Re: Deadlock with ShareLocks? - Mailing list pgsql-hackers

From Mario Weilguni
Subject Re: Deadlock with ShareLocks?
Date
Msg-id 200512131657.54984.mweilguni@sime.com
Whole thread Raw
In response to Re: Deadlock with ShareLocks?  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Deadlock with ShareLocks?  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Am Dienstag, 13. Dezember 2005 16:52 schrieb Tom Lane:
> Mario Weilguni <mweilguni@sime.com> writes:
> > Since the type field is PK, there cannot be 2 rows with type='list', I
> > guess the deadlock must have some different explanation.
>
> Then the deadlock must involve rows in two different tables.  What else
> are you doing in the same transaction(s) as updating last_modified?

That's what I think too, unfortunatly, I can't say for sure. The application 
uses around 1000 different queries, and I had no logging on for this case. So 
I guess two backends might have issued interleaved updates.

I think without logging this cannot be solved here. 

Best regards,Mario Weilguni


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Deadlock with ShareLocks?
Next
From: "Francisco Figueiredo Jr."
Date:
Subject: Re: [Bulk] Re: [Bug] Server Crash, possible security exploit,