tag:blogger.com,1999:blog-3743585359130906790.comments2023-10-08T13:21:47.030+01:00Don't Believe Acronyms!petercmoorehttp://www.blogger.com/profile/17366331102790168465noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-3743585359130906790.post-24010076734936594502016-09-07T16:08:21.729+01:002016-09-07T16:08:21.729+01:00I realize I'm posting on a 4 year-old thread b...I realize I'm posting on a 4 year-old thread but thanks for this information. I've been scratching my head for the past hour trying to figure out why the LAST_DDL_TIME had changed on a table that *clearly* has not had any alters done to it. Best I can figure is the automated stats process (11.2.0.4) did something to the PK index causing the LAST_DDL_TIME on the table to change as well. Frustrating because I'm trying to do a flashback query well ahead of the oldest flashback entry but can't because of this LAST_DDL_TIME change on the table.<br />Anonymoushttps://www.blogger.com/profile/12312274019110231124noreply@blogger.comtag:blogger.com,1999:blog-3743585359130906790.post-72063565953880938022013-05-28T08:40:53.484+01:002013-05-28T08:40:53.484+01:00Not my choice, Noons. I used Amazon's ranking...Not my choice, Noons. I used Amazon's rankings to compile the list. :-)petercmoorehttps://www.blogger.com/profile/17366331102790168465noreply@blogger.comtag:blogger.com,1999:blog-3743585359130906790.post-84607362413660197262013-05-28T06:41:51.024+01:002013-05-28T06:41:51.024+01:00What? No RAC/FastProvisioning/CloudControl/Exadat...What? No RAC/FastProvisioning/CloudControl/Exadata/ODA/DBA1.0/BigData in that lot?<br />You can't possibly be taken for serious!<br />(But as another DBA with feet solidly resting on the ground, I do compliment you on your choice!)Noonshttps://www.blogger.com/profile/04285930853937157148noreply@blogger.comtag:blogger.com,1999:blog-3743585359130906790.post-1307445106417323682012-10-24T16:20:07.622+01:002012-10-24T16:20:07.622+01:00I can sort of understand changing grants on a tabl...I can sort of understand changing grants on a table will change the LAST_DDL_TIME. It's not exactly changing the table definition, but it is changing permissions on that table, so I guess it qualifies.petercmoorehttps://www.blogger.com/profile/17366331102790168465noreply@blogger.comtag:blogger.com,1999:blog-3743585359130906790.post-76208125716151524532012-10-17T08:39:13.741+01:002012-10-17T08:39:13.741+01:00Even granting SELECT access on a table will update...Even granting SELECT access on a table will update its LAST_DDL_TIME. You can see what I mean at the URL below:<br /><br />http://international-dba.blogspot.co.uk/2010/11/grant-select-updates-lastddltime.htmlAndrew Reidhttps://www.blogger.com/profile/10622696064206411002noreply@blogger.comtag:blogger.com,1999:blog-3743585359130906790.post-37201095257310660552011-03-28T16:07:07.505+01:002011-03-28T16:07:07.505+01:00Pete,
It simply piqued my interest - I recall doi...Pete,<br /><br />It simply piqued my interest - I recall doing some work on LAST_DDL_TIME about 18 months ago but that was with an old client and I don't have the details any more. I've never really trusted the LAST_DDL_TIME, but I do wonder why it's updating the parent table.<br /><br />The LAST_DDL_TIME on a table also updates on a CREATE INDEX or DROP INDEX command, so there must be some sort of additional dependency. I've done a quick trace check and tried with a DDL trigger but nothing obvious showing up. If you work out why, I hope you post it up!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3743585359130906790.post-65283810595042392982011-03-28T14:29:15.936+01:002011-03-28T14:29:15.936+01:00Hi Neil,
Thanks for taking an interest.
I probab...Hi Neil,<br /><br />Thanks for taking an interest.<br /><br />I probably shouldn't have used the word 'structural' - it was inaccurate for the example. I'll edit the post to make it clearer.<br /><br />I would expect moving a table or adding a comment on a table (or one of its columns) to change the last DDL time on the table. After all, we've performed some DDL against the table!<br /><br />However, my example below is the unexpected case that performing DDL against an INDEX updates the LAST_DDL_TIME for the index's parent table.<br /><br />I have just tested recompiling a procedure and it does NOT cause a change in LAST_DDL_TIME for any dependent table. I have amended the post to show this, as a counter example.<br /><br />Cheers,<br />Petepetercmoorehttps://www.blogger.com/profile/17366331102790168465noreply@blogger.comtag:blogger.com,1999:blog-3743585359130906790.post-67136895483493518752011-03-28T14:07:04.702+01:002011-03-28T14:07:04.702+01:00Peter,
There's an awful lot of non intrusive ...Peter,<br /><br />There's an awful lot of non intrusive commands that can cause last_ddl_time to update. Recompiling a procedure, moving a table (including ONLINE for IOT's), adding a comment to a table and many other non-structural changedAnonymousnoreply@blogger.com