DB issue with Shutdown and AI not wanting to start correctly

C

ctgivens

New Member
Hello All,

We are running OpenEdge version 10.1B03. Progress is running under Red Hat Enterprise Server Version 5.7. Kernal 2.6.18-274.e15 on an x86_64. This is running in a Virtual environment.

Okay, the problem is that when we shut down the database we get the following message with regards to AI:

(829) ** The File is not the correct after image file.

Before the restart we would have the following happen:

Switched to ai extent /ai/nxt.a4
This is after-image file number 410 since the last AIMAGE Begin
Aimage new session end
Before-image cluster size set to 1024
Before-image block size set to 16 kb
.bi file truncated
Truncate bi session end

run prostrct statistics

Multi-user session begin
---Then error message from above

This happens whenever we shut down the database. Is there something that we should be doing with AI before we shut it down, or when it starts up?

Thanks in advance

Charles
Progress user since 1992 (WDSII, TREND, SX, SYMIX, QAD, various homegrown applications)
 
For starters you should not be truncating the bi file with every shutdown.

How are you shutting down? With proshut? Or some other technique?
 
Tom,

Thanks for the quick reply. We are running SX Enterprise so we are using standard scripts. It shuts down the appserver then the dbbroker.

$DLC/bin/asbman -name $APPBROKER -port 7180 -stop

$DLC/bin/proshut -by ${DBNAME} -F

In looking at the dbbroker we are forcing a shutdown rather than the following:

$DLC/bin/dbman -database ${DBNAME} -port 7180 -stop

Any help would be appreciated.

Thanks

Charles
 
The first thing I do for most Infor SXE customers is to get rid of those standard scripts. Or at least make some key modifications ;)

You should NOT be doing a proshut -F as a matter of routine. It should never be necessary and it certainly shouldn't be the default.

Is there a "kill -9" buried in there somewhere? (I forget if there is or not -- if there is it is another one of those things that needs to go.)

The "error message" that you are showing is not complete -- it should also be showing the file name as part of the message. That's important. There are also likely to be a few messages surrounding that message. The complete text of all of them may provide important clues about what is going on.

Beyond that I don't have many ideas -- are there any other interesting entries in the .lg files prior to or during shutdown or startup that might be in any way related to the AI files?
 
The first thing I do for most Infor SXE customers is to get rid of those standard scripts. Or at least make some key modifications ;)

You should NOT be doing a proshut -F as a matter of routine. It should never be necessary and it certainly shouldn't be the default.

Is there a "kill -9" buried in there somewhere? (I forget if there is or not -- if there is it is another one of those things that needs to go.)

The "error message" that you are showing is not complete -- it should also be showing the file name as part of the message. That's important. There are also likely to be a few messages surrounding that message. The complete text of all of them may provide important clues about what is going on.

Beyond that I don't have many ideas -- are there any other interesting entries in the .lg files prior to or during shutdown or startup that might be in any way related to the AI files?
 
Wow - I guess they deploy scripts like that to ensure you depend upon their support...

It would be interesting to know what the db logs say after using $DLC/bin/proshut -by ${DBNAME} instead.

Are you applying the AI files to a database to see if they are valid ? I would wonder about that until I've tried it.

Not too far away from Springfield... Olathe, KS.
 
You must log in or register to reply here.
Back
Top