toasters

toasters@lists.teaparty.net

13515 discussions

RE: Snapmirror initialization aborts consistently
by Mike Partyka 07 Nov '07

07 Nov '07

Sorry, wrong snapmirror.conf, here is right one: rdfile /etc/snapmirror.conf #Regenerated by registry Wed Oct 24 02:29:43 GMT 2007 hci:vol1 hci2:rcv_vol1 - 0 23 * * hci:data hci2:rcv_data - 0 21 * * -----Original Message----- From: Chris Blackmor [mailto:chris.blackmor@amd.com] Sent: Tue 11/6/2007 5:26 PM To: Mike Partyka Subject: Re: Snapmirror initialization aborts consistently This is a VSM - yes? Could you send the df from the source and destination, the snapmirror.conf entry, the initialize command, and the log entries from /etc/log/snapmirror on the source and the destination? C- Mike Partyka wrote: > Hello, > > I'm having a problem with Async SnapMirror where i have a 500G flexvol on both source and destination. When i initialize it fails at 83G consistently, I've destroyed the volume and rebuilt it several times but the problem reoccurs each time. I've run the source snapmirror logs relating to the failure through the syslog translator and all it really says is that "this is a generic snapmirror error on source", which just isn't very helpful. > > Here are the logs from the source: > Tue Nov 6 16:04:46 CST [pipeline_3:notice]: snapmirror: Network communication error > Tue Nov 6 16:04:46 CST [snapmirror.src.err:error]: SnapMirror source transfer from data to hci2:rcv_data : transfer failed. > Tue Nov 6 16:26:36 CST [pipeline_3:notice]: snapmirror: Network communication error > Tue Nov 6 16:26:36 CST [snapmirror.src.err:error]: SnapMirror source transfer from data to hci2:rcv_data : transfer failed. > > The source 3050a is running DOT 7.0.5 and the destination is running DOT 7.0.6 > > The volume options are identical as far as i can tell based on the "vol options -v data_vol" command. > > The filers sites are connected via a GbE MAN, so bandwidth isn't the problem. > > I've checked for errors on the ethernet interfaces on both ends and none of the bad counter are not incrementing. > > Has anyone on the list experienced something like this? Or have any troubleshooting advice? > > Thx > > Mike Partyka > > > > -- --------------------------------------------------------------------- * Chris Blackmor _______ | Oh my God... * * Advanced Micro Devices \____ | | We're all temps! * * Phone: (512) 602-1608 /| | | | * * Fax: (512) 602-5155 | |___| | | Daisy Adair * * Email: chris.blackmor(a)amd.com |____/ \| | Dead Like Me * --------------------------------------------------------------------- * My comments are mine, and mine alone. * ---------------------------------------------------------------------

3 4

How many 750G disks can can fit into an aggregate?
by Stephen C. Losen 07 Nov '07

07 Nov '07

We're going to buy some shelves of 750G ATA drives. An aggregate is limited to 16T "raw" (which I presume includes parity disks). Not having any 750G drives yet, we don't know what they size down to. Can anyone tell us how many 750G disks (including parity) that you can stuff into an aggregate? Steve Losen scl(a)virginia.edu phone: 434-924-0640 University of Virginia ITC Unix Support

