[Date Prev][Date Next] [Chronological] [Thread] [Top]

RE: back-bdb deadlocks

To: "Lawrence Greenfield" <leg+@andrew.cmu.edu>, "Marijn Meijles" <marijn@bitpit.net>
Subject: RE: back-bdb deadlocks
From: "Howard Chu" <hyc@highlandsun.com>
Date: Fri, 18 Jan 2002 06:54:14 -0800
Cc: <openldap-devel@OpenLDAP.org>
Importance: Normal
In-reply-to: <200201181439.g0IEdi4i020600@smtp6.andrew.cmu.edu>

> -----Original Message-----
> From: Lawrence Greenfield [mailto:leg+@andrew.cmu.edu]

> Running the deadlock detector automatically is almost certainly what
> you want.  When I was doing similiar tests for Cyrus, this improved
> performance and simplified coding, since I no longer had to schedule a
> process or thread to do deadlock detection.
> 
> I also remember some tests (somewhat artificial perhaps) that showed
> that "youngest" was faster.

Thanks for the feedback, it's reassuring to know that the results make
sense and have been observed by others.

  -- Howard Chu
  Chief Architect, Symas Corp.       Director, Highland Sun
  http://www.symas.com               http://highlandsun.com/hyc  
  Symas: Premier OpenSource Development and Support

References:
- Re: back-bdb deadlocks
  - From: Lawrence Greenfield <leg+@andrew.cmu.edu>

Prev by Date: Re: back-bdb deadlocks
Next by Date: RE: back-bdb deadlocks
Index(es):
- Chronological
- Thread