toasters April 2008

toasters@lists.teaparty.net

92 participants
54 discussions

Oddball SnapManager for SQL retention issue
by Raj Patel 15 Apr '08

15 Apr '08

Hi all, Is there a definitive command for snapshot deletion/retention - we've been struggling with this for awhile now and no combination of the '-D' or '-R' flag seems to do what we'd like it to do. For most our our SQL servers this isn't a real issue - once every few months we can manually remove old snaps; however we have a server that has a big database and a high-rate of change so we'd like to take snaps more often but lower the retention time - unfortunately as the deletions don't occur we fill the volume fairly quickly. The string we're using as a SQL 2005 agent job is similar to this - C:\Program Files\NetApp\SnapManager for SQL Server\SmSqlBi.exe -H "INFSQL1" -D 4 -R -G -S "INFSQL1" -C 8 "ArchiveManager" -Q "INFSQL1" -UM Has anyone else had similar issues or is there something I'm missing in the command string ? Alternatively is there a sdcli.exe / smsqlbi.execommand I can schedule to remove x number of old snaps ? Thanks in advance, Raj.

1 0

Wafl susp in a sim
by Nicholas Bernstein 15 Apr '08

15 Apr '08

I'm running the command "wafl_susp" and am getting the error: "found zero suspended messages" -- has anyone seen this before? Very strange. I tried this on a sim and a filer. Ontap 7.2.1. -- Nicholas Bernstein Technologist, Consultant, Instructor http://nicholasbernstein.com

6 5

FW: Serious A-sis bugs
by Freeman, Larry 14 Apr '08

14 Apr '08

Hi Stephan, regarding the issue you've described below, we are in the process of writing a Customer Service Bulletin and notifying users that all dedupe (A-SIS) systems require a mandatory upgrade to 7.2.4. I don't want to diminish the severity of this issue, but I think its important to point out that very few users have experienced panics as a result of this, but of course if you are one of those few that has...that is of little consolation. I apologize for the aggravation this issue caused you. NetApp felt it was very important to first identify the root cause and then define a corrective action plan before notifying our dedupe users. The corrective action plan will include the above mentioned upgrade to 7.2.4 and also tools for users to scan and repair any inconsistencies in their ACL or directory metadata without having to live through the kind of experience you were faced with. I'd also like to mention that no file data corruption occurs as a result of this issue, only inconsistencies between metadata and data blocks which can be detected and repaired with the maintenance tools we will provide. Again, not trying to diminish the seriousness of this issue, believe me it has a very high priority at NetApp. Notifications will begin next week. If anyone else on toasters has questions about this issue or would like more info, please contact NetApp Customer Support. Thanks, Larry Freeman NetApp > From: "Darragh, Stephen J \(CSC\) \(US SSA\)" > <stephen.j.darragh(a)baesystems.com> > Date: April 10, 2008 8:04:33 AM PDT > To: <toasters(a)mathworks.com> > Subject: Serious A-sis bugs > > If you are running a-sis with 7.2.3 or 7.2.3, there are a few bugs > that can cause the filer to panic and mark that volume offline forcing > a wafl_check on the containing aggregate. > Not only that, but we had to netboot a developer release and had > extensive outages. > > We have been told that 7.2.4 had fixed these bugs, but if > you have run > asis on an old version, the bug may be there dormant. > > I believe these were the bugs that were identified in our incident > report. > > bug 256799 > Bug 266312 > Bug 276084 > bug 266312 > bug 251673 > > The bug hit our primary storage, and it also his the VSM volumes on > our DR filer, and it would is in the ACL/directory of the backup tapes > as well. > > > > >

2 1

OFCS2 clones in Oracle RAC environments
by Daniel Keisling 11 Apr '08

11 Apr '08

Toasters, Is anyone using clones of OCFS2 filesystems in an Oracle RAC environment? I am testing cloning database LUNs to present back to the same server in order to accomplish zero downtime backup and ease of file-level and LUN-level recovery. The problem I am running into is that the UUID of the OCFS2 filesystem is cloned into the new filesystem (as expected), which causes the OCFS2 daemons to refuse to mount an already mounted filesystem (even though it's really not). While it's possible to change the UUID on OCFS2, it cannot be changed until all filesystems sharing that UUID are unmounted (and there is no "force" option). The only solutions I see so far are: 1) After the clone is taken, unmount the live database and change the UUID on the clone. Obviously this is probably not the wisest solution for high-availability databases. 2) Mount the clone on a Linux host not in this RAC environment. Any Linux host running the OFCS2 modules and on the SAN can be used. The post-processing by the DBA (copying the last ARC and control file) can be done via SCP to the mounted filesystem on the secondary host. autofs would also be a nice solution to enable access of the filesystem from the RAC environment. I'd rather not have to rely on a secondary server to accomplish this. 3) Edit the filesystem superblock of the clone to change the UUID. This is highly technical, but possible. If anyone has ever done this, let me know. I ran into the same problem with ASM (which does not let you change the label at all), which led me to try out OCFS2. I'm curious as to how other people are accomplishing ZDB and having a clone of an RAC-enabled database mounted on the same server. TIA, Daniel ______________________________________________________________________ This email transmission and any documents, files or previous email messages attached to it may contain information that is confidential or legally privileged. If you are not the intended recipient or a person responsible for delivering this transmission to the intended recipient, you are hereby notified that you must not read this transmission and that any disclosure, copying, printing, distribution or use of this transmission is strictly prohibited. If you have received this transmission in error, please immediately notify the sender by telephone or return email and delete the original transmission and its attachments without reading or saving in any manner.

