durusmail: durus-users: Re: what I learned
Conflict in sequential atomic read/modify/write
2006-07-26
Re: what I learned
2006-07-27
2006-07-27
2006-07-27
2006-08-01
2006-07-27
2006-07-27
2006-07-27
2006-07-27
2006-07-29
Re: what I learned
Mike Sandford
2006-07-27
On Thursday 27 July 2006 14:14, David Binger wrote:
> On Jul 27, 2006, at 7:41 AM, Mike Sandford wrote:
> >  I do wonder if Quixote needs to update the access time
> > every time, but that is a question for another list :-)
>
> Quixote does not need to update the access time every time.
> The 'resolution' argument to Session._set_access_time() is supposed
> to slow down this activity.  Maybe you should try setting
> the SessionManager.ACCESS_TIME_RESOLUTION to a higher value.
>
Point taken.

> Are you sure this is what triggers your commit() storms?

Pretty much. I've been through the quixote and durus code (in rather more
detail than I planned) and the only update appears to be comming from the
access time. I've now laid out the application so that stuff that uses the
session for real work is down a different path from all the javascript and
images. To match this I've set the cookie path to the application path.
Result, no more commit storms.

I do have something that appears to be doing the job at the moment, so I'm no
longer particularly desparate (that was about three weeks ago!). That's why I
brought up the durus issue as that seemed more of an ongoing risk.

Mike S.
--
Mike Sandford
Director
Campbell Carr
12 Broadway
Amersham
HP7 0HP

Tel: +44 (0)1494-432323
Fax: +44 (0)1494-434888
Mob: +44 (0)7946-710026

www.campbellcarr.co.uk
m_sandford@campbellcarr.co.uk
reply