modifed timestamp of a table

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Sumanth - 06 Jun 2006 21:06 GMT

Hi,

Information with regard to creation of a table is stored in SYSTABLES.
Is modified time stamp of a table stored in
DB2, if so where.

Thanks,
Sumanth

Reply to this Message

Sumanth - 06 Jun 2006 21:19 GMT

By modified time stamp I mean "time when the content of the table changes"

Thanks,
Sumanth

> Hi,
>
[quoted text clipped - 4 lines]
> Thanks,
> Sumanth

Reply to this Message

Shashi Mannepalli - 07 Jun 2006 03:20 GMT

No this information is not stored in any of the CATALOG tables.
But all the changes will be LOGGED. You can use db2urlog or something
like that
to read those LOGS and check for a particular table....

This entire process is complex...

cheers...
Shashi Mannepalli

> Hi,
>
[quoted text clipped - 4 lines]
> Thanks,
> Sumanth

Reply to this Message

Knut Stolze - 07 Jun 2006 11:37 GMT

> Hi,
>
> Information with regard to creation of a table is stored in SYSTABLES.
> Is modified time stamp of a table stored in

You should use only the views in the SYSCAT schema, i.e. SYSCAT.TABLES. All
tables in the SYSIBM schema are not documented and they may change at any
time, breaking your application if you rely on them.

> DB2, if so where.

The actual timestamp of the last data modification is not automatically
tracked. This would be a huge bottle-neck performance-wise if it were
done.

Signature

Knut Stolze
DB2 Information Integration Development
IBM Germany

Reply to this Message

aj - 07 Jun 2006 15:15 GMT

...snip...

> The actual timestamp of the last data modification is not automatically
> tracked. This would be a huge bottle-neck performance-wise if it were
> done.

But its been requested over and over and over again. Seems like those
smart IBM engineers could figure out some way to do it w/o screwing up
performance.

aj

Reply to this Message

Gregor Kovač - 07 Jun 2006 15:47 GMT

> ...snip...
>>
[quoted text clipped - 7 lines]
>
> aj

And I'd only need the time of last ALTER TABLE :))

Signature

-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~

| Gregor Kovac | Gregor.Kovac@mikropis.si |

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

| In A World Without Fences Who Needs Gates? |
| Experience Linux. |

-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~

Reply to this Message

Knut Stolze - 09 Jun 2006 19:06 GMT

>> ...snip...
>>>
[quoted text clipped - 5 lines]
>> smart IBM engineers could figure out some way to do it w/o screwing up
>> performance.

As soon as someone implements a general purpose sorting mechanism that runs
faster than O(n log n), we can probably discuss the tuple counter
again. ;-)

> And I'd only need the time of last ALTER TABLE :))

Indeed, I've been wondering why this information isn't available. :-(

Signature

Knut Stolze
DB2 Information Integration Development
IBM Germany

Reply to this Message