toasters

toasters@lists.teaparty.net

13520 discussions

Re: Hardware for F330
by sthaug＠nethelp.no 04 May '99

04 May '99

> I was wondering if anyone can help. We are a small shop with 2 F330's with > 2 disk cabinets on each one. We were looking into adding another tray on > each, but were told by NetApps that they are no longer available. > > All I'm looking for is to upgrade the memory, add another scsi controller, > and another tray (is that asking TOO MUCH???). Questions: Can I use off > the shelf memory (and what kind)? What about SCSI controllers? Where can I > get disk trays? Could we instead replace the drives with 9G or 18G drives > (and what model would we use)? > > NetApps solution is to spend $60k+ to upgrade, and all we need is another > 10G - 20G. The current system works fine, isn't overused, and could handle > the expansion IF WE CAN FIND THE COMPONENTS... We are in a somewhat similar situation here. We have one F330, with Seagate Hawk 4 GB disks (ST15230N), running 4.3.3D2. The system is 2.5 years old. We have lost *one* disk (so the hot spare was put into use), and we are thus running a system with no hot spare. We don't particularly like this situation. The Netapp vendor here in Norway tells us that they can no longer get hold of new ST15230Ns, and in general there seems to be quite a problem to get hold of *narrow* disks that can be used with the existing controllers. Questions: - Does anybody know of narrow disks that can be purchased today and which will work with 4.3.3D2? (I have come to point that I no longer care if the disks are Netapp approved - I just want to have a reasonable chance of the disks working.) - We have a number of (used) DEC RZ29B 4 GB disks, fw revision 0007. Will these work? - We feel that the F330 is more or less abandoned, and any kind of upgrade with new disk shelves etc. is going to extremely expensive. Is this a correct assessment? Steinar Haug, Nethelp consulting, sthaug(a)nethelp.no

2 1

Re: Hardware for F330
by sirbruce＠ix.netcom.com 04 May '99

04 May '99

On 05/04/99 16:15:26 you wrote: >- Does anybody know of narrow disks that can be purchased today and which >will work with 4.3.3D2? (I have come to point that I no longer care if the >disks are Netapp approved - I just want to have a reasonable chance of the >disks working.) Netapp has other narrow approved disks they can sell you. You should buy those. >- We have a number of (used) DEC RZ29B 4 GB disks, fw revision 0007. Will >these work? Sure, unless Netapp says there's a problem with that fw revision. You don't have to have all your narrow drives the same brand! Use DECs or Fujitsus or whatever Netapp can sell you. Heck, you can mix brands, mix sizes of drives, even mix narrow and wide in some cases. >- We feel that the F330 is more or less abandoned, and any kind of upgrade >with new disk shelves etc. is going to extremely expensive. Is this a >correct assessment? I think abandoned is a little strong; although it's EOL-ed I was under the impression that Netapp had a reserve stock of components (shelves, etc.) for replacements and upgrades of existing equipment. But if your F330 is already at maximum configuration, it is correct to say that any further expansion/upgrade requires you to buy new equipment. But this is not really anything new; the same was essentially the case when F330s were still in production, except for new cards and the like. Bruce

1 0

eight-shelf F630's
by Brian Rice 04 May '99

04 May '99

Hey, all. I have an F630 with one dual SCSI host adapter and four Shelf 2 shelves...that is, 8a and 8b are fully populated. I am inheriting four more shelves, and I suspect that I'm going to have to scare up another dual SCSI host adapter. When I've seen eight-shelf F630's, I see them with the second SCSI card in slot 9, so the additional shelf pairs are 9a and 9b. But slot 9 in my filer has the NVRAM in it, or so says sysconfig -v. Would someone who has an eight-shelf SCSI F630 send me their sysconfig -v output? Tips on getting to that configuration from the land of four shelves? (My filer is running 5.2.1D5. Yes, we are talkin' multiple RAID groups here. Everybody say "OnTAP 5.x, uh-huh.") Thanks! Brian brice(a)gnac.com

1 0

Hold Music.
by Shaun T. Erickson 04 May '99

04 May '99

Hey NetApp: Your hold music volume is WAY too loud, in relation to the "please keep waiting" voice-over that it alternates with. I'm gonna go deaf if you don't do something about it. -ste

