| From: | <david(dot)sahagian(at)emc(dot)com> |
|---|---|
| To: | <pgsql-general(at)postgresql(dot)org> |
| Subject: | explicit deadlock-victim-priority mechanism |
| Date: | 2011-10-25 21:28:40 |
| Message-ID: | F3CBFBA88397EA498B22A05FFA9EC49D4C38A581@MX22A.corp.emc.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
ref = [ E1QZDJc-0000XV-C8(at)gemulon(dot)postgresql(dot)org ]
I note with interest that [deadlock_timeout] can be used as . . .
"a poor-man's deadlock priority mechanism: a transaction with a high [deadlock_timeout] is less
likely to be chosen as the victim than one with a low [deadlock_timeout]"
I for one, would definitely be interested in "a more explicit priority mechanism".
My use case is:
We have some "batch work" processes, whose "deadlock-victim-priority" I would like to set to HIGH.
(at the expense of smaller transactions who I would like to be the victim)
-dvs-
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2011-10-25 21:50:18 | Re: writing a foreign data wrapper for hdfs, but getting and undefined symbol error for hdfsConnect |
| Previous Message | Rob_pg | 2011-10-25 21:24:47 | Re: writing a foreign data wrapper for hdfs, but getting and undefined symbol error for hdfsConnect |