14 16

RE: Snapmirror initialization aborts consistently
by Mike Partyka 06 Nov '07

06 Nov '07

Yes, sorry, VSM df, i can't do while the destination is restricted, i get this when i try to get df output hci2> vol status -b rcv_data rcv_data: volume is offline or is unavailable. hci2> vol online rcv_data vol online: Volume 'rcv_data' was left in an inconsistent state by an aborted vol copy or an aborted snapmirror initial (level 0) transfer. In order to bring it online, you must either destroy and re-create the volume, or complete an initial snapmirror transfer or vol copy. I did compare the source and destination with the "vol status data -b" and they were the same number of blocks. rdfile /etc/snapmirror.conf #Regenerated by registry Wed Oct 24 02:29:43 GMT 2007 netapp1:vol1 netapp2:rcv_vol1 - 0 23 * * netapp1:data netapp2:rcv_data - 0 21 * * Here are the destination snapmirror logs from today dst Tue Nov 6 12:04:56 CST hci:data hci2:rcv_data Request (Initialize) dst Tue Nov 6 12:05:08 CST hci:data hci2:rcv_data Start dst Tue Nov 6 13:50:04 CST hci:data hci2:rcv_data Abort (snapmirror transfer fa iled to complete) dst Tue Nov 6 14:12:36 CST hci:data hci2:rcv_data Request (Initialize) dst Tue Nov 6 14:12:37 CST hci:data hci2:rcv_data Restart (@ 82881 MB) sys Tue Nov 6 14:20:15 CST SnapMirror_off (registry) dst Tue Nov 6 14:20:17 CST hci:data hci2:rcv_data Abort (snapmirror transfer fa iled to complete) sys Tue Nov 6 14:25:28 CST SnapMirror_on (registry) dst Tue Nov 6 14:25:31 CST hci:data hci2:rcv_data Request (Initialize) dst Tue Nov 6 14:25:35 CST hci:data hci2:rcv_data Start dst Tue Nov 6 16:10:38 CST hci:data hci2:rcv_data Abort (snapmirror transfer fa iled to complete) sys Tue Nov 6 16:11:31 CST SnapMirror_off (registry) sys Tue Nov 6 16:16:07 CST SnapMirror_on (registry) dst Tue Nov 6 16:18:37 CST hci:data hci2:rcv_data Request (Initialize) dst Tue Nov 6 16:18:38 CST hci:data hci2:rcv_data Restart (@ 82347 MB) dst Tue Nov 6 16:32:28 CST hci:data hci2:rcv_data Abort (snapmirror transfer fa iled to complete) -----Original Message----- From: Chris Blackmor [mailto:chris.blackmor@amd.com] Sent: Tue 11/6/2007 5:26 PM To: Mike Partyka Subject: Re: Snapmirror initialization aborts consistently This is a VSM - yes? Could you send the df from the source and destination, the snapmirror.conf entry, the initialize command, and the log entries from /etc/log/snapmirror on the source and the destination? C- Mike Partyka wrote: > Hello, > > I'm having a problem with Async SnapMirror where i have a 500G flexvol on both source and destination. When i initialize it fails at 83G consistently, I've destroyed the volume and rebuilt it several times but the problem reoccurs each time. I've run the source snapmirror logs relating to the failure through the syslog translator and all it really says is that "this is a generic snapmirror error on source", which just isn't very helpful. > > Here are the logs from the source: > Tue Nov 6 16:04:46 CST [pipeline_3:notice]: snapmirror: Network communication error > Tue Nov 6 16:04:46 CST [snapmirror.src.err:error]: SnapMirror source transfer from data to hci2:rcv_data : transfer failed. > Tue Nov 6 16:26:36 CST [pipeline_3:notice]: snapmirror: Network communication error > Tue Nov 6 16:26:36 CST [snapmirror.src.err:error]: SnapMirror source transfer from data to hci2:rcv_data : transfer failed. > > The source 3050a is running DOT 7.0.5 and the destination is running DOT 7.0.6 > > The volume options are identical as far as i can tell based on the "vol options -v data_vol" command. > > The filers sites are connected via a GbE MAN, so bandwidth isn't the problem. > > I've checked for errors on the ethernet interfaces on both ends and none of the bad counter are not incrementing. > > Has anyone on the list experienced something like this? Or have any troubleshooting advice? > > Thx > > Mike Partyka > > > > -- --------------------------------------------------------------------- * Chris Blackmor _______ | Oh my God... * * Advanced Micro Devices \____ | | We're all temps! * * Phone: (512) 602-1608 /| | | | * * Fax: (512) 602-5155 | |___| | | Daisy Adair * * Email: chris.blackmor(a)amd.com |____/ \| | Dead Like Me * --------------------------------------------------------------------- * My comments are mine, and mine alone. * ---------------------------------------------------------------------

