Hi Oleg
> It was a conscious decision to provide only very basic protocol handlers
> out of the box in HttpCore, primarily to not be seen as competition to
> other Apache projects. This probably is no longer relevant. So, If
> anyone would be willing to work on improving protocol handlers in
> H
On Sun, 2010-04-25 at 22:00 +0530, Ruwan Linton wrote:
> Hi Oleg,
>
> Thanks a lot, I have got it fixed.
>
Hi Ruwan
I am glad I was of some help
> Moving forward, do you think that this can be incooperated with the
> IOReactor#pause??
Ideally I/O reactors should remain protocol agnostic. They
Hi Oleg,
Thanks a lot, I have got it fixed.
Moving forward, do you think that this can be incooperated with the
IOReactor#pause??
Why I am saying this is, IOReactor#pause should make sure that the server is
not accepting any more messages, right? but if you have keep-alive turned
on, those persi
On Thu, 2010-04-22 at 16:10 +0530, Ruwan Linton wrote:
> Oleg,
>
> I wrote the code, expecting that the IOControl#requestOutput will be calling
> the NhttpServiceHandler#outputReady method or the
> NhttpServiceHandler#responseReady method, but non of those get called?
>
> Do you have any clue or
Oleg,
I wrote the code, expecting that the IOControl#requestOutput will be calling
the NhttpServiceHandler#outputReady method or the
NhttpServiceHandler#responseReady method, but non of those get called?
Do you have any clue or is it a different method that I should be expected
be calling by the
On Sun, Apr 18, 2010 at 4:17 PM, Oleg Kalnichevski wrote:
> On Sun, 2010-04-18 at 08:00 +0530, Ruwan Linton wrote:
> > Hi Oleg,
> >
> >
> > > Hi Ruwan
> > >
> > > You have got to be very careful here. HTTP connection objects are not
> > > thread safe! One can easily run into a race condition when
On Sun, 2010-04-18 at 08:00 +0530, Ruwan Linton wrote:
> Hi Oleg,
>
>
> > Hi Ruwan
> >
> > You have got to be very careful here. HTTP connection objects are not
> > thread safe! One can easily run into a race condition when interacting
> > with connection objects from any other but an I/O dispatc
Hi Oleg,
> Hi Ruwan
>
> You have got to be very careful here. HTTP connection objects are not
> thread safe! One can easily run into a race condition when interacting
> with connection objects from any other but an I/O dispatch thread.
>
I have already encountered the above issue, and had a look
On Sat, 2010-04-17 at 09:03 +0530, Ruwan Linton wrote:
> Hi Oleg,
>
> I was able to get hold of the server side connection pool with using a list
> and recording them using the above mentioned methods.
>
> Now I was trying to check the condition, whether a connection is being used
> or not at a g
Hi Oleg,
I was able to get hold of the server side connection pool with using a list
and recording them using the above mentioned methods.
Now I was trying to check the condition, whether a connection is being used
or not at a given point in time. Meaning that whether this particular
connection h
Hi Oleg,
Sorry for the late response, I was on a vacation. I will try this and get
back to you with the result.
Thanks a lot for the help.
Thanks,
Ruwan
On Tue, Apr 6, 2010 at 6:02 PM, Oleg Kalnichevski wrote:
> On Tue, 2010-04-06 at 12:59 +0530, Ruwan Linton wrote:
> > Hi Devs,
> >
> > We (S
On Tue, 2010-04-06 at 12:59 +0530, Ruwan Linton wrote:
> Hi Devs,
>
> We (Synapse) has a requirement to close the persisted connections
> (keep-alive) when the IOReactor is being paused, and I have looked at the
> code and tried to get hold of all the persisted connections that are being
> used at
Hi Devs,
We (Synapse) has a requirement to close the persisted connections
(keep-alive) when the IOReactor is being paused, and I have looked at the
code and tried to get hold of all the persisted connections that are being
used at the time of the IOReactor pause, without any success :-(
1) Is it
13 matches
Mail list logo
