UNSOLVED

CWFNetman

updated

11 years ago

C

CWFNetman

1 Rookie

•

13 Posts

0

3793

October 26th, 2015 10:00

VNXe3200 - Cannot delete snapshots

snap_error.jpg

Getting this error trying to delete snapshots taken by NDMP.

How do I delete them?

  • Rainer_EMC

    6 Operator

    •

    8645 Posts

    2829

    0

    Posted October 27th, 2015 14:00

    they should get auto-deleted after the NDMP backup is done.

    If your NDMP backup is finished and the snapshot is there please contact EMC support

  • CWFNetman

    1 Rookie

    •

    13 Posts

    2809

    0

    Posted October 28th, 2015 10:00

    Well, they're *supposed* to get auto-deleted after the backup job has run.

    Problem is sometimes they are getting left behind as orphans.

    Just spent nearly 2 hours on the phone & WebEx session with EMC support and apparently there's a problem in the VNXe3200's operating environment code and the issue is being escalated to development for a fix.

    The support engineer was finally able to manually delete the orphaned snapshots via commands in the dartconsole.

  • Daya1991

    5 Posts

    2581

    0

    Posted February 7th, 2020 19:00

    @CWFNetman 

    Do you have procedure how  support engineer was able to manually delete the orphaned snapshots via commands in the dartconsole. I am looking for help as I dont have EMC support on my VNXe array.

    Thanks ,

    Dayanand

  • DELL-Sam L

    Community Manager

    •

    8148 Posts

    •

    34709 Points

    2558

    0

    Posted February 10th, 2020 10:00

    Hello Daya1991,

    Here is a link to a knowledge base article that may help you resolve your issue.  https://dell.to/2uraQZk

    Please let us know if you have any other questions.

  • Daya1991

    5 Posts

    2043

    0

    Posted July 19th, 2020 06:00

    @DELL-Sam L 

     

    I am unable to open the shared link can you please share any alternate link to view this article. Appreciate your reply.

    Regards,

    Dayanand

  • DELL-Sam L

    Community Manager

    •

    8148 Posts

    •

    34709 Points

    2017

    0

    Posted July 20th, 2020 10:00

    Hello Daya1991,

    Here is the article.

     

    Issue

    When using the VNX Snapshots auto-delete function and if the Pool that contains the snaps is very full, the auto-delete can impact the performance of both auto-tiering completion and general performance issues on the array.


    The following errors will be in the Event log:

    MLU    712d8106     Could not complete operation Destroy Snapshot due to 0xE16D8018. Operation Status: 0xE16D8018, Commit Status: N/A
    0x716D8018     "The operation failed because the specified snapshot is currently being destroyed" message

    VNX Snapshot "Destroying_20130429160931.558+00-000".
    You may receive this error when the VNX Snap is still in the  Delete  stage and cannot be relocated by auto-tiering until the delete is complete.

     

    Cause

    The VNX snapshot Destroy is a background operation. If you are using low performance, high capacity NL-SAS disk to support the VNX Snap feature, you may experience this issue.
    The slower the physical drive, the more time will be required to complete the "Destroy" operation. 3-TB NL-SAS drives may not be able provide the level of performance needed.
    The biggest limiting factor on the speed of the snapshot destroys is the back-end disk speed as the snap truncate/destroy are very I/O intensive (see article 172629 ).

     

    Change

    VNX Snapshots deleting makes pool LUNs very busy.

     

    Resolution

    Configure the Pool to have sufficient free space for the VNX Snaps and the type and speed of drives that can handle the extra workload required for auto-delete.  

    The following steps will help with this issue.

    Execute the following steps to improve the situation:
     
    1. Set all snaps to not 'auto-expire'. This can be done via the GUI or via the cli. 
       - Use CLI command "naviseccli snap -modify -id "VNX snapid" -keepFor 0"
    When the pool starts auto-deleting all the snaps, the pool starts running out of space, this again places a large BE load on the array.

     2. A longer term solution would be to expand the affected pool with faster disks [SSD/EFD drives], however if the Auto Tiering window is very large, this would NOT be a good long term solution.

    3. Another 'medium' term solution would be to migrate the source LUNs to new LUNs in a different pool [user will need to create new LUNs] made up of mix drives [SSD & SAS], and then destroy the existing LUNs.

    4. Check to see if MV/A is used and if the RLP are also designated to disks in the affected pool.

    5. See article 172629 for corrective actions that can be taken.

  • Daya1991

    5 Posts

    1999

    1

    Posted July 21st, 2020 00:00

    @DELL-Sam L 

    Hi Dell-Sam L 

    Is this article applies for the VNXe3200 system as well ? I dont have active support with EMC also I dont have root credentials to execute advanced commands, The only option I have is to perform the SP failover  to delete this snapshot automatically . It would be great if you share some additional options to solve this error which is blocking NDMP session in VNXe due to which my NDMP backup is failing .

     

  • DELL-Sam L

    Community Manager

    •

    8148 Posts

    •

    34709 Points

    1980

    0

    Posted July 21st, 2020 16:00

    Hello Daya1991,

    What is the current version of your OE on your VNXe?

  • Daya1991

    5 Posts

    1793

    0

    Posted September 12th, 2020 06:00

    @DELL-Sam L 

    my current OE version is 3.1.8.9340299

  • DELL-Sam L

    Community Manager

    •

    8148 Posts

    •

    34709 Points

    398

    0

    Posted September 14th, 2020 16:00

     Hello Daya1991,

    Do you have auto snapshot configured?