please? - Mailing list pgsql-hackers

From Pablo Funes
Subject please?
Date
Msg-id 199905311708.NAA32523@mancha.cs.brandeis.edu
Whole thread Raw
Responses Re: [HACKERS] please?  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Please take a sec to read this question. I've posted
it several times but got no comments at all. Thanx, Pablo.

---

Forwarded message:
From pablo Thu May 27 18:42:11 1999
Subject: nonblocking lock?
To: pgsql-hackers@postgresql.org
Date: Thu, 27 May 1999 18:42:11 -0400 (EDT)
Content-Type: text
Content-Length: 890

Is it possible to do a nonblocking lock? That is,
I want several clients to execute,

    begin
    if table A is locked
    then
        go around doing stuff on other tables
    else
        lock A and do stuff on A that takes a long time
    endif

the problem is, if I use normal lock, then
after one client has locked and is doing stuff on A
the other one will block and thus it won't be able
to go around doing stuff on other tables. Is it
possible to do a nonblocking lock that will just
fail if the table is locked already?


NOTE: I tried using PQrequestCancel but it won't
cancel the request. It still blocks for as long
as the lock lasts. The only way around I've found so
far is to use PQreset. That's crude but works. But
it leaves a dangling postmaster process that lives
until the orignal lock is freed. Any other ideas?

Thanks a lot

Pablo Funes
Brandeis University
pablo@cs.brandeis.edu

pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] Backend sent 0x45 type while idle
Next
From: Tom Lane
Date:
Subject: Re: [HACKERS] Open 6.5 items