You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Jim Jagielski <ji...@jaguNET.com> on 2008/09/30 13:59:56 UTC
2.2.10 (Was: Re: STATUS file, 2.2.10 and voting)
I'm looking to T&R 2.2.10 this week, for a release early next
week... Complain now :)
Re: 2.2.10 T&R on Oct 7
Posted by Ruediger Pluem <rp...@apache.org>.
On 10/07/2008 04:54 PM, Greg Ames wrote:
> On Tue, Oct 7, 2008 at 5:11 AM, Ruediger Pluem <rp...@apache.org> wrote:
>
>>> The
>>> problem is that they all stay ESTABLISHED for KeepAliveTimeout seconds.
>> I
>>> was expecting to see 2 of them drop after 30 seconds. Any suggestions?
>> You need one more request after 30 seconds to trigger this since the
>> reslist
>> maintenance function only runs when a backend connection was given back to
>> the pool.
>
>
> ok. I did that and it works as expected, including smax=0. +1
>
> Thanks for the balancer tutorial, Rüdiger.
Thanks for reviewing Greg.
Regards
Rüdiger
Re: 2.2.10 T&R on Oct 7
Posted by Greg Ames <am...@gmail.com>.
On Tue, Oct 7, 2008 at 5:11 AM, Ruediger Pluem <rp...@apache.org> wrote:
> > The
> > problem is that they all stay ESTABLISHED for KeepAliveTimeout seconds.
> I
> > was expecting to see 2 of them drop after 30 seconds. Any suggestions?
>
> You need one more request after 30 seconds to trigger this since the
> reslist
> maintenance function only runs when a backend connection was given back to
> the pool.
ok. I did that and it works as expected, including smax=0. +1
Thanks for the balancer tutorial, Rüdiger.
Greg
Re: 2.2.10 T&R on Oct 7
Posted by Ruediger Pluem <rp...@apache.org>.
On 10/06/2008 11:30 PM, Greg Ames wrote:
>
> ok, KeepAliveTimeout 300 makes a big difference. But I haven't been able to
> observe smax and ttl working. WIth this,
>
> <Proxy balancer://mycluster>
> BalancerMember http://localhost:8093 keepalive=on smax=2 ttl=30
> </Proxy>
> ProxyPass /proxy balancer://mycluster/
>
> KeepAliveTimeout 300
>
> ...and using ab -c 4 to drive it, I now see 4 back end connections. The
> problem is that they all stay ESTABLISHED for KeepAliveTimeout seconds. I
> was expecting to see 2 of them drop after 30 seconds. Any suggestions?
You need one more request after 30 seconds to trigger this since the reslist
maintenance function only runs when a backend connection was given back to the pool.
Regards
Rüdiger
Re: 2.2.10 T&R on Oct 7
Posted by Greg Ames <am...@gmail.com>.
On Mon, Oct 6, 2008 at 4:34 PM, Ruediger Pluem <rp...@apache.org> wrote:
>
> Thanks for reviewing. First of all I guess you should increase the
> KeepAliveTimeout
> to a large value like 5 minutes to make observations easier. Furthermore I
> would increase
> the ttl parameter of your BalancerMember to something like 30.
> The above configuration makes only sense if you are using a threaded MPM
> like worker / event / WINNT
> with more than 2 threads per process.
> Then start your httpd with a limitation to one process (-X or ServerLimit
> 1) and try to do
> 3 requests to /proxy in parallel e.g. with telnet or nc. Afterwards you
> should
> have 3 backend connections of which one should vanish about 30 seconds
> after your requests.
>
ok, KeepAliveTimeout 300 makes a big difference. But I haven't been able to
observe smax and ttl working. WIth this,
<Proxy balancer://mycluster>
BalancerMember http://localhost:8093 keepalive=on smax=2 ttl=30
</Proxy>
ProxyPass /proxy balancer://mycluster/
KeepAliveTimeout 300
...and using ab -c 4 to drive it, I now see 4 back end connections. The
problem is that they all stay ESTABLISHED for KeepAliveTimeout seconds. I
was expecting to see 2 of them drop after 30 seconds. Any suggestions?
thanks,
Greg
Re: 2.2.10 T&R on Oct 7
Posted by Ruediger Pluem <rp...@apache.org>.
