toasters

toasters@lists.teaparty.net

3 participants
13514 discussions

Re: OT - swag
by Martin Hughes 20 Apr '01

20 Apr '01

I second that one. And we've got the SAM agreement to boot. Hmmmm...... Regards Martin Enterprise Storage/Enterprise Server Group <mailto:martin.hughes@halliburton.com> cellphone: +1 713 303 9214 office: +1 281 871 4124 -------------------------- Sent from BlackBerry Wireless Handheld (www.BlackBerry.net) -----Original Message----- From: Chuck Ross <chuck.ross(a)signiant.com> To: Net App List (E-mail) <toasters(a)mathworks.com> Sent: Fri Apr 20 13:24:04 2001 Subject: OT - swag Man, you guys suck! When I went to Net App training all we got was a travel coffee mug and a really lame tote bag. Chuck Ross > > > I think I've got everyone on this thread beat. I am the > proud owner of a > black leather NetApp jacket. Currently the best vendor swag > in my whole > wardrobe. > > -- Jeff > > > On Fri, Apr 20, 2001 at 02:54:44PM +0200, Leif Nixon wrote: > > "Timothy Cook" <TCOOK(a)joy.com> writes: > > > > > Maybe I'm missing out on the really important aspect of netapps??? > > > > > > Other than a jacket I got at Admin training I've never received my > > > new clothing line from Netapp! Is this an optional part of the > > > Support Agreement, or should I start complaining that I'm not > > > receiving the level of support from Netapp that I should be! > > > > Yes, you might want to consider that. > > > > A couple of months ago I added a new 36-gigger to my F720 setup, and > > since I failed to notice the little sticker on the bottom > of the disk > > about supported ONTAP versions, the filer halted and production > > stopped. Ouch. > > > > Not two hours later I received an overseas FedEx package containing > > a NetApp sweater. Now that's service! > > > > -- > > Leif Nixon Sysadm/developer Ericsson SoftLab AB > > ---------------------------------------------------------- > > E-mail: nixon(a)softlab.ericsson.se Phone: +46 13 23 57 61 > > ---------------------------------------------------------- > > >

1 0

RE: Vif and Cluster setup questions
by Krzesinski, John 20 Apr '01

20 Apr '01

In a clustered environment, partnering a VIF needs the VIF name of the other filer's VIF rather than it's IP address (and it makes it easier if you use names that are easily differentiated between the two filers)... > -----Original Message----- > From: Art Hebert [mailto:art@arzoon.com] > Sent: Friday, April 20, 2001 7:04 PM > To: 'Sam Cramer' > Cc: 'toasters(a)mathworks.com' > Subject: RE: Vif and Cluster setup questions > > > > > > > Here is my rc lines: > > > > ifconfig e6 mediatype 100tx-fd > > ifconfig e0 mediatype 100tx-fd > > vif create multi vif01 e6 e0 > > vif favor e6 > > ifconfig vif01 `hostname`-e6 mediatype 100tx-fd partner vif01 > ^^^^^ > >When specifying the partner of a vif, use the IP address of > the partner > >interface, not the interface name. > > >Sam > > > Sam -- I tried using the partner address and it complained. > I also tried > the hostname of the partner and it also complained. Then I > read the archives > and someone said you needed to use the name of the virtual > interface to get > the cluster to work. > > > > >

1 0

Re: Vif and Cluster setup questions
by Sam Cramer 20 Apr '01

20 Apr '01

> I can live without load balancing (though I would think that once the one > link was saturated, packets from another source would go over the > ethernet connection in the vif). In most circumstances it should load balance, but the algorithm used by the switch actually doesn't attempt to "load balance" per se; it simply distributes traffic across the links based on the source and destination address of the incoming packet. In other words, the switch doesn't pay attention to the "load" on each link. By using a simple hash function on the incoming packets it distributes them over the available links; if the packets vary in such a way that the hash function distributes them fairly evenly, the load on each link on each link should be roughly equal. Sam

