Forum Discussion

Claudio_Veronez's avatar
Claudio_Veronez
Level 6
16 years ago
Solved

Great Catalog problem / daylight saving error

Hello


I think I´m w/ a great problem in my hands

there was a script ntpdate to automatically change the time using the AD server and stuff

I don´t know why the time was set to 03/20/10

I have manualy  changed the server time  and didn´t note about the script that changed again to the 03/20/10 date,

so netbackup runs for 3 days w/ that date. 

now the date is ok

Problem 1:
Some jobs R not running automaticlly

problem 2:

 I think that I´ll have to "delete "  those backups that runs w/ wrong date.

Problem 3:

even the catalog backup isn´t running automatically (and it is set to run ever 2 hours)



someone can help me please?


That is a real problem isn´t it /??


Thanks for any adivice 





7.1.x and Earlier
Backup and Recovery
Configuring
NetBackup
Recovering
Solaris
  • Andy_Welburn's avatar
    Andy_Welburn
    16 years ago
    Frequency Schedule

    Especially the nbpemreq -predict_all stuff to see when it expects to run its next jobs & removal/renaming of STREAMS files (maybe just test on one to start - also only relevant if multistreaming is on). Switching multistreaming off may have the same effect i.e. give it a 'kick' - next backup will be a full & can then switch on again.

    There must be a more obvious way to 'reset' but I can't think of it - short of copying all your Policies to new which seems a bit much!

    As a matter of interest - what does the BAR GUI show for available restores - are they dated March? If so, maybe you will have to expire all the "future" backups?

    Sorry I can't be more help, but it's not something you come across everyday! Maybe your best recourse would be to log a call & see what Symantec suggest? Or wait a few hours until the rest of the world wakes up & starts to log back into the forum?! :D

15 Replies

Sort By
  • Claudio_Veronez's avatar
    Claudio_Veronez
    Level 6
    16 years ago
    so almost no policy is starting automatically

    I have to  start manually even the archive log policy.


    Any Idea??

     
  • Andy_Welburn's avatar
    Andy_Welburn
    Level 6
    16 years ago
    but not necessarily why some are running!

    If your server has had an incorrect date set in the future and then backups have run, then they will probably be waiting until they have caught up before they run again. e.g. incorrect date on server of 20th March, job runs with a frequency of 1 day, next due time therefore is 21st March  -  possibly!

    What's the difference between those jobs that are running & those that aren't? Calendar & frequency? Non-multi-streamed & multi-streamed?

    How many policies/schedules are affected? Just wondering if it's worth making a minor change to each just to try & give it a 'kick'. e.g. delete a backup window & reinstate, remove multi-streaming & re-adding (may make next save a FULL) etc. If there are too many policies then this may not be feasible for you.

  • Claudio_Veronez's avatar
    Claudio_Veronez
    Level 6
    16 years ago
    So Andy,

    I think that the policies that R running ok, wan't active when the date was Wrong.


    I have tryed to get some archive policies that ran hourlly, copy to new, to a temp policy, remove the policy w/ problem, and copy to new again the temp policy to the name of the policy before.


    and it doesn't

    is there some way of removing that backup from the future  from the catalog (that cam become a movie haha)


    I had to run some policy manually, it ran ok, status 0, the problem must be that thing U mention,
    next backup 03/21

    any idea of how pass throught that problem?


  • Claudio_Veronez's avatar
    Claudio_Veronez
    Level 6
    16 years ago
    just a try,

    deactivete ALL policy,
    put the date wrong again? is there any way of removing ahh jobs/backups/images/tapes everything?

    and back tho the past?


  • Andy_Welburn's avatar
    Andy_Welburn
    Level 6
    16 years ago
    Frequency Schedule

    Especially the nbpemreq -predict_all stuff to see when it expects to run its next jobs & removal/renaming of STREAMS files (maybe just test on one to start - also only relevant if multistreaming is on). Switching multistreaming off may have the same effect i.e. give it a 'kick' - next backup will be a full & can then switch on again.

    There must be a more obvious way to 'reset' but I can't think of it - short of copying all your Policies to new which seems a bit much!

    As a matter of interest - what does the BAR GUI show for available restores - are they dated March? If so, maybe you will have to expire all the "future" backups?

    Sorry I can't be more help, but it's not something you come across everyday! Maybe your best recourse would be to log a call & see what Symantec suggest? Or wait a few hours until the rest of the world wakes up & starts to log back into the forum?! :D
  • Claudio_Veronez's avatar
    Claudio_Veronez
    Level 6
    16 years ago
    Don't Worry Andy, U R being very helpfull for sure.


    the activivity monitor is showing the Mars jobs, and w/ the JobID lower than that runs today

    no multistreaming were set.



  • Claudio_Veronez's avatar
    Claudio_Veronez
    Level 6
    16 years ago
    Bingo Andy, the nbpemreq shows that DUE TIME column w/ date set for Mar.


    I 'll try the

    nbpemreq -updatepolicies,

    that Marianne seid at the other post
    ]

    I have moved the /db/images/"STREAM"  w/ mars date, to a TEMP directory.

    restarting the service ...

    now the predict has no job

    I will run manually a archive backup to see if something appears for the next hour..



  • Claudio_Veronez's avatar
    Claudio_Veronez
    Level 6
    16 years ago
    Nothing,

    maybe I have to reset ALL that predict_all shows,,

    Do You know if  there is a way of doing that?