1 0

nfsd.invalid.nsec.time
by Willeke, Jochen 11 Apr '08

11 Apr '08

Hello toasters, since some time we do see strange error messages in the logs of one filer. Here is the message: [nafiler: nfsd.invalid.nsec.time:warning]: Client 192.168.139.43 has an invalid nsec mtime (V3):2196581216 on inode:700615 The client is a SLES10 SP1 and has the kernel 2.6.16.46-0.12-smp. I have searched NOW and the web but did not find a proper explanation of what is causing this error messages :( We do see this message only for that client, never seen it for any other system. Best Regards Jochen -- Wincor Nixdorf International GmbH Sitz der Gesellschaft: Paderborn Registergericht Paderborn HRB 3507 Gesch�ftsf�hrer: Eckard Heidloff (Vorsitzender), Stefan Auerbach, Dr. J�rgen Wunram Vorsitzender des Aufsichtsrats: Karl-Heinz Stiller Steuernummer: 339/5884/0020 - Ust-ID Nr.: DE812927716 - WEEE-Reg.-Nr. DE44477193 Diese E-Mail enth�lt vertrauliche Informationen. Wenn Sie nicht der richtige Adressat sind oder diese E-Mail irrt�mlich erhalten haben, informieren Sie bitte sofort den Absender und vernichten Sie diese E-Mail. Das unerlaubte Kopieren sowie die unbefugte Weitergabe dieser E-Mail ist nicht gestattet. This e-mail may contain confidential information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden.

2 2

CIFS BUG!!! Solution ID: kb23196 (???)
by Milazzo Giacomo 09 Apr '08

09 Apr '08

Hi all A customer of ours migrating to a new NetApp all their shares (from a Samba server on Linux) got this well documented error, both from NetApp and Microsoft. Data Ontap is 7.2.4. The problem is that neither me or some collegues of mine have had this issue, from XP clients or other Windows OS!!! So, why this customer is getting this bug now? He ha also tried to change the registry key (its clients are all XP SP2) but nothing change! Any file system operation performer by an XP client need a refresh (F5) of the explorer... Any idea? NetApp says: Solution ID: kb23196 CIFS: Files created by Windows XP are not visible without pressing F5 Symptoms When a Windows XP workstation is mapping filer shares through a DFS root and a new file is added, the new file may not be visible in Windows Explorer until the user explicitly refreshes the window (in Windows Explorer by pressing F5). CIFS: Files created by Windows XP are not visible without pressing F5 Solution Microsoft says: http://support.microsoft.com/?scid=kb%3Ben-us%3B823291&x=22&y=7 Dott. Giacomo Milazzo Technical Account Manager Sinergy SpA Filiale di Roma + 00148. viale Castello della Magliana, 38 ' (+39) 3406001045 0665970252 7 +39 02 26922048 - g.milazzo(a)sinergy.it

3 3

UTF-8 appended to language code.
by Milazzo Giacomo 08 Apr '08

08 Apr '08

Hi all I need to append the .UTF-8 to the language code of some volume with a mixed security. I need to know if this can change or alter the name of the files there contained, accessing them both from NFS and CIFS. This because using Enterprise Vault SP2 on a NetApp share and placeholder (policy ad hoc for ev) the size of archived files reports "0" bytes instead of the real value reflecting the original data... My EV is 7.5 SP2 but I've only found a doc related to EV v.6 SP2 (http://seer.entsupport.symantec.com/docs/282785.htm) and I put here an extract of the readme attached to EV installation: Anyway supposing that they're right, can I add this UTF-8 and be sure? Thanks a lot NetApp Filer placeholder shortcuts appear to be 0 KB [Ref E1029910, E1017992, DOR1006] Placeholder shortcuts created on NetApp Filers fail to reflect the original file size, appearing instead to be 0 KB, if either of the following conditions applies: * The language setting on the NetApp Filer is Shift JIS (Japanese) * The language setting on the NetApp Filer is not appended with .UTF-8, and the name of the file that is converted to a placeholder contains Unicode characters. To avoid the second condition, you can append .UTF-8 to the language setting on the NetApp Filer. To do this, enter the following command from the NetApp Filer's administration console: vol lang volume_name language_code.UTF-8 where volume_name specifies the volume, and language_code is the code for the language setting. For example: vol lang vol0 en.UTF-8 Note: You can view a list of the language settings the NetApp Filer supports by entering the following command: vol lang

2 1

MPIO and path failure
by Jon Hill 08 Apr '08

08 Apr '08

We had an incident the other day where we lost one of our two paths to storage. We noticed the path failure quite by accident, and some time after it actually happened. Does anyone know whether there's a way that the NTAP fibre DSM can be configured to generate an alert (by SNMP or e-mail) whenever a path fails?