1 0

Re: Vif and Cluster setup questions
by Sam Cramer 20 Apr '01

20 Apr '01

I wrote > When specifying the partner of a vif, use the IP address of the partner > interface, not the interface name. Sigh... this is exactly backwards. As the man page (na_vif(1)) notes, use the interface name, not the address. Sorry for any confusion, and I promise not to skip that second cup of coffee before replying next time! Sam

1 0

RE: How would you respond to this -Reply
by Joe Luchtenberg 20 Apr '01

20 Apr '01

You could probably find retail space available at one of the 38 Gateway Country stores that they announced plans to close recently... ;-) Come to think of it, there's no way NetApp management would sign off on it. You wouldn't make the same kind of profit (clothing retail is pretty cut throat, I understand), and shareholders would not stand for overall gross margins dropping below XX% (unless EMC's dropped, too) [1]. Oops, shareholder opinion may be a touchy subject nowadays... [1] Sanitized gross margin figure in case this is a touchy matter to anyone; if it really matters to you the information is public and easy to find; the way I see it NetApp can only charge for their product what the market will bear, so no you didn't pay too much. Joe Joe Luchtenberg Dataline, Inc. 757.858.0600 757.285.1223 (cell) 757.858.0606 (fax) joe.luchtenberg(a)data-line.com www.data-line.com -----Original Message----- From: Mohler, Jeff [mailto:jeff.mohler@netapp.com] Sent: Friday, April 20, 2001 1:16 PM To: Benn, Paul; toasters(a)mathworks.com Subject: RE: How would you respond to this -Reply I still think we need a company store...*chuckle* -----Original Message----- From: Benn, Paul [mailto:paul.benn@netapp.com] Sent: Friday, April 20, 2001 9:16 AM To: toasters(a)mathworks.com Subject: RE: How would you respond to this -Reply Gee, NetApp clothing must be valuable. I've seen a few items on eBay: http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1227231380 http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1227655648 -----Original Message----- From: Gavin Kelman [mailto:gavin@husky.muscat.com] Sent: Friday, April 20, 2001 7:11 AM To: Timothy Cook Cc: toasters(a)mathworks.com Subject: Re: How would you respond to this -Reply On Fri, Apr 20, 2001 at 08:35:16AM -0400, Timothy Cook wrote: > Maybe I'm missing out on the really important aspect of netapps??? > > For the benefit of you netapp guys who monitor this board, I wear an Extra Large. Just out of interest, can you get ahold of netapp clothing out with them giving you the odd t-shirt at meetings and sales opportunites? My single netapp t-shirt is wearing thin... :) -- Gavin Kelman Phone : 01223 715043 gavin(a)husky.muscat.com Mobile : 07967 360822 Systems Administrator Fax : 01223 715050

1 0

RE: Vif and Cluster setup questions
by Art Hebert 20 Apr '01

20 Apr '01

> Here is my rc lines: > > ifconfig e6 mediatype 100tx-fd > ifconfig e0 mediatype 100tx-fd > vif create multi vif01 e6 e0 > vif favor e6 > ifconfig vif01 `hostname`-e6 mediatype 100tx-fd partner vif01 ^^^^^ >When specifying the partner of a vif, use the IP address of the partner >interface, not the interface name. >Sam Sam -- I tried using the partner address and it complained. I also tried the hostname of the partner and it also complained. Then I read the archives and someone said you needed to use the name of the virtual interface to get the cluster to work.

1 0

RE: How would you respond to this -Reply
by Mohler, Jeff 20 Apr '01

20 Apr '01

I still think we need a company store...*chuckle* -----Original Message----- From: Benn, Paul [mailto:paul.benn@netapp.com] Sent: Friday, April 20, 2001 9:16 AM To: toasters(a)mathworks.com Subject: RE: How would you respond to this -Reply Gee, NetApp clothing must be valuable. I've seen a few items on eBay: http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1227231380 http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1227655648 -----Original Message----- From: Gavin Kelman [mailto:gavin@husky.muscat.com] Sent: Friday, April 20, 2001 7:11 AM To: Timothy Cook Cc: toasters(a)mathworks.com Subject: Re: How would you respond to this -Reply On Fri, Apr 20, 2001 at 08:35:16AM -0400, Timothy Cook wrote: > Maybe I'm missing out on the really important aspect of netapps??? > > For the benefit of you netapp guys who monitor this board, I wear an Extra Large. Just out of interest, can you get ahold of netapp clothing out with them giving you the odd t-shirt at meetings and sales opportunites? My single netapp t-shirt is wearing thin... :) -- Gavin Kelman Phone : 01223 715043 gavin(a)husky.muscat.com Mobile : 07967 360822 Systems Administrator Fax : 01223 715050

