toasters April 2000

toasters@lists.teaparty.net

121 participants
139 discussions

RE: Filer volume configuration question
by Sims, Aaron 17 Apr '00

17 Apr '00

The configuration you specified is perfectly fine for a cluster install. You do not need two volumes for CFO. You do need two volumes for a destination filer in a SnapMirror configuration. Aaron -- Aaron Sims Professional Services Engineer Network Appliance aaron.sims(a)netapp.com 404-226-7382 (cell) 408-822-8947 (voice mail) -----Original Message----- From: Ari [mailto:ari@genuity.net] Sent: Monday, April 17, 2000 8:31 AM To: toasters(a)mathworks.com Subject: Filer volume configuration question Hi, I am setting up a pair of clustered 740's for the first time and have a question. Each filer currently has one disk shelf. 6 of the disks are in the root volume, /vol/vol0, and the 7th is a hot spare. Is this configuration OK for a clustered pair? I was told by someone that for a cluster, I would need 2 distinct volumes on each head, although I am not entirely sure why. If that is the case, how do I remove drives from the root volume so that I can add them into a newly created volume? Any help is greatly appreciated. Ari Ari Globerman eBusiness Web Hosting Services e-mail: ari(a)genuity.com <mailto:ari@genuity.com> phone: 617-873-5843 fax: 617-873-4033 <http://www.genuity.com/>

1 0

RE: Filer volume configuration question
by Cody, Jason 17 Apr '00

17 Apr '00

Ari- You do NOT need more than one volume on each filer to install a cluster. Your configuration is fine. The person you were talking to may have been thinking about snapmirror. That requires more than one volume to be successful. hope this helps- Jason Cody Network Appliance Professional Services Engineer East Mid-Atlantic Region jcody(a)netapp.com 410-299-6290 Cell 888-374-7504 Pager 408-822-8895 Voice Mail 801-340-7018 eFax -----Original Message----- From: Ari [mailto:ari@genuity.net] Sent: Monday, April 17, 2000 8:31 AM To: toasters(a)mathworks.com Subject: Filer volume configuration question Hi, I am setting up a pair of clustered 740's for the first time and have a question. Each filer currently has one disk shelf. 6 of the disks are in the root volume, /vol/vol0, and the 7th is a hot spare. Is this configuration OK for a clustered pair? I was told by someone that for a cluster, I would need 2 distinct volumes on each head, although I am not entirely sure why. If that is the case, how do I remove drives from the root volume so that I can add them into a newly created volume? Any help is greatly appreciated. Ari Ari Globerman eBusiness Web Hosting Services e-mail: ari(a)genuity.com <mailto:ari@genuity.com> phone: 617-873-5843 fax: 617-873-4033 <http://www.genuity.com/>

1 0

Filer volume configuration question
by Ari 17 Apr '00

17 Apr '00

Hi, I am setting up a pair of clustered 740's for the first time and have a question. Each filer currently has one disk shelf. 6 of the disks are in the root volume, /vol/vol0, and the 7th is a hot spare. Is this configuration OK for a clustered pair? I was told by someone that for a cluster, I would need 2 distinct volumes on each head, although I am not entirely sure why. If that is the case, how do I remove drives from the root volume so that I can add them into a newly created volume? Any help is greatly appreciated. Ari Ari Globerman eBusiness Web Hosting Services e-mail: ari(a)genuity.com phone: 617-873-5843 fax: 617-873-4033

1 0

large disk activity but no net io or nfs?
by mgx＠ornl.gov 17 Apr '00

17 Apr '00

hmmm ... just having a look at sysstat on my F520's this morning: CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache in out read write read write age 79% 1 0 0 2 0 11619 0 0 0 >60 79% 0 0 0 2 0 11314 52 0 0 >60 75% 1 0 0 2 0 11158 0 0 0 >60 76% 8 0 0 48 2 11517 0 0 0 >60 83% 17 0 0 6 3 11700 157 0 0 >60 76% 2 0 0 3 0 11199 0 0 0 >60 80% 12 0 0 4 15 11743 63 0 0 >60 74% 1 0 0 2 0 11329 0 0 0 >60 74% 1 0 0 3 0 10791 45 0 0 >60 72% 2 0 0 3 0 10904 1 0 0 >60 73% 1 0 0 2 0 11728 0 0 0 >60 seems like a lot of CPU and DISK IO for no nfs ops. any idea what could be going on here? thanks! -- michael

