Le 2010-04-16 19:19, Scott Marlowe a écrit :
> On Fri, Apr 16, 2010 at 2:57 PM, Alexandre Leclerc<aleclerc@ipso.ca> wrote:
>
>> Thank you guys. I wanted to rush and vacuum the other tables and try, but I
>> decided to make a copy. This is actually running. (Enough mistakes in one
>> day to not take the time to do it.)
>>
>> After that we try to launch the DB and hopefully it will be working good
>> enough before the next maintenance. Else, a big week-end is coming.
>>
> Don't forget to schedule an upgrade to at least 8.2 since 8.1 is
> considered broken and unsupported / unsupportable on windows. 8.3
> changes some casting behaviour so you might want to wait until you can
> test / fix code to go there, but 8.2 is usually a painless upgrade
> from 8.1
>
Thank you Scott for this note.
I wanted to give some feedback on the situation:
- The vacuum completed after about 6-7 hours (by 16:15 PM). We
immediately made a backup before proceeding further. Then we vacuumed
the postgres et template1 databases.
- We were then successful at restarting postmaster and working with the
database. Everything works as expected.
Among all the help received (thank you all) I want to specially thank
Kevin and Tom for their excellent help and time through this urgent
problem we had to fix. I want to underline their detailed and precise
posts that were of great value for acting and taking decisions to
recover the DB. Thank you guys!
Again, thank you to all.
Best regards,
--
Alexandre Leclerc