4 6

impact of setting snap reserve to 0
by Anna M 07 Apr '08

07 Apr '08

I have a volume with following details: [xxxhost ~yyy/zzz]# rsh xxxfiler df -h /vol/vol_1111_222 Filesystem total used avail capacity Mounted on /vol/vol_1111_222/ 453GB 422GB 31GB 93% /vol/vol_1111_222/ /vol/vol_1111_222/.snapshot 0GB 101GB 0GB ---% /vol/vol_1111_222/.snapshot when I clone a new flexclone volume using this volume and try to get the used space of that newlyc loned volume, I see that used space value changes drastically in couple of seconds We have noticed that with in 17 seconds the used space reported for this newly cloned volume was altered to the tune of 76 GBs (starting from ~387 GB to ~311 GB). Any clue as to why would this be heppening? Can it be because snap reserve is not set properly(we have set it to zero)? -- View this message in context: http://www.nabble.com/impact-of-setting-snap-reserve-to-0-tp16468497p164684… Sent from the Network Appliance - Toasters mailing list archive at Nabble.com.

8 11

Re: impact of setting snap reserve to 0
by Webster, Stetson 07 Apr '08

07 Apr '08

Hello Anna, I believe your snapshot is indeed zero (0) at first. Then as the data changes, your snapshot is holding on to it's read-only blocks. So it makes sense that the snapshot is growing. The more data you turn over, the bigger thw snapshot will be. A good test: Create a small volume, dump some data into it and then take a snapshot. Examine the snapshot carefully. Now completely overwrite (delete and then re-write) the data and then look at the same snapshot again. Theory: your snapshot keeps changing because your 'active filesystem' is changing. Cheers ...... Sent from my network synchronized laptop running Microsoft Outlook 2003 (www.microsoft.com) -----Original Message----- From: Anna M [mailto:annpurna@gmail.com] Sent: Monday, April 07, 2008 10:18 AM Pacific Standard Time To: toasters(a)mathworks.com Subject: Re: impact of setting snap reserve to 0 Since I am totally new to the world of filers, I would really appreciate if you can just explain me as to why the snapshot size for a newly created flexclone volume is not zero? After all the source for it also was a snapshot. The issues in space calculation for a newly clones volume is actually because of the changes in the snapshot size for this vol. For one of my newly cloned flexclone volume I am seeing that with in 3 minutes, the snapshot size reported is altered from /vol/prod_v/ 475777656KB 400203924KB 75573732KB 84% /vol/prod_v/ /vol/prod_v/.snapshot 0KB 74248140KB 0KB ---% /vol/prod_v/.sn to /vol/prod_v/ 475777656KB 325971708KB 149805948KB 69% /vol/prod_v/ /vol/prod_v/.snapshot 0KB 16112KB 0KB ---% /vol/prod_v/.snapshot Similarly the agg show_space output that I took at the same time as I took above df command output shows change in allocated and Used space values for this volume as: prod_v 75574680KB 32KB volume . ... prod_v 149823776KB 17816KB volume O/p for both these commands are still changing... We have found that values take approx 19 minutes to settle down...19 minutes seems to be a BIG duration :-( Any help in this regard will really be appreciated. -Anna bluezman wrote: > > Anna M wrote: >> snip >> > >> Any clue as to why would this be heppening? Can it be because snap >> reserve >> is not set properly(we have set it to zero)? >> > > i believe that when you take a clone, it takes a snapshot of the src, > since your snapshot space comes out of the actual volume, you will see > the size change > > correct me if i am wrong. > > -- View this message in context: http://www.nabble.com/impact-of-setting-snap-reserve-to-0-tp16468497p165353… Sent from the Network Appliance - Toasters mailing list archive at Nabble.com. DISCLAIMER: This email and any attachments are confidential and privileged. It is intended solely for the addressee. Access to this email by anyone else is unauthorized. If received in error, do not open, read, copy, disclose, use or store it in any way, or permit others to do so but please notify the sender immediately and delete the original email and attachments from your machine. If you are not the addressee or the intended recipient, any use, disclosure, copying, distribution or action taken in reliance of it, is prohibited and unlawful. The information, attachments, opinions or advice contained in this email are not necessarily the views or opinions of Celtel Nigeria Limited or its affiliates and without more, shall not commit Celtel Nigeria Limited or its affiliates. Celtel Nigeria Limited and its affiliates do not accept liability for claims, losses, or damages caused by the inaccuracy, incorrectness, or lack of integrity of this email or any attachments.

1 0

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 2008