1 0

RE: HTTP admin access denied
by Learmonth, Peter 06 Nov '07

06 Nov '07

Make sure you're pointing your browser at http://filer_name_or_ip/na_admin Http.rootdir only applies when serving general-purpose http (statis content). Nothing to do with admin. Your http.admin options look fine. If you did an upgrade or fresh install, make sure everything is there. Type ndmpd and see if the command is found. If not, then like tmac said, re-extract the ONTAP archive and reboot. Share and Enjoy! Peter ... Sent from one of those phone+email doohicky things... I reserve the right to be wrong. -----Original Message----- From: Roger G [mailto:ro_gi@msn.com] Sent: Tuesday, November 06, 2007 02:29 PM Pacific Standard Time To: toasters(a)mathworks.com Subject: HTTP admin access denied Hello, For some reason I can not access the Netapp admin Interface. The server returns with the following errors: Error 503 HTTP is not licensed.To administer this filer, use /na_admin/ . Error 500 Servlets not enabled httpd and httpd.admin are enabled, but something is wrong with the httpd.rootdir and I can't reset it. filer> options httpd httpd.access legacy httpd.admin.access legacy httpd.admin.enable on httpd.admin.hostsequiv.enable off httpd.admin.ssl.enable off httpd.autoindex.enable on httpd.enable on httpd.log.format common httpd.rootdir XXX httpd.timeout 300 httpd.timewait.enable off filer> options httpd.rootdir httpd.rootdir XXX filer> options httpd.rootdir /vol/http warning: httpd.rootdir (/vol/http) not accessible Is there a way to recreate the /vol/http with all of its content in case it was deleted ? Do you have any idea how to solve this? -- View this message in context: http://www.nabble.com/HTTP-admin-access-denied-tf4760881.html#a13615379 Sent from the Network Appliance - Toasters mailing list archive at Nabble.com.

1 0

accessing lun as a raw device using dm-multipathing from a rhel host
by Alexander Schalek 06 Nov '07

06 Nov '07

Hi! I have some troubles with a oracle rac installation on netapp fcp luns. We use rhel4 u5 with dm-multipathing with a 2.6 kernel Currently we have 4 partitioned and multipathed luns maped to the hosts(mpath0-3) but the rac installation exits with an error. I found the following lines in the fcp host utilities setup guide: Accessing LUNs using dm-multipath Accessing LUNs as a raw devices Overview Instead of creating a file system, you can access a LUN as a raw device. Be sure to use the /dev/mapper/ device created by dm-multipath to access the raw device. For the Linux 2.6 kernel used by RHEL 4, do not use the raw interface. Instead open the device using the open call with the O_DIRECT flag. ***** So I have to access the devices using the open call with the O_DIRECT flag!? Can anyone of you guys explain this sentence to me? How do I do that? Many thanks and best regards, Alexander

2 4

Single command to determine # of files on a volume
by Robert M Bailey 06 Nov '07

06 Nov '07

Besides using quota report and adding up the "File Used" for all the qtrees, is there a simpler way to get a summary file usage on a volume? Bob Bailey Network Centric Systems (NCS) Information Solutions Enterprise Operations Server Management & Central Operations Desk Phone: 972-952-2518 Pager: 972-670-7336 Alpha Pager: 9726707336(a)messaging.nextel.com

5 4

Symantec Antivirus Corporate Edition 10.1 and vscan on filer
by Richard Distrajo 04 Nov '07

