Forum Discussion

candelorom's avatar
candelorom
Level 3
3 years ago

AIR & Image on Media

Goodmorning everyone,
I inherited a NetBackup 8.2 structure with 2 master servers and 2 DataDomain configured in AIR.

Some time ago the password of the DataDomain was changed only on the source master, so the target master server stopped working: all the import jobs failed.
Currently I have solved the situation, so I am able to import correctly.

I am carrying out some verifications and I have noticed that on the target master server, there are some expired backup images. I noticed this through the GUI: NetBackup Management> Reports> Images on Media.

In the target master server, I have images that have expired between 2020 and 05/10/2021. While in the source master server I have the backup images which will expire from now on.

On the target master server, I have already run the Image Cleanup, but to no avail. How is this possible?

I also noticed that the same backup images have different retention between the source master server and the target master server.

Eg:
backup id XXXXXXXXXXXX_XXXXXXXXXX has a retention of 2 years on the source master server and 1 week on the target master server. On both, however, the expiration date is the same.

I have checked the SLP config and they are in sync. They are automatically created on the target master server with the same retention.

Could you kindly explain to me what could have happened?

Many thanks in advance,
Matteo

 

Other
  • davidmoline's avatar
    davidmoline
    3 years ago

    Hi candelorom 

    The only issue I see (and cannot explain) is that the original report for the image on the target system is showing a retention period of 1 week (although based on the imported image information FRAG fields, the retention level here (field 19) is importing as 0 rather than retention level + 65536 - this may be a feature of SLP imports, I'm not sure. On the original backup this is set to 65549 (which is indicating 2 year retention). 

    The source SLP sets a target retention of 2 years for the replication operation, and the import operation on the target SLP uses this retention. 

    So there is nothing really wrong as such. 

    David

4 Replies

Sort By
  • davidmoline's avatar
    davidmoline
    Level 6
    3 years ago

    Hi candelorom 

    Interesting problem. Can you provide details on the SLP used to perform the replication and import (e..g nbstl <SLP> -L), and bpimagelist output for the backupID on both the source and target (bpimagelist -backupid <BID>)

    I suspect that the SLP hasn't completed hence why it is not being expired as you expect. Did you suspend or inactive SLP's while debugging your DD issue?

    You could also look at the output from the nbstlutil command to see if this helps - some commands to try are:

    • nbstlutil stlilist -backupid -U (can be done on both source and target)
    • nbstlutil list -backupid -U
    • nbstlutil report

    One last thing to check although I don't think this would cause the behaviour you describe is that there is no NOexpire flag in the NetBackup bin directory (/usr/openv/netbackup/bin or <INSTALLPATH>\Veritas\NetBackup\bin) This file will stop image expiration.

    Cheers
    David

    • candelorom's avatar
      candelorom
      Level 3
      3 years ago

      Hi davidmoline,
      Thank you for your answer.

      I have already checked if the NOexpire file was present in the path you indicated and does not actually exist.

      I don't know if any SLPs have been suspended or disabled in the past, because it's an infrastructure I've been managing for a few months. Certainly some SLPs have been deleted from the source master. In fact in the destination master I have 75 SLP and in the source master there are only 64 SLPs. As far as you know, the SLPs deleted from the source master will be automatically deleted from the destination master? The creation of the import SLPs occurs automatically as configured in the master settings.

      I will provide you with the output of the commands required to investigate and solve my problem.

      Thank you for your support,
      Matteo

    • candelorom's avatar
      candelorom
      Level 3
      3 years ago

      Good morning davidmoline 
      these are the outputs you asked me:

      The first image is always from the source master server. The second image is of the target master server

      - nbstl <SLP> -L

       

       

      - bpimagelist -backupid <BID>

       

      - nbstlutil stlilist -backupid -U

      - nbstlutil list -backupid -U

      - nbstlutil report

      Thanks again for your support,
      Matteo

      • davidmoline's avatar
        davidmoline
        Level 6
        3 years ago

        Hi candelorom 

        The only issue I see (and cannot explain) is that the original report for the image on the target system is showing a retention period of 1 week (although based on the imported image information FRAG fields, the retention level here (field 19) is importing as 0 rather than retention level + 65536 - this may be a feature of SLP imports, I'm not sure. On the original backup this is set to 65549 (which is indicating 2 year retention). 

        The source SLP sets a target retention of 2 years for the replication operation, and the import operation on the target SLP uses this retention. 

        So there is nothing really wrong as such. 

        David