1 0

RE: How would you respond to this -Reply
by Mohler, Jeff 20 Apr '01

20 Apr '01

i ran into the same problem with [i believe] the cheetah 18G drives. i later found the insignificant little piece of paper stating that there might be a problem if you didn't download the latest disk defs. i can understand the filer not recognizing/using the disk. but slamming to a halt? brrrrrr. the netapp engineer agreed that the notice should be large red and very visable. sad to see that this did not happen. if anyone in netapp-land is listening, could this packaging bug be fixed? thank you. i'd be happy to provide case details to netapp upon request. --- This has a fix. This is part of the 6.1 release which is currently in FCS. You can check details on this on the NOW site. The root of this is..we are very particular over what drives we allow in our systems. Different vendors have different way of "working" and we thoroughly test all drives, models, and specific firmware versions before we release it to the wild. Data protection is whats on our mind here. Clearly there is always a better way, and feedback into Engineering asking why we did it that way resulted in reocnstructing how we deal with non-qualified drives and it no longer causes downtime in the current FCS release of ONTAP 6.1. Please review the FCS documentation on NOW to better understand what our OS release procedures and definitions are. If anyone ever has features for different ways of doing things they want to be heard here at Netapp..open a case and ask. Your technical support rep. will find the answers..and if they dont exist will open an RFE. Your sales contacts are also open to suggestions from the field. Free leather jackets are not considered RFE material..just so were all in sync on that. :^)

1 0

RE: How would you respond to this -Reply
by Benn, Paul 20 Apr '01

20 Apr '01

Gee, NetApp clothing must be valuable. I've seen a few items on eBay: http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1227231380 http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1227655648 -----Original Message----- From: Gavin Kelman [mailto:gavin@husky.muscat.com] Sent: Friday, April 20, 2001 7:11 AM To: Timothy Cook Cc: toasters(a)mathworks.com Subject: Re: How would you respond to this -Reply On Fri, Apr 20, 2001 at 08:35:16AM -0400, Timothy Cook wrote: > Maybe I'm missing out on the really important aspect of netapps??? > > For the benefit of you netapp guys who monitor this board, I wear an Extra Large. Just out of interest, can you get ahold of netapp clothing out with them giving you the odd t-shirt at meetings and sales opportunites? My single netapp t-shirt is wearing thin... :) -- Gavin Kelman Phone : 01223 715043 gavin(a)husky.muscat.com Mobile : 07967 360822 Systems Administrator Fax : 01223 715050