04 Nov '07

Any help on registering the filer to the Symantec antivirus server. I tried to put the filer's hostname /ip address to the symantec antivirus center console but no success. Is this version of antivirus can register the CIFS only filer (ONTAP 7.2.2)? I'm getting error "no network provider accepted the given network path" every time I add the filer's IP address to the symantec antivirus console center...both sides can able to ping each other's hostname and user's permission is already been added as operator on the filer's side.... Thanks for your time on this.... Best regards, Richard Send instant messages to your online friends http://uk.messenger.yahoo.com

5 6

Re: R: VMWare CIFS driver
by Alan McLachlan 04 Nov '07

04 Nov '07

Hi Milazzo, You are absolutely correct, and beyond lower performance there's an issue of maintaining the CIFS connection in the case of cluster failover. However, if there's one thing that's true about I.T. it's that if something is _possible_, someone somewhere will have a set of circumstances (not necessarily technical, often financial or political), where that solution makes sense for them. 10 years ago it was a very hard sell to get people to access enterprise databases over any sort of file protocol, now Oracle themselves have vast deployments of Oracle over NFS. And Microsoft SQL Server over CIFS, while not so common, has been used for many years with both reasonable performance and high availability by many NetApp customers (although now we would recommend iSCSI with SnapDrive instead of course), and is a configuration that was originally supported by Microsoft only in the context of NetApp storage. It's very easy to say "that's a stupid idea", but eventually you learn that something that appears to make little sense on it's own can make more sense for a particular environment once all of the issues (technical and non-technical) are considered. There are serious scalability issues with using iSCSI or FCP for VMWare system images (leaving the data volumes aside as a seperate issue that is handled on it's own merits), and the NFS implementation is a very good solution for this apart from improving overall manageability of a large VMWare environment. While I would hestitate before using CIFS for this in any production environment, it may have value in creating a simple and scaleable dev/test solution for a client who then would still use iSCSI or FCP for their production vm images. regards, Alan. ----- Original Message ----- From: "Milazzo Giacomo" To: "M. Vaughn Stewart" , "Alan McLachlan" Subject: R: VMWare CIFS driver Date: Thu, 1 Nov 2007 12:13:11 +0100 Hi vmdk images streamed on CIFS? :-) Sorry if I smile but performances will be...terrible!!! That's the reason for the NFS usage available on VMWare.And, after all, VMWare is a RHEL 4.x Linux distribution..to "read" CIFS the service console should have a SAMBA interface on it...heavy role!!! I agree with the license cost of NFS and CIFS together but, after all..iSCSI is free on NetApp. Create an iSCSI LUN and map it to VMWare...best practice! Bye -------------------------------------------------------------------- Da: owner-toasters(a)mathworks.com per conto di M. Vaughn Stewart Inviato: lun 29/10/2007 5.45 A: Alan McLachlan Cc: toasters(a)mathworks.com Oggetto: Re: VMWare CIFS driver Thanks for bringing this to our attention V Alan McLachlan wrote: > Hi Peter, > > I can't disagree with anything you have said, particularly re. failover (see my post that followed). > > Incidentally though, it was an internal Network Appliance whitepaper that alerted me to the possibility of using CIFS for .vmdk files. > > See section 9.1.3 on page 58 of TR-3505i. > > I understand that being an internal paper I shouldn't be mentioning it here, and the published version of that paper on A-SIS eliminates the discussion of VMWare, I suspect because support issues haven't been nailed down yet. > > Still waiting on feedback from my contact in VMWare, will post when I hear something. > > There are other reasons beside statefulness of CIFS connections to prefer NFS (performance, etc.), but that's the main issue. I only mentioned it so as not to exclude it as a technical possibility, my followup posts were intended to stop anyone getting too excited about it at this stage. Also even if VMWare were to build it into the kernal, I'd still be pushing clients to pay for an NFS licence instead. The issue here is the cost of the NFS licence, which can be considered high for some Windows-only shops. > > regards, > > Alan. > > >> ----- Original Message ----- >> From: "Learmonth, Peter" <Peter.Learmonth(a)netapp.com> >> To: "Alan McLachlan" <alanmac(a)technologist.com>, toasters(a)mathworks.com >> Subject: RE: VMWare CIFS driver >> Date: Sun, 28 Oct 2007 19:10:27 -0700 >> >> >> Hi Alan >> Would you happen to have a URL instead of a IP address? >> >> Given that this comes from an unidentifiable web site, I highly doubt >> this will get any support from VMware, and I can pretty much guarantee >> that this will get no support from NetApp. >> >> More on the RPM