> On Monday 11 February 2008 16:17, Karl Pauls wrote:
> > However, do you (and others) think
> > it is acceptable to deliver log events asynchronously or at least
> > later then the actual point they occur?
>
> If you are only worried about Pax Logging case, then you need to do nothing.
> The locking in Log4J is severe enough for us to consider it "our problem" and
> I will probably later in the week do the "buffering" for you, which will be
> in proper order.
>
> I am not sure if there are any other Log Services out there, which will have
> similar problems.
>
> The broader concern probably still remains; Should the framework have any
> locks in place when calling outside itself, whether being log message, events
> or callbacks? The spec is pretty silent about it, and I have no clue how
> Felix measure up against the others on this point. But I think it is a valid
> question to ask both here and to the spec guys.

Well, we only hold locks in the case of log messages. For the other calls we
should be lock free .

regards,

Karl

> Cheers
> --
> Niclas Hedhman, Software Developer
>
> I  live here; http://tinyurl.com/2qq9er
> I  work here; http://tinyurl.com/2ymelc
> I relax here; http://tinyurl.com/2cgsug
>



-- 
Karl Pauls
[EMAIL PROTECTED]

Reply via email to