• Re: [Info-ingres] Recover Database

    From Martin Bowes@21:1/5 to Rich Ford on Tue Jan 25 14:11:04 2022
    To: info-ingres@lists.planetingres.org (info-ingres@lists.planetingres.org)

    Hi Rich,

    Short answer: Don't.

    This is not the right way to do it as there are huge problems in getting the database to believe the checkpoint file is cool. You could end up with a very scrambled database configuration file

    This is not what a checkpoint file is for. A checkpoint file is created so that you can do time based recoveries using th edatabsse journals.

    What you want is a static backup created with either unloaddb or copydb. Those are designed for exactly this sort of task.

    Marty

    -----Original Message-----
    From: Rich Ford <richford990@gmail.com>
    Sent: 25 January 2022 14:02
    To: info-ingres@lists.planetingres.org
    Subject: [Info-ingres] Recover Database

    Is it possible to save a checkpoint file in a backup directory then destroy the database, create the database and then use this checkpoint file to recover the database the state prior to the destroy?

    Rich
    _______________________________________________
    Info-ingres mailing list
    Info-ingres@lists.planetingres.org https://lists.planetingres.org/mailman/listinfo/info-ingres

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Rich Ford@21:1/5 to Martin Bowes on Tue Jan 25 06:16:22 2022
    On Tuesday, 25 January 2022 at 06:12:09 UTC-8, Martin Bowes wrote:
    Hi Rich,

    Short answer: Don't.

    This is not the right way to do it as there are huge problems in getting the database to believe the checkpoint file is cool. You could end up with a very scrambled database configuration file

    This is not what a checkpoint file is for. A checkpoint file is created so that you can do time based recoveries using th edatabsse journals.

    What you want is a static backup created with either unloaddb or copydb. Those are designed for exactly this sort of task.

    Marty
    -----Original Message-----
    From: Rich Ford <richf...@gmail.com>
    Sent: 25 January 2022 14:02
    To: info-...@lists.planetingres.org
    Subject: [Info-ingres] Recover Database

    Is it possible to save a checkpoint file in a backup directory then destroy the database, create the database and then use this checkpoint file to recover the database the state prior to the destroy?

    Rich
    _______________________________________________
    Info-ingres mailing list
    Info-...@lists.planetingres.org https://lists.planetingres.org/mailman/listinfo/info-ingres

    Thanks for the quick reply.

    Rich

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Karl Schendel@21:1/5 to Ingres and related product discussi on Tue Jan 25 09:49:49 2022
    On Jan 25, 2022, at 9:01 AM, Rich Ford <richford990@gmail.com> wrote:

    Is it possible to save a checkpoint file in a backup directory then destroy the database, create the database and then use this checkpoint file to recover the database the state prior to the destroy?


    Possible, yes. It's not the simplest thing in the world to get right,
    and you can't just rollforwarddb the checkpoint. I don't advise it
    unless there is no other way to recover, and then I'd want to get
    Tech Support involved.

    Karl

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Adrian Williamson@21:1/5 to Ingres and related product discussi on Tue Jan 25 15:13:48 2022
    Ah yes, the dreaded database identifier and binary editing of database
    (CNF?) files....keep you all night long that will.


    -----Original Message-----
    From: info-ingres-bounces@lists.planetingres.org <info-ingres-bounces@lists.planetingres.org> On Behalf Of Karl Schendel
    Sent: 25 January 2022 14:50
    To: Ingres and related product discussion forum <info-ingres@lists.planetingres.org>
    Subject: Re: [Info-ingres] Recover Database


    On Jan 25, 2022, at 9:01 AM, Rich Ford <richford990@gmail.com> wrote:

    Is it possible to save a checkpoint file in a backup directory then
    destroy the database, create the database and then use this checkpoint
    file to recover the database the state prior to the destroy?


    Possible, yes. It's not the simplest thing in the world to get right, and
    you can't just rollforwarddb the checkpoint. I don't advise it unless there
    is no other way to recover, and then I'd want to get Tech Support involved.

    Karl

    _______________________________________________
    Info-ingres mailing list
    Info-ingres@lists.planetingres.org https://lists.planetingres.org/mailman/listinfo/info-ingres

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Paul White@21:1/5 to Ingres and related product discussi on Wed Jan 26 12:00:06 2022
    This work ok for replication.

    I drop a production database, which removes ckp jnl dmp tmp data. (Non x100 db).

    Then:
    Createdb
    Ckpdb +j
    Rsync from DR server:ckp dmp jnl
    Copy aaaaaaaa.* from dmp to data
    Infodb to make sure
    Rollforwarddb to point in time
    Ckpdb +j

    No editing configs or iidbdb required.

    Am i missing something?



    On Wed, 26 Jan 2022, 2:25 am Roy Hann, <specially@processed.almost.meat>
    wrote:

    Rich Ford wrote:

    Is it possible to save a checkpoint file in a backup directory
    then destroy the database, create the database and then use this
    checkpoint file to recover the database the state prior to the
    destroy?