itself... >> It contains only an object module, which appears to install in the >> service console Linux, not the VMKernel, although I could be wrong on >> that. >> >> The version on that RPM matches the ESX 3.0.1 COS Linux kernel version. >> >> Besides, while I think NFS is great for VMware, I think CIFS is >> horrible. People who have tried it with VMware server or GSX have had >> some surprises when the storage fails over and CIFS drops the >> connections. It doesn't make any difference whose CIFS server you're >> using. I personally don't think VMware should waste any time developing >> an ESX kernel client. >> >> Thanks anyway. >> >> Peter >> >> >> -----Original Message----- >> From: Alan McLachlan [mailto:alanmac@technologist.com] >> Sent: Sunday, October 28, 2007 4:24 PM >> To: toasters(a)mathworks.com >> Subject: VMWare CIFS driver >> >> >> Hi, >> >> Find it here: >> >> http://89.105.41.70/sites/esx/Lists/Posts/Post.aspx?ID=1 >> >> Enjoy! >> >> regards, >> >> Alan McLachlan >> >> >> >> >>> ----- Original Message ----- >>> From: "Davies,Matt" <MDAVIES(a)generalatlantic.com> >>> To: Peter.Learmonth(a)netapp.com, alanmac(a)technologist.com, >>> toasters(a)mathworks.com >>> Subject: RE: VMware / A-SIS >>> Date: Sun, 28 Oct 2007 21:48:16 -0000 >>> >>> >>> I was about to ask the same question..... >>> >>> -----Original Message----- >>> From: owner-toasters(a)mathworks.com <owner-toasters(a)mathworks.com> >>> To: Alan McLachlan <alanmac(a)technologist.com>; >>> toasters(a)mathworks.com <toasters(a)mathworks.com> >>> Sent: Sun Oct 28 21:09:20 2007 >>> Subject: RE: VMware / A-SIS >>> >>> Ummmm.... >>> What ESX CIFS driver? >>> >>> -----Original Message----- >>> From: Alan McLachlan [mailto:alanmac@technologist.com] >>> Sent: Sunday, October 28, 2007 8:26 AM >>> To: toasters(a)mathworks.com >>> Subject: VMware / A-SIS >>> >>> Hi Vaughn, >>> >>> For small environments on a budget where the cost of additional >>> licenses is an issue, you can use RDM's over iSCSI or FCP for >>> both data LUN's _and_ the VM OS images. You still need VMDK >>> pointer files to enable vmotion. >>> >>> By using RDM's for the VM images you can use lun cloning for R/W >>> snapshots within the same FlexVol, without paying for a FlexClone >>> license and achieve a reasonable level of deduplication without >>> paying >>> >>> for an A-SIS licence. i.e. create "golden images" as source LUNs >>> then use lun clones for the prod, dev, test and backup images of >>> the VM's. >>> >>> However, for larger environments using vmdk's on NAS from NFS >>> exports or CIFS shares for the VM images is a far more scalable >>> and manageable >>> >>> solution as long as you're happy to pay for A-SIS to manage >>> deduplication. I still wouldn't be putting data volumes inside >>> the vmdk's though. >>> >>> (Yes, you can use CIFS for vmdk's in ESX. Not as ideal as NFS for >>> a whole bunch of reasons, but it works via the VMWare ESX CIFS >>> driver and useful if you have to have CIFS for file server >>> consolidation and don't want to pay for NFS licencing as well). >>> >>> Either way, storing your application data in either RDM LUN's or >>> as direct NAS shares makes managing snapshots for dev/test copies >>> and online backups of the production data much simpler to manage. >>> Putting data inside vmdk's limits your functionality particularly >>> for backups. >>> >>> MSCS requires an RDM for the quorum disk, but you can still use a >>> NAS share for the data if your data profile and application >>> characteristics indicate that it makes more sense to do that than >>> use an RDM for the data. >>> >>> regards, >>> >>> Alan McLachlan. >>> >>> -----Original Message----- >>> From: owner-toasters(a)mathworks.com on behalf of M. Vaughn Stewart >>> Sent: Sun 10/28/2007 8:07 AM >>> To: Jack Lyons >>> Cc: Page, Jeremy; toasters(a)mathworks.com >>> Subject: Re: A-SIS questions >>> >>> FlexClone will clone a datastore, for VM level cloning granularity you >>> >>> gonna have to wait I think something is just around the corner. >>> >>> As for MSCS you need RDMs as VMDKs are not supported with MSCS >>> >>> RDMs can be either FCP or iSCSI >>> VMDKs can be on NFS or VMFS (which is over FCP or iSCSI) >>> >>> >>> >>> -- >>> Want an e-mail address like mine? >>> Get a free e-mail account today at www.mail.com! >>> >>> >>> >>> >>> _____________________________________________________________ >>> This e-mail (including all attachments) is confidential and may be >>> >> privileged. >> >>> It is for the exclusive use of the addressee only. If you are not >>> the addressee, you are hereby notified that any dissemination of >>> this communication is strictly prohibited. If you have received >>> this communication in error, please erase all copies of the >>> message and its >>> >>> attachments and notify us immediately at help(a)generalatlantic.com >>> <mailto:help@generalatlantic.com>. Thank You. >>> >> -- >> Want an e-mail address like mine? >> Get a free e-mail account today at www.mail.com! >> > > > > > -- Want an e-mail address like mine? Get a free e-mail account today at www.mail.com!