4 5

NetApp F7XX + Snapmirror
by TTSG 16 Apr '00

16 Apr '00

Hi, Thanks to all those who helped me over the Networking problem... Now on to bigger/better/harder things. As I mentioned, the customer had a F720 shipped in. When I configured it, I just threw everything into vol0. Now when I look to use SNAPMIRROR to mirror from "A" filer to "B" filer, I can do it by volumes only. Am I correct isn ASSuM(E)ing that if I mirror "vol0" I'll clobber the configuration of "B" filer? I decided to believe this was correct, and reloaded the "B" filer. I created a "vol1" on it with the remaining 5 disks. In attempting to setup the snapmirror, it now complains that the volume is on-line. Um, how is it going to let people still access it if it goes offline? So to see where I can get myself into more trouble, I offline it, reboot, and now its complaining they are different sizes. I have a feeling there is A BUNCH of confusion on my part. Maybe if I say what I want, someone can shoot the clue-gun at me. I want only 1 computer to have update access to filer "A". This will be the host that updates the content on the disk. 3 other computers will have this mounted RO. I will also have filer "B". 4 computers will have this mounted RO. I'd like to have "A" filer copy the content to "B" filer once every (??? not sure yet, say minute to be on the safe side ???). Thanks, Tuc/TTSG

2 3

1400 model
by John Stacey 16 Apr '00

16 Apr '00

IM sure this has come up before, but has anyone attempted overclocking any of the older model filers? Just curious how well they can (or cant) take the upgrades commonly available for 486 based systems. Thanks

3 2

Problem in Automounting
by bhaskar.g＠philips.com 15 Apr '00

15 Apr '00

Hi all, Here is a real good problem. I shall explain it to you in detail. We have a filer F720 with 2*18GB*7disks shelves and we have configure a single volume on that /vol/vol0 We have raid 4 configured on it. The total capacity of the vol0 is about 150GB. We have started dumping data on it and almost crossed 70GB. We have 3 NIS indirect maps for mounting of the directories one for the user home directories one for Project directories one for System Binary directories. All the mount points were accessible till yesterday. Problem is as follows: we had to change the subnet mask of the whole network from 255.255.255.0 to 255.255.254.0 We have done change of netmaks on all the active switches, servers, clients and also the filer. The problem started now Some of the users home directories were not accessible from the this maps but for me and some others there is no problem at all accessing the data root@austin# cd /home/www root@austin# ll . not found root@austin# bdf . bdf: .: Permission denied root@austin# root@austin# cd /home/bhaskarg root@austin# ls .ICAClient NetHood fmfilesvisited .ICEauthority PC Database.xls frame .WindU Personal friend.doc .Xauthority Pop3Server-2_14.zip friendship.pps .acrorc PrintHood gcc-2.95.2-sol7-sparc-local .acrosrch RECEIPT.doc gcc-2.95.2.tar .addrbk Recent general.asp .addressbook SOFT_OF2.DOC ghostscript-5.50-sol7-sparc-local root@austin# root@austin# bdf . Filesystem kbytes used avail %used Mounted on drona:/vol/vol0/home/bhaskarg 150223752 72078752 78145000 48% /tmp_mnt/home/bhaskarg root@austin# when I tried to change the user www's home directory to some other unix system I am able to see the files in the directory. This means that there is not problem in the NIS Indirect map file. Do anyone have solution for this problem or anyone has come across such a type of problem. Please help us in solving the problem Thanks & Regards Bhaskar G Systems Engineer, Philips Semiconductors, #3, Sterling Square,Madras Bank Road, Off St Mark's Road,BANGALORE-560 001. Voice : +91 80 222 9200 Ext 1478 Fax: +91 80 2229700