On 10/06/2008 10:18 PM, Greg Ames wrote:
> On Mon, Oct 6, 2008 at 3:13 PM, Ruediger Pluem <rp...@apache.org> wrote:
>
>>
>> On 10/06/2008 04:42 PM, Jim Jagielski wrote:
>>> Subj says it all...
>> Thanks for doing RM Jim. So c'mon guys. There must be someone
>> out there that reviews the remaining patch that misses only
>> *one* vote.
>>
>
> what do I need for a minimal config to test it and see a behavior change? I
> tried this, proxying to a different port on the same server, but couldn't
> get the back end connections to stay alive for more than a few seconds with
> the old code.
>
> <Proxy balancer://mycluster>
> BalancerMember http://localhost:8093 keepalive=on smax=2 ttl=1
> </Proxy>
> ProxyPass /proxy balancer://mycluster/
Thanks for reviewing. First of all I guess you should increase the KeepAliveTimeout
to a large value like 5 minutes to make observations easier. Furthermore I would increase
the ttl parameter of your BalancerMember to something like 30.
The above configuration makes only sense if you are using a threaded MPM like worker / event / WINNT
with more than 2 threads per process.
Then start your httpd with a limitation to one process (-X or ServerLimit 1) and try to do
3 requests to /proxy in parallel e.g. with telnet or nc. Afterwards you should
have 3 backend connections of which one should vanish about 30 seconds after your requests.
Regards
Rüdiger
Re: 2.2.10 T&R on Oct 7
Posted by Greg Ames <am...@gmail.com>.
On Mon, Oct 6, 2008 at 3:13 PM, Ruediger Pluem <rp...@apache.org> wrote:
>
>
> On 10/06/2008 04:42 PM, Jim Jagielski wrote:
> > Subj says it all...
>
> Thanks for doing RM Jim. So c'mon guys. There must be someone
> out there that reviews the remaining patch that misses only
> *one* vote.
>
what do I need for a minimal config to test it and see a behavior change? I
tried this, proxying to a different port on the same server, but couldn't
get the back end connections to stay alive for more than a few seconds with
the old code.
<Proxy balancer://mycluster>
BalancerMember http://localhost:8093 keepalive=on smax=2 ttl=1
</Proxy>
ProxyPass /proxy balancer://mycluster/
Greg
Re: 2.2.10 T&R on Oct 7
Posted by Ruediger Pluem <rp...@apache.org>.
On 10/06/2008 04:42 PM, Jim Jagielski wrote:
> Subj says it all...
Thanks for doing RM Jim. So c'mon guys. There must be someone
out there that reviews the remaining patch that misses only
*one* vote.
Regards
Rüdiger
2.2.10 T&R on Oct 7
Posted by Jim Jagielski <ji...@jaguNET.com>.
Subj says it all...
Re: 2.2.10 (Was: Re: STATUS file, 2.2.10 and voting)
Posted by Jim Jagielski <ji...@jaguNET.com>.
On Oct 2, 2008, at 10:21 AM, Jim Jagielski wrote:
>
> On Sep 30, 2008, at 7:59 AM, Jim Jagielski wrote:
>
>> I'm looking to T&R 2.2.10 this week, for a release early next
>> week... Complain now :)
>>
>
> There is one small proposed patch in STATUS that is lacking
> 1 single vote to be folded into 2.2.10... Anyone have cycles
> to look, test and vote?
>
FWIW, I'm hoping to say a 1/2 today, so the T&R will likely
happen Monday. This gives us the weekend to review the
remaining proposed backport as well as do some prelim testing
before the tag. Note which versions of APR/APU I intend
to bundle (1.3.3/1.3.4)
Re: 2.2.10 (Was: Re: STATUS file, 2.2.10 and voting)
Posted by Jim Jagielski <ji...@jaguNET.com>.
On Sep 30, 2008, at 7:59 AM, Jim Jagielski wrote:
> I'm looking to T&R 2.2.10 this week, for a release early next
> week... Complain now :)
>
There is one small proposed patch in STATUS that is lacking
1 single vote to be folded into 2.2.10... Anyone have cycles
to look, test and vote?