Prostrct issues

[bjag]

Hi, folks

I hope that someone can help me; I'm in a bit of a bind....

I was creating some new storage areas, when I realized that there was not enough disk space. When I killed the prostrct command (before it could exceed the available diskspace), it left partial data files, apparently in an indeterminate state.

The files could not be deleted via prostrct remove, nor could I add any new files. A "prostrct repair"/"prostrct list" went fine, but did not help. I went ahead and deleted the files via the op system and deleted the reference of them from the .st file manually, but whenever I try to add any new s.a.'s, I get an error, still looking for those bad files.

I then tried copying those same files from the DR system, but now prostrct complains that the creation date is invalid. Does anyone know of anything else that could help, or is my database now toast?

Thanks, in advance, for any light that anyone can shed on this!

 

[TomBascom]

For the benefit of people who come along later (Brenda and I worked through this offline...)

Remove the bogus dr extents and then use:

prostrct unlock dbname -extents

 

[bjag]

For the benefit of people who come along later (Brenda and I worked through this offline...)

Remove the bogus dr extents and then use:

prostrct unlock dbname -extents

Tom,

First, I want to say, Thanks for your help - you were a real life-saver!!! I did the "unlock" command and deleted the bad extents, which allowed me to finish creating the new S.A.'s. Now, however, when I bring up the database, I get the following:


$
$ start_db unifiprd
OpenEdge Release 10.1B01 as of Tue Apr 3 21:12:59 EDT 2007
12:00:07 BROKER 0: Multi-user session begin. (333)
12:00:07 BROKER 0: Begin Physical Redo Phase at 0 . (5326)
12:00:07 BROKER 0: Physical Redo Phase Completed at blk 0 off 235 upd 0. (7161)
12:00:07 BROKER 0: At end of Physical redo, transaction table size is 2048. (13547)
12:00:07 BROKER 0: ** Your database was damaged. Dump its data and reload it. (37)
12:00:07 BROKER 0: Started for unifiprd_sql using tcp, pid 20072. (5644)
Batch processing will be performed using:
OpenEdge Release 10.1B01 as of Tue Apr 3 21:12:59 EDT 2007
OpenEdge Release 10.1B01 as of Tue Apr 3 21:12:59 EDT 2007
12:00:09 BROKER 1: Started for unifiprd using tcp, pid 20205. (5644)
12:00:09 BROKER 1: This is an additional broker for this protocol. (5645)
12:00:09 BROKER 1: This broker supports 4GL server groups only. (8863)


The "database damaged" message looks rather ominous. I don't have 50+ hours to D&L the whole database this week end. Does this error message indicate corruption? Is there any less-timely way to resolve this error?

Thanks again,
Brenda

 

[TomBascom]

That's a consequence of having done the "unlock". You just needed a "mock index rebuild" to clear the "damaged" flag. (Which was ok in this case because you had truncated the bi file earlier and the only activity was manipulating extents...)