1 0

RE: IIS and netapp in existing unix world
by Benn, Paul 15 Apr '00

15 Apr '00

Clarification: With the later versions of Data ONTAP you can leave oplocks on. Regards, Paul -----Original Message----- From: Benn, Paul [mailto:paul.benn@netapp.com] Sent: Friday, April 14, 2000 2:11 PM To: 'Mathew Anderson'; toasters(a)mathworks.com Subject: RE: IIS and netapp in existing unix world There are a number of NetApp customers back ending their IIS servers with filers. Make sure you disable oplocks if you are pointing more than one IIS machine to the same data on the filer. When using IIS 4.0, the recommended release of Data ONTAP is 5.3.5P2 or later. You can configure the home directory with any user account you wish; in the lab we usually use the account that the IIS installation app creates. Just make sure it has the appropriate share level and NTFS level permissions (usually read-only for IIS). As I recall, the IUSER* account is configured without a password (I like to change that) and IIS will let you configure the home directory with an incorrect password (even though it may match when you're prompted to verify it). You should be fine using a qtree with mixed permissions. We just published an IIS white paper which should be helpful for initial setup. It can be found at: http://www.netapp.com/tech_library/3078.html Regards, Paul Benn Network Appliance -----Original Message----- From: Mathew Anderson [mailto:mathew@nm.net] Sent: Friday, April 14, 2000 1:35 PM To: toasters(a)mathworks.com Subject: IIS and netapp in existing unix world Hello -- I have a 740 in production in a unix environment. I am looking forward to using this as a backend server for my NT web farm. I created a qtree and set it with mixed permissions (I also would like to set my linux web farm up and mount the same partitions). The problem that I am having is that IIS 4, (on NT4-sp6a) gives the error: Unable to enumerate files and directories because the following error occurred: Access is denied for the home directory setting, I am using a network directory and connecting to the share. I can do the a net view \\machine without a problem. An additional question is, what user should I be connecting as? and with unix permissions set for the volume (and mixed for the qtree) should there be a problem with IIS looking at the mount? I can mount the filesystem without a problem. Thanks for any help. Mathew Anderson Systems Administration NM Technet Oso Grande Technologies, Inc. mathew(a)nm.net mathew(a)osogrande.net

1 0

RE: How do I move /vol/vol0?
by Brookes, Matt 15 Apr '00

15 Apr '00

Actually, you don't have to destroy the old volume until you're happy with the transition, though you might want to offline it. It's also possible to rename the old volume to something other than vol0, and to rename the new volume to vol0, this should save some hassle with exports & other config info. You may still run in to problems with stale file handles if you have clients with active mounts, but I guess since this isn't something you'll be doing regularly that shouldn't be a big deal. (That's why management created weekends, right? ;-) Regards, Matt. ________________________________________________________________ Matthew Brookes mailto:mbrookes@netapp.com Systems Engineer http://www.netapp.com Network Appliance (Sales) Ltd. Tel: +353 1 4064600 Rathfarnham Gate Fax: +353 1 4064601 Dublin 14, IRELAND Mob: +353 86 8575127 "Anyone looking for storage on the Net turns to Network Appliance" Fortune, December 1999. > -----Original Message----- > From: Fox, Adam [mailto:Adam.Fox@netapp.com] > Sent: 14 April 2000 19:42 > To: 'Cliff Nadler'; toasters(a)mathworks.com > Subject: RE: How do I move /vol/vol0? > > > > > > I'm in a situatio where we have one F760 with 4 shelves of > > 9Gb SCSI disk, and > > three shelves of 36Gb FC attached disks. The FC disks were > > added later, so the > > original install of vol0 was done to the SCSI disks. > > > > We'd like to get rid of the SCSI disks (and use one shelf of > > FC to hold it), > > but I'm concerned since we have /vol/vol0 on it. Has anyone > > moved /vol/vol0 off > > SCSI to FC before, and if so, what problems did you run into? > > > > Many customers have used vol copy to move the data over. Just > make sure that the destination fiber-based volume is at least as > large as the original SCSI volume. You vol copy the data over. > Set the new volume to be the root volume. Then you can destroy > the SCSI volume. I'd probably type download, just to be safe. > Then I'd reboot...the filer won't actually destroy the old vol0 > and start using the new volume until you actually reboot. Then > if you want to keep the vol0 name, you can rename the new root > volume. > > This has worked for many customers that I've talked to. Has anyone > seen differently? > > -- Adam Fox > NetApp Professional Services, NC > adamfox(a)netapp.com >

1 0

RE: Disk-to-disk backups? (was Re: Need new backup solution)
by mitch＠netline.com 14 Apr '00

14 Apr '00

/* Andrew Bond [abond(a)netapp.com] writes: */ >I'm not clear how a hacker could destroy data in snapshots - remember that >they are read-only, and that's why they have proved useful in cases of >virus attack. > I can't say what Bruce was thinking, but a cracker could do the obvious: snap delete <vol-name> name once they gain access to the filer either directly or via the admin host. That pretty much destroys data in a snapshot. :-) -- do svidaniya, ~mitch

1 0

← Newer
1
...
4
5
6
7
8
9
10
...
14
Older →

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 April 2000