3 3

RE: adding second LRC to FC7 FC-AL shelves
by Norman, Paul 04 May '99

04 May '99

> -----Original Message----- > From: Jay Soffian [mailto:jay@cimedia.com] > Sent: Tuesday, May 04, 1999 9:40 AM > To: toasters(a)mathworks.com > Subject: adding second LRC to FC7 FC-AL shelves > > > Can someone from Netapp give me the run down on upgrading FC-AL > shelves for clustering. I've heard conflicting information. > > We have three FC-AL FC7 shelves that only have a single LRC card in > them. I know that for clustering, the shelves will need an additional > LRC card (for the second FC-AL loop). > > It is not clear to me what is involved in adding the LRC card. I don't > see this documented in the cluster conversion guide nor in the shelf > hardware guide. I have heard that this is a field procedure that > Netapp needs to perform and that after it is done, the shelves need to > be calibrated. I heard the down time for all of this is supposed to be > several hours. Try looking at the Fibre Channel StorageShelf Hardware Installation Guide, page 56, "Installing a Second LRC Card" http://now.netapp.com/knowledge/docs/hardware/fcshelf.pdf It is field installable. We prefer that NetApp SE does the install as the card is sometimes hard to seat properly. There is no calibration, but we have a recipe for the SEs to follow which calls out checking that all disks are visible from both "a" and "b" loop. This is really critical to do BEFORE you attempt any takeovers/givebacks. We prefer that NetApp SEs perform the Cluster conversion to insure that everything is set up correctly. The procedure takes 10 minutes per shelf. You should probably expect about 30 to 45 minutes for the upgrade to include time to add Cluster Interconnect and FC-AL adapters to each filer, then cable up the interconnect & "b" loops. We usually double estimates, so 1.5 to 2 hours should be set aside for this. Paul Norman, Product Manager Direct: (408) 367-3756 Network Appliance E-Mail: pnorman(a)netapp.com 2770 San Tomas Expressway Corp: (408) 367-3100 Santa Clara, CA 95051 Web: WWW.NetApp.Com

1 0

Re: Hardware for F330
by George Kahler 04 May '99

04 May '99

Greetings... sthaug(a)nethelp.no writes: -- snip -- snip -- > Questions: > > - Does anybody know of narrow disks that can be purchased today and which > will work with 4.3.3D2? (I have come to point that I no longer care if the > disks are Netapp approved - I just want to have a reasonable chance of the > disks working.) When Seagate discontinued the ST15230N drives, NetApp replaced them with Fujitsu Alegro M2954E. See if you can get these from your local dealer. I have few running in my F330 with NetApp Release 4.3.5D4. George UNIX Systems Administrator humans: (416) 736-5257 x.22699 Computing Services, York University machines: (416) 736-5830 Ontario, Canada, M3J-1P3 > > - We have a number of (used) DEC RZ29B 4 GB disks, fw revision 0007. Will > these work? > > - We feel that the F330 is more or less abandoned, and any kind of upgrade > with new disk shelves etc. is going to extremely expensive. Is this a > correct assessment? > > Steinar Haug, Nethelp consulting, sthaug(a)nethelp.no

1 0

snap trace
by Jay Soffian 03 May '99

03 May '99

Okay, this is on an F630 running a pretty old DOT (NetApp Release 4.3.4D5: Tue May 26 14:05:43 PDT 1998), so I don't expect any answers, but I thought I'd ask anyway. We just had this box stop passing traffic on it 1000fdx interface. I logged in on console and tried to cycle the 1000fdx interface and got a pretty strange message: whirlpool> ifconfig -a e2: flags=42<BROADCAST,RUNNING> mtu 1500 ether 00:c0:95:f8:2f:6d (Twisted pair) f3: flags=43<UP,BROADCAST,RUNNING> mtu 4352 inet 192.168.254.250 netmask ffffff00 broadcast 192.168.254.255 ether 00:00:f8:bd:c6:04 (Dual Attach Fiber) e10: flags=60043<UP,BROADCAST,RUNNING,TCPCKSUM,IPHDRCKSUM> mtu 1500 inet 192.168.250.250 netmask ffffff00 broadcast 192.168.250.255 ether 00:60:cf:20:48:61 (1000fx) lo: flags=9<UP,LOOPBACK> mtu 1536 inet 127.0.0.1 netmask ff000000 broadcast 127.0.0.1 whirlpool> ifconfig e10 down Couldn't do snap trace, one already active We ended up rebooting the box and it came back up okay. There was no snapshot running. The filer wasn't doing anything out of the ordinary or under heavy load. Nothing in the logs either. It looked fine, except it just stopped passing traffic. The interface was not pingable. I did not try cycling the port on the the switch it's plugged into (a Lucent Cajun P550). The box had been up for over 183 days w/o a single problem. Any idea what that message means? Anyone seen a NetApp just stop passing traffic like this on an 1000fdx interface? Thanks. j. -- Jay Soffian <jay(a)cimedia.com> UNIX Systems Administrator 404.572.1941 Cox Interactive Media