1 0

Windows SnapDrive Query
by Raj Patel 02 Nov '07

02 Nov '07

Simple question - We've just started to utilise some space on a second SAN head and for the life of me I can't remember what I needed to do on the SAN to allow our SnapDrive domain account to connect and create luns on the SAN. I've had a look 'useradmin user list' and 'useradmin domain list' but can't find anything obvious to match our domain 'sduser' account to a SAN account. When I use SnapDrive I get this in the event log - Error code : Access is denied. Verify that you have entered the correct domain and user name during SnapDrive installation and that this account has sufficient privileges to perform this operation. And this as a popup - SnapDrive error Error Description: User sduser does not have the capability to invoke API system-get-version I figure theres something really simple I'm missing or have taken for granted on our other SAN heads in terms of initial setup to use with SnapDrive. Cheers, Raj.

6 8

A-SIS questions
by Page, Jeremy 02 Nov '07

02 Nov '07

I have read some of the info on how A-SIS works but I have a couple of questions: 1) Assuming an average compression rate (say /home with random docs in it) how much of a performance hit does it impose 2) Does it change how much data is sent over the wire when you use SnapMirror/SnapVault? I.e. if I get 30% compression on a 1 gig snapshot do I send 1 gig or 700 mb to update my target? Oh, and for the folks who helped earlier, we are moving our Oracle and ESX systems to NetApp, the pSeries will be fibre connected, at least at first but the ESX stuff is all going on NFS. ~Jeremy This message (including any attachments) contains confidential and/or proprietary information intended only for the addressee. Any unauthorized disclosure, copying, distribution or reliance on the contents of this information is strictly prohibited and may constitute a violation of law. If you are not the intended recipient, please notify the sender immediately by responding to this e-mail, and delete the message from your system. If you have any questions about this e-mail please notify the sender immediately.

11 18

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