1 0

RE: Vif and Cluster setup questions
by Art Hebert 20 Apr '01

20 Apr '01

Eyal -- I tried vif favor in both multi and single. I can live without load balancing (though I would think that once the one link was saturated, packets from another source would go over the ethernet connection in the vif). My main concern is it wants to use e0 all the time. We have found here that we had to move the 100mb nic to slot 6 to get performance, we had netapp come out and we finally had to use the following configuration. Copying a 100mb file to e0 would take 24 sec, copying to the nic card in another slot would take 24 sec. It wasn't until we moved it to slot 6 and used 8k packets did we get 10 sec copy times. So the e6 is 100mb. smf760-01> sysconfig -v NetApp Release 5.3.6R2: Sat Aug 5 09:40:44 PDT 2000 System ID: 0016800487 (smf760-01); partner ID: 0016800485 (smf760-02) slot 0: System Board (NetApp System Board V N8) Model Name: F760 Serial Number: 235663 Firmware release: 2.3_a2 Memory Size: 1024 MB slot 0: Fibre Channel Host Adapter 0a (Qlogic ISP2100 rev. 3, 64-bit) Firmware rev: 1.19.6 Host Loop Id: 7 FC Node Name: 2:000:00a098:03988f Cacheline size: 8 FC Packet size: 2048 slot 0: SCSI Host Adapter 0b (Qlogic ISP 1040B) Fast/Wide, Differential Firmware Version 4.65 Clock Rate 60MHz. slot 0: Ethernet Controller e0 MAC Address: 00:a0:98:00:7d:b4 (auto--down) slot 4: Fibre Channel Host Adapter 4 (Qlogic ISP2100 rev. 3, 32-bit) Firmware rev: 1.19.6 Host Loop Id: 7 FC Node Name: 2:000:00e08b:018aa7 Cacheline size: 8 FC Packet size: 2048 0: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 1: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 2: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 3: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 4: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 5: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 6: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) Shelf 0: VEM Kernel Version : 2.5 App. Version : 3.0 slot 6: FastEnet-10/100 Ethernet Controller Device Type: 82559 e6 MAC Address: 00:90:27:be:a8:b5 (auto-100tx-fd-up) slot 7: Fibre Channel Host Adapter 7 (Qlogic ISP2100 rev. 3, 32-bit) Firmware rev: 1.19.6 Host Loop Id: 7 FC Node Name: 2:000:00e08b:0189a7 Cacheline size: 8 FC Packet size: 2048 0: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 1: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 2: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 3: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 4: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 5: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) 6: SEAGATE ST318203FC NA07 Size=17.0GB (35566480 blocks) Shelf 0: VEM Kernel Version : 2.5 App. Version : 3.0 slot 8: Interconnect adapter (Tandem ServerNet SPA 1.5) slot 9: NVRAM (NetApp NVRAM II) Revision: A0 Serial Number: 23271 Memory Size: 32 MB Battery 1 Status: 100% (3.22v) Battery 2 Status: 100% (3.22v) thanks art -----Original Message----- From: Traitel, Eyal [mailto:eyal@netapp.com] Sent: Friday, April 20, 2001 8:52 AM To: 'Art Hebert'; 'toasters(a)mathworks.com' Subject: RE: Vif and Cluster setup questions I'm almost 100% percent sure that vif favor will affect only single vifs. I will check later on, but you should be able to see that in docs as well. Regarding load balancing - EtherChannel is more about failover than balancing. It distributes to ports based on last 2 bits of MAC address. In real life, one device is becoming the one that's more used than the others. Is e6 GbE ? Can you post your sysconfig -v ? 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: Art Hebert [mailto:art@arzoon.com] Sent: Friday, April 20, 2001 4:55 PM To: 'toasters(a)mathworks.com' Subject: Vif and Cluster setup questions > I've read all the posts and I am having some success and some failures. > > > 2 F760 filers clustered together, onboard ethernet (e0), 10/100 nic > (e6) > Cisco 6509 switch trunked together > > I'm trying to get the e0 and e6 ports to work failover for each > other and ideally > load balance to get better performance. > > I've been able to get the virtual interface running and fail the > cluster and everything is > fine but I can't get it to load balance (I even tried 4 clients > hitting it copying 500mb files) > and it seems to always want to use e0, even if I do a "vif favor e6" > > Here is my rc lines: > > ifconfig e6 mediatype 100tx-fd > ifconfig e0 mediatype 100tx-fd > vif create multi vif01 e6 e0 > vif favor e6 > ifconfig vif01 `hostname`-e6 mediatype 100tx-fd partner vif01 > > Now I added the vif favor e6 even though its multi (shouldn't have > to it should load balance), right? > Does it matter what order the e6 and e0 commands come in, seems that > it affects how vif stat reports? > Can I even use e0 and e6 together or should I have another card? its > on the 32 bit bus > and performance is half of e6. > Are the nic cards in the netapps anything special or can I use a > well known brand ? > > > Thanks > > Art Hebert Email: art(a)arzoon.com <mailto:art@arzoon.com> > Arzoon IT Dept. Phone: 650 522-6125 > Mobile: 650 444-8667 >

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