2 1

Re: Problems with CIFS and 5.3.1
by Joan Pearson 03 May '99

03 May '99

In 5.3 support for Unicode was added. CIFS can only access directories that have been converted to the new Unicode format. Directories that are not in Unicode format are converted automatically to Unicode format when first accessed by CIFS. But since snapshots are read only, directories in snapshots cannot be converted and are not accessible from CIFS. Before 5.3, there were two directory formats: those accessed by CIFS (cifs format) and those never accessed by CIFS(non-cifs format). The first time CIFS accessed a directory it would be converted to CIFS format. Non-cifs format directories in snapshots were not accessible from CIFS. In 5.3, both cifs-format and non-cifs format directories are converted to unicode format when accessed by CIFS. Directories created by CIFS are created in unicode format. There is an option wafl.create_ucode that can be used to create unicode directories by default from NFS. Refer to the System Administration Guide for more information. Joan Pearson At 11:08 AM 5/3/99 , you wrote: >I upgraded to 5.3.1 this morning. After the upgrade I noticed I was unable >to access any snapshots via CIFS. The snapshots would show up in Explorer >but when I tried to access the snapshots, even logged in as root, I received >a error message "Access is Denied". > >I tried several ideas to resolve the issue without any luck. I then ran >"cifs testdc" and received the following message: > >_______________________________________________________________ >Using Established configuration >Current Mode of NBT is H Mode > >Netbios scope "" >Registered names... > NEWTOASTER < 0> WINS Broadcast > NEWTOASTER < 3> WINS Broadcast > NEWTOASTER <20> WINS Broadcast > OSSRTPPROD < 0> WINS Broadcast > OSSRTPPROD < 3> WINS Broadcast > OSSRTPPROD <20> WINS Broadcast > >Testing Primary Domain Controller >found 1 addresses > >Testing all Domain Controllers >found 3 addresses >trying 192.168.255.4...192.168.255.4 is alive >trying 192.168.255.4...192.168.255.4 is alive >trying 192.168.255.14...192.168.255.14 is alive >found DC OSSRTPNT40 at 192.168.255.4 >Not able to communicate with DC 192.168.255.4 >found DC OSSRTPBDC at 192.168.255.14 >______________________________________________________ > >Thinking the problem was on my end, I terminated CIFS, deleted the filer on >my PDC and readded the filer. I rebooted the filer and reran "cifs setup" >and kept receiving a error message stating "PDC is not active" and cifs >setup would terminate. > > I finally dropped back to 5.21. > >I reran "cifs testdc" and received the following: >____________________________________________________________________________ >____ >testfiler> cifs testdc >Using Established configuration >Current Mode of NBT is H Mode > >Netbios scope "" >Registered names... > TESTINGFILER < 0> WINS > TESTINGFILER < 3> WINS > TESTINGFILER <20> WINS > OSSRTPPROD < 0> WINS > OSSRTPPROD < 3> WINS > OSSRTPPROD <20> WINS > >Testing Primary Domain Controller >found 1 addresses >trying 192.168.255.4...192.168.255.4 is alive >found PDC OSSRTPNT40 > >Testing all Domain Controllers >found 3 addresses >trying 192.168.255.4...192.168.255.4 is alive >trying 192.168.255.4...192.168.255.4 is alive >trying 192.168.255.14...192.168.255.14 is alive >found DC OSSRTPBDC >found DC OSSRTPNT40 >found DC OSSRTPBDC >____________________________________________________________________________ >____________ > > >Finally, I reran cifs setup without a problem. I was able to get back into >the snapshots. Has anyone else came across a similar issue with 5.3.1 and if >so how did you resolve? >

