Rollforward recovery algorithm...

Collapse
This topic is closed.
X
X
 
  • Time
  • Show
Clear All
new posts
  • Raquel

    Rollforward recovery algorithm...

    I have asked a similar question before, but am still not very clear on
    this......expec ting some insights on this one....

    Ok, so these are the two pieces of information regarding archived logs
    during a rollforward recovery in a recoverable database:

    1. OVERFLOWLOGPATH configuration parameter can specify a location for
    DB2 to find log files that are needed for rollforward operations.

    2. The database configuration file parameter 'userexit' specifies
    whether the database manager invokes a user exit program to retrieve
    log
    files during rollforward recovery of databases.

    So, it seems that the algorithm during a rollforward recovery is:

    1. Check the active log path directory for the required logs. If logs
    are not found here, go to step 2.

    2. Check to see if OVERFLOWLOGPATH db. Config. Parm. is specified (or
    OVERFLOW LOG PATH parameter is specified in the ROLLFORWARD DATABASE
    command).
    a. If OVERFLOWLOGPATH is specified, see if the required log files
    exist
    there. If log files are not present there then check if 'userexit' is
    enabled.
    i. If 'userexit' is enabled, invoke the userexit to find the location
    of the log files. If log found in the location given by userexit,
    continue with the rollforward operation else give the 'log not found'
    error.

    ii. If 'userexit' is diabled, give the 'log not found' error.

    b. If OVERFLOWLOGPATH is not specified, check if 'userexit' is
    enabled.

    i. If 'userexit' is enabled, invoke the userexit to find the location
    of the log files. If log found in the location given by userexit,
    continue with the rollforward operation else give the 'log not found'
    error.
    ii. If 'userexit' is diabled, give the 'log not found' error.

    So, basically, the sequence is: active log path directory,
    OVERFLOWLOGPATH directory and lastly the 'userexit'.

    Is my understanding correct or flawed.

    TIA
    Raquel.
  • Pierre Saint-Jacques

    #2
    Re: Rollforward recovery algorithm...

    The sequence you use at the end of your note is right.
    One more point.
    The overflow logpath parm. must b specified if you logpath is defined over
    raw devices. The logpath will be checked, if log file is not there, the
    overflow is checked; if not found againand the exit is used (which in thic
    case should always be) then the FILES stored in the archive can ONLY be
    retrieved to the overflow logpath as the exit is not able to deal with raw
    devices.
    HTH, Pierre.
    "Raquel" <raquel_rodrigu ezus@yahoo.com> a écrit dans le message de
    news:9a73b58d.0 401152234.12452 771@posting.goo gle.com...[color=blue]
    > I have asked a similar question before, but am still not very clear on
    > this......expec ting some insights on this one....
    >
    > Ok, so these are the two pieces of information regarding archived logs
    > during a rollforward recovery in a recoverable database:
    >
    > 1. OVERFLOWLOGPATH configuration parameter can specify a location for
    > DB2 to find log files that are needed for rollforward operations.
    >
    > 2. The database configuration file parameter 'userexit' specifies
    > whether the database manager invokes a user exit program to retrieve
    > log
    > files during rollforward recovery of databases.
    >
    > So, it seems that the algorithm during a rollforward recovery is:
    >
    > 1. Check the active log path directory for the required logs. If logs
    > are not found here, go to step 2.
    >
    > 2. Check to see if OVERFLOWLOGPATH db. Config. Parm. is specified (or
    > OVERFLOW LOG PATH parameter is specified in the ROLLFORWARD DATABASE
    > command).
    > a. If OVERFLOWLOGPATH is specified, see if the required log files
    > exist
    > there. If log files are not present there then check if 'userexit' is
    > enabled.
    > i. If 'userexit' is enabled, invoke the userexit to find the location
    > of the log files. If log found in the location given by userexit,
    > continue with the rollforward operation else give the 'log not found'
    > error.
    >
    > ii. If 'userexit' is diabled, give the 'log not found' error.
    >
    > b. If OVERFLOWLOGPATH is not specified, check if 'userexit' is
    > enabled.
    >
    > i. If 'userexit' is enabled, invoke the userexit to find the location
    > of the log files. If log found in the location given by userexit,
    > continue with the rollforward operation else give the 'log not found'
    > error.
    > ii. If 'userexit' is diabled, give the 'log not found' error.
    >
    > So, basically, the sequence is: active log path directory,
    > OVERFLOWLOGPATH directory and lastly the 'userexit'.
    >
    > Is my understanding correct or flawed.
    >
    > TIA
    > Raquel.[/color]

    Comment

    • Raquel

      #3
      Re: Rollforward recovery algorithm...

      Thank you so much Pierre for your response. Your responses are always appreciated.

      Raquel.

      "Pierre Saint-Jacques" <sescons@invali d.net> wrote in message news:<XCSNb.757 56$ld7.862565@w agner.videotron .net>...[color=blue]
      > The sequence you use at the end of your note is right.
      > One more point.
      > The overflow logpath parm. must b specified if you logpath is defined over
      > raw devices. The logpath will be checked, if log file is not there, the
      > overflow is checked; if not found againand the exit is used (which in thic
      > case should always be) then the FILES stored in the archive can ONLY be
      > retrieved to the overflow logpath as the exit is not able to deal with raw
      > devices.
      > HTH, Pierre.
      > "Raquel" <raquel_rodrigu ezus@yahoo.com> a écrit dans le message de
      > news:9a73b58d.0 401152234.12452 771@posting.goo gle.com...[color=green]
      > > I have asked a similar question before, but am still not very clear on
      > > this......expec ting some insights on this one....
      > >
      > > Ok, so these are the two pieces of information regarding archived logs
      > > during a rollforward recovery in a recoverable database:
      > >
      > > 1. OVERFLOWLOGPATH configuration parameter can specify a location for
      > > DB2 to find log files that are needed for rollforward operations.
      > >
      > > 2. The database configuration file parameter 'userexit' specifies
      > > whether the database manager invokes a user exit program to retrieve
      > > log
      > > files during rollforward recovery of databases.
      > >
      > > So, it seems that the algorithm during a rollforward recovery is:
      > >
      > > 1. Check the active log path directory for the required logs. If logs
      > > are not found here, go to step 2.
      > >
      > > 2. Check to see if OVERFLOWLOGPATH db. Config. Parm. is specified (or
      > > OVERFLOW LOG PATH parameter is specified in the ROLLFORWARD DATABASE
      > > command).
      > > a. If OVERFLOWLOGPATH is specified, see if the required log files
      > > exist
      > > there. If log files are not present there then check if 'userexit' is
      > > enabled.
      > > i. If 'userexit' is enabled, invoke the userexit to find the location
      > > of the log files. If log found in the location given by userexit,
      > > continue with the rollforward operation else give the 'log not found'
      > > error.
      > >
      > > ii. If 'userexit' is diabled, give the 'log not found' error.
      > >
      > > b. If OVERFLOWLOGPATH is not specified, check if 'userexit' is
      > > enabled.
      > >
      > > i. If 'userexit' is enabled, invoke the userexit to find the location
      > > of the log files. If log found in the location given by userexit,
      > > continue with the rollforward operation else give the 'log not found'
      > > error.
      > > ii. If 'userexit' is diabled, give the 'log not found' error.
      > >
      > > So, basically, the sequence is: active log path directory,
      > > OVERFLOWLOGPATH directory and lastly the 'userexit'.
      > >
      > > Is my understanding correct or flawed.
      > >
      > > TIA
      > > Raquel.[/color][/color]

      Comment

      Working...