From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Oleg Bartunov <oleg(at)sai(dot)msu(dot)su> |
Cc: | Teodor Sigaev <teodor(at)sigaev(dot)ru>, Pgsql Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Gist Recovery testing |
Date: | 2005-06-12 17:57:38 |
Message-ID: | 25636.1118599058@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Oleg Bartunov <oleg(at)sai(dot)msu(dot)su> writes:
> There is a problem with gist_xlog_cleanup function - we have no
> information about user defined functions and indices, because postmaster
> doesn't started yet and system catalog isn't available. For btree it's not
> necessary, but for GiST we certainly need them.
btree manages to avoid calling the index support functions during WAL
restore --- can't you do the same? Perhaps you need to be including
more information in the initial xlog records, so that the cleanup step
has everything it needs. Or resort to brute-force search (which is more
or less what btree does). I don't think this operation needs to be very
efficient, since it's a corner case that should only seldom be invoked.
It's a really bad idea to be calling any support functions during WAL
recovery, since you won't have a transaction open for them to run in...
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Jim C. Nasby | 2005-06-12 19:52:32 | Re: mirroring oracle database in pgsql |
Previous Message | Oleg Bartunov | 2005-06-12 17:37:03 | Re: Gist Recovery testing |