quetz-mod_python-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jim Gallacher <jg.li...@sympatico.ca>
Subject Re: FileSession.py - improved? version
Date Tue, 12 Apr 2005 15:25:55 GMT
Hi Grisha,

Gregory (Grisha) Trubetskoy wrote:
> On Tue, 12 Apr 2005, Jim Gallacher wrote:
>> However, when lock=0, there will be a race condition. Two process can 
>> have their own copies of FileSession data for a specific session at 
>> the same time.
> OK, but that's not specific to the FileSession? It sounded like there is 
> something inherent in FileSession about this.

It's not specific to FileSession - that's just the context of the 
discussion. I wanted to note that there was a race condition in case 
anyone was looking at the evolving FileSession code.

> This is basically the main reason why session locking is necessary, it'd 
> be good add this to documentation since it's not very easy to grasp why 
> session locking is necessary. But it belongs in BaseSession docs I think.

Agreed. When the FileSession code has settled down, I'll write some 
documentation. I can modify BaseSession documentation section detailing 
this limitatation as well, if you like. Or would it make sense to have a 
separate section describing the importance of session locking, since it 
is, as you say, difficult to grasp?


View raw message