    Like everyone else has said, I would discourage that approach.
    The preferred tool for that situation is unloaddb.

    That said, your disaster recovery plan should include all the steps
    required to do just what you suggest. In the event of a total loss of
    the facility (fire, flood, carpet moths) you need to be able to
    reinstate your database on a new machine using the checkpoint.

    I get the impression you are new to the site. You should ask if
    anyone else there knows where the disaster plan is.

    If they don't, your next step is to become a hero by writing it.

    Roy
    _______________________________________________
    Info-ingres mailing list
    Info-ingres@lists.planetingres.org https://lists.planetingres.org/mailman/listinfo/info-ingres


    <div dir="auto">This work ok for replication.<div dir="auto"><br></div><div dir="auto">I drop a production database, which removes ckp jnl dmp tmp data. (Non x100 db).</div><div dir="auto"><br></div><div dir="auto">Then:</div><div dir="auto">Createdb</
    <div dir="auto">Ckpdb +j</div><div dir="auto">Rsync from DR server:ckp dmp jnl</div><div dir="auto">Copy aaaaaaaa.* from dmp to data</div><div dir="auto">Infodb to make sure</div><div dir="auto">Rollforwarddb to point in time</div><div dir="auto">
    Ckpdb +j</div><div dir="auto"><br></div><div dir="auto">No editing configs or iidbdb required.</div><div dir="auto"><br></div><div dir="auto">Am i missing something?</div><div dir="auto"><br></div><div dir="auto"><br></div></div><br><div class="gmail_
    quote"><div dir="ltr" class="gmail_attr">On Wed, 26 Jan 2022, 2:25 am Roy Hann, &lt;specially@processed.almost.meat&gt; wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Rich Ford wrote:<


    &gt; Is it possible to save a checkpoint file in a backup directory<br>
    &gt; then destroy the database, create the database and then use this<br>
    &gt; checkpoint file to recover the database the state prior to the<br>
    &gt; destroy?<br>

    Like everyone else has said, I would discourage that approach.<br>
    The preferred tool for that situation is unloaddb.<br>

    That said, your disaster recovery plan should include all the steps<br> required to do just what you suggest. In the event of a total loss of<br>
    the facility (fire, flood, carpet moths) you need to be able to<br>
    reinstate your database on a new machine using the checkpoint. <br>

    I get the impression you are new to the site. You should ask if<br>
    anyone else there knows where the disaster plan is. <br>

    If they don&#39;t, your next step is to become a hero by writing it.<br>

    Roy<br>
    _______________________________________________<br>
    Info-ingres mailing list<br>
    <a href="mailto:Info-ingres@lists.planetingres.org" target="_blank" rel="noreferrer">Info-ingres@lists.planetingres.org</a><br>
    <a href="https://lists.planetingres.org/mailman/listinfo/info-ingres" rel="noreferrer noreferrer" target="_blank">https://lists.planetingres.org/mailman/listinfo/info-ingres</a><br>
    </blockquote></div>

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Roy Hann@21:1/5 to you on Wed Jan 26 13:12:03 2022
    Wednesday, January 26, 2022, 2:00:06 AM, you wrote:

    This work ok for replication.

    I drop a production database, which removes ckp jnl dmp tmp data. (Non x100 db).

    Then:
    Createdb
    Ckpdb +j
    Rsync from DR server:ckp dmp jnl
    Copy aaaaaaaa.* from dmp to data
    Infodb to make sure
    Rollforwarddb to point in time
    Ckpdb +j

    No editing configs or iidbdb required.

    Am i missing something?

    I would update iidatatabase.db_id in iidbdb to make it the database ID
    of the old database. It's not a big deal but it means IPM will show
    the correct database name.

    And for DR purposes I'd make sure I was keeping all my Ingres user IDs
    from iidbdb, and also my area and location definitions so they could
    be recreated after the smoke clears.

    Roy

    On Wed, 26 Jan 2022, 2:25 am Roy Hann,
    <specially@processed.almost.meat> wrote:

    Rich Ford wrote:

    Is it possible to save a checkpoint file in a backup directory
    then destroy the database, create the database and then use this
    checkpoint file to recover the database the state prior to the
    destroy?

    Like everyone else has said, I would discourage that approach.
    The preferred tool for that situation is unloaddb.

    That said, your disaster recovery plan should include all the steps
    required to do just what you suggest. In the event of a total loss of
    the facility (fire, flood, carpet moths) you need to be able to
    reinstate your database on a new machine using the checkpoint.

    I get the impression you are new to the site. You should ask if
    anyone else there knows where the disaster plan is.

    If they don't, your next step is to become a hero by writing it.

    Roy
    _______________________________________________
    Info-ingres mailing list
    Info-ingres@lists.planetingres.org
    https://lists.planetingres.org/mailman/listinfo/info-ingres

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)