondering if the status has changed and any1 on the list has been successfull in doing so. THanks Igor

1 0

RE: How are you mounting CIFS homedirs with Win9x?
by Uren, Jim 04 Apr '00

04 Apr '00

[snip] > > Thanks for the info. Are there any plans to support variable > expansion a la > SAMBA? For example, sample allows you to define a share using > "%u", which > gets expanded to the username on the SAMBA side. That llows you to put > "\\samba-server\\home" in the logon script, and have SAMBA do the > exapansion for you. > > Tim Yes, this capability is in-plan. The bug report (burt) number for this RFE is 13856.

1 0

Re: EMC Celerra vs NetApp Filer
by Todd C. Merrill 04 Apr '00

04 Apr '00

Kendall Libby wrote: > > right now. Going by units of "1 cpu, 1gb cache, 200gb disk", EMC > > comes in cheaper after a small number of units. For us. YMWV. Bruce Sterling Woodcock wrote: > But these wouldn't be units that would be equal comparison. > What matters is primarily ops/$ and GB/$ for roughly the same Kendall indeed pointed out one "unit" measure, combining performance and storage, that works well *for us*. ops/$ and GB/$ may not be a primary concern for everyone [0], nor may it be a majority concern [1]; I don't think it's fair to make such a blanket statement. There are subjective factors, of course. And, if you ask NetApp for their super-duper maintenance and support plans with bacon and extra cheese, in order to try to narrow the service level gap and compare apples-to-apples with what EMC offers by default, the price gap narrows quite a bit, too. Bruce wrote: > not doubting what you say; I'm just annoyed customers are > comparing the two and yet don't appear to have numbers on > this. [...] > Since we don't know how much all the EMC and related stuff cost, it's > hard to accurately compare the two solutions. I bet it is much higher, [...] > Personally I don't mind who prefers what just so long as all the facts > are on the table. Vendor quotes are often confidential, and info gained from NDA's certainly are, so release of numbers can be a big no-no. In a public forum, we can only quote list prices and independent performance numbers and currently shipping features. But here, as in our personal consumer lives, "nobody pays list price." ;) It's hard to make an orange look like an apple; and, it's equally hard to make a pear look like an apple, too. Vendors in a comparison have to be sqeeuzed into a framework that is not complementary to any of them. Apples-to-apples comparisons are very difficult when the paradigms and company philosophies are vastly different, too. If we were talking about CheapPCDiskDriveVendor #1 and #2, it might come down soley to parameter/$ and we probabaly wouldn't be having this interesting discussion. ;) [0] I ran across this quote this weekend which was rather interesting: "There is hardly anything in the world that some man cannot make a little worse and sell a little cheaper, and the people who consider price only are this man's lawful prey." --John Ruskin [2] [1] I'm with Brian (?) on the ops issue, though. More, more, more! [2] I'm not commenting on either vendor in question specifically. I just thought it was a nice thought to think about since so many people get wrapped up in one thing ($) when making purchases, for business or pleasure. Step back from the tree and look at the forest. Until next time... The Mathworks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill(a)mathworks.com http://www.mathworks.com ---

2 1

RE: Problem with WINS registration on 5.3.4R3
by Hawley, Rob 04 Apr '00

04 Apr '00

The rule on WINS is that the WINS server must be able to ping every address on the filer. We added the -wins option on ifconfig since WINS will not work properly if the filer has a private net. The option says to exclude this interface from WINS registration. If you rc_toggle_basic you can use the nbtstat command to show the current registration status. If the entries say broadcast instead of WINS, then we were not able to complete registration with the WINS server. As Andrew said, NetApp customer support would be happy to help you solve this problem. -----Original Message----- From: Bond, Andrew [mailto:abond@netapp.com] Sent: Tuesday, April 04, 2000 2:12 AM To: johan.gillquist(a)cn.saab.se Cc: toasters(a)mathworks.com Subject: RE: Problem with WINS registration on 5.3.4R3 Maybe the WINS server is not accessible to the filer (in which case there should be errors in the messages file). Some versions of ONTAP allow you to run "cifs testdc" and this will report whether the name was registered by WINS or by broadcast because the WINS server was not available (I think this shows as "WINS" or "WINS broadcast" respectively). It is also possible to disable wins registration for specific IP address using the -wins flag to ifconfig. Have you opened a call with NetApp support? There are other tools to debug this.... Regards, Andrew > -----Original Message----- > From: owner-dl-toasters(a)netapp.com > [mailto:owner-dl-toasters@netapp.com]On Behalf Of Johan Gillquist > Sent: 04 April 2000 07:16 > To: netapp mailing list > Subject: Problem with WINS registration on 5.3.4R3 > > > Hi all, > > I have have a problem with my F720 running 5.3.4R3, CIFS and NFS. The > toaster does not seem to register itself in WINS. When I run a "cifs > sessions" command it states "WINS servers: xxx.xxx.x....." > and it's the > correct address. When a check the WINS server there is no entry's for > the filer. If I enter a static entry in the WINs database it > works fine > > Suggestions ? > > /Johan >

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