1 0

Problems with CIFS and 5.3.1
by Mike Ball 03 May '99

03 May '99

I upgraded to 5.3.1 this morning. After the upgrade I noticed I was unable to access any snapshots via CIFS. The snapshots would show up in Explorer but when I tried to access the snapshots, even logged in as root, I received a error message "Access is Denied". I tried several ideas to resolve the issue without any luck. I then ran "cifs testdc" and received the following message: _______________________________________________________________ Using Established configuration Current Mode of NBT is H Mode Netbios scope "" Registered names... NEWTOASTER < 0> WINS Broadcast NEWTOASTER < 3> WINS Broadcast NEWTOASTER <20> WINS Broadcast OSSRTPPROD < 0> WINS Broadcast OSSRTPPROD < 3> WINS Broadcast OSSRTPPROD <20> WINS Broadcast Testing Primary Domain Controller found 1 addresses Testing all Domain Controllers found 3 addresses trying 192.168.255.4...192.168.255.4 is alive trying 192.168.255.4...192.168.255.4 is alive trying 192.168.255.14...192.168.255.14 is alive found DC OSSRTPNT40 at 192.168.255.4 Not able to communicate with DC 192.168.255.4 found DC OSSRTPBDC at 192.168.255.14 ______________________________________________________ Thinking the problem was on my end, I terminated CIFS, deleted the filer on my PDC and readded the filer. I rebooted the filer and reran "cifs setup" and kept receiving a error message stating "PDC is not active" and cifs setup would terminate. I finally dropped back to 5.21. I reran "cifs testdc" and received the following: ____________________________________________________________________________ ____ testfiler> cifs testdc Using Established configuration Current Mode of NBT is H Mode Netbios scope "" Registered names... TESTINGFILER < 0> WINS TESTINGFILER < 3> WINS TESTINGFILER <20> WINS OSSRTPPROD < 0> WINS OSSRTPPROD < 3> WINS OSSRTPPROD <20> WINS Testing Primary Domain Controller found 1 addresses trying 192.168.255.4...192.168.255.4 is alive found PDC OSSRTPNT40 Testing all Domain Controllers found 3 addresses trying 192.168.255.4...192.168.255.4 is alive trying 192.168.255.4...192.168.255.4 is alive trying 192.168.255.14...192.168.255.14 is alive found DC OSSRTPBDC found DC OSSRTPNT40 found DC OSSRTPBDC ____________________________________________________________________________ ____________ Finally, I reran cifs setup without a problem. I was able to get back into the snapshots. Has anyone else came across a similar issue with 5.3.1 and if so how did you resolve?

1 0

Re: Bypassing rc
by Jason D. Kelleher 03 May '99

03 May '99

In message <Pine.BSF.4.10.9905011132060.249-100000(a)schizo.cdsnet.net>, Jaye Mathisen writes: > > >While I sit on hold at netapp, I'm just wondering if anybody's out there. > >I have a filer running 5.3, and I was trying one of the tips on the Netapp >web page about making one of the interfaces a primary for CIFS, by >rearranging things in /etc/rc, and while all I did was move one of >the 2 ifconfig lines, something is boobered up now, and the filer dies >with: > >Sat May 1 11:27:51 PDT [rc]: de_main: e10 : Link down. >Sat May 1 11:27:51 PDT [rc]: de_main: e10 : Link down. >Sat May 1 11:27:52 PDT [rc]: timed: time daemon started >Sat May 1 11:27:52 PDT [Java Thread]: TimeDaemon: Can't connect to time se >rver >'10.0.1.3' >NFS server is running. >timed not found. Type '?' for a list of commands >quota: vol0: run 'quota off' before 'quota on'. > >PANIC: ../common/nbt/nbns_receive.c:207: Assertion failure. on Sat May 1 1 >8:27: >53 1999 > >version: NetApp Release 5.3: Thu Apr 1 11:18:39 PST 1999 >cc flags: 1 >dumping core: ..................................................Old core pr >esent > on disk --- not dumped. > > > >Anyway, I guess I'll try some floppies, and see how far I get while I >wait... > >Any tips appreciated. Stay away from 5.3 w/out speaking to tech support first. Not sure about 5.3.1, but it took three "D" releases to get 5.3 working properly for me. And since the "Closed Bugs by fix release" for 5.3.1 on NOW looks shorter than the list of bugs fixed in my 5.3Dx release, I'd be real careful. The "Bugs" section in NOW has gotten alot more usable. (In fact, all of NOW has gotten better recently.) But I still want a "Known Bugs by release". jason

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters