> Hi, Alan
Thanks, I know little about python, when I execute
./qpid-cpp-benchmark -q 1 -s 1 -r 1 -m 20 --summarize --repeat 5
This is the result:
send-tp recv-tp l-min l-max l-avg total-tp
815081270.9085.12 33.34 8112
> Well, when qpidd -h, the result is different between 0.28 and 0.29
qpidd 0.29: --auth yes | no (1)
qpidd 0.28: --auth yes | no (0)
>
Does it mean 0.29 --auth no is default, 0.28 is the opposite?
在 2014年7月21日，下午11:58，Alan Conway 写道：
> On Tue, 2014-07-15 at 16:47
Hi, everyone. Alpha is now available:
Release artifacts: https://dist.apache.org/repos/dist/dev/qpid/0.30-alpha/
Release tool log output:
http://people.apache.org/~jross/qpid-trunk-2014-07-21.log
Maven local install:
http://people.apache.org/~jross/maven-trunk-2014-07-21/
Note that the jav
For the client stuff, it's important to work in the name of the particular
API, as you suggest.
For the broker, I'd like to leave room for the possibility of packaging
other server types and servers implemented in other languages. Our website
uses the following names for our servers:
- C++ bro
Currently the upstream packages in Fedora provide the following virtual
packages:
qpid-cpp-client -> qpid(client)
qpid-cpp-client-devel -> qpid(client-devel)
qpid-cpp-client-rdma -> qpid(client-rdma)
qpid-cpp-server -> qpid(server)
qpid-cpp-server-ssl -> qpid(server-ssl)
qpid-
On Mon, 2014-07-21 at 11:45 +0200, Jakub Scholz wrote:
> Hi,
>
> Qpid project contains two utilities: qpid-perftest and qpid-latency-test.
> You can use these instead of your own program. If you use these you can
> share the complete command which you used to start the performance test.
> Right no
On 07/21/2014 04:56 PM, Alan Conway wrote:
On Mon, 2014-07-21 at 13:16 +0100, Gordon Sim wrote:
On 07/21/2014 01:12 PM, Andreas Welchlin wrote:
Funny that it seemed to work with "decode()" as long as I only used C++
applications. I think the reason was that the C++ senders used also
"encode()".
On Tue, 2014-07-15 at 16:47 +0800, 郑勰 wrote:
> Hi，
> I have fixed this problem a few minutes ago, I must start qpidd with
> --auth no, although -auto no is default, it does seem not effect, if
> without —auth no, type link list showing that B->C is not
> connected.
Actually the default is aut
On Mon, 2014-07-21 at 13:16 +0100, Gordon Sim wrote:
> On 07/21/2014 01:12 PM, Andreas Welchlin wrote:
> > Funny that it seemed to work with "decode()" as long as I only used C++
> > applications. I think the reason was that the C++ senders used also
> > "encode()".
>
> Yes, that would result in a
On 07/21/2014 01:12 PM, Andreas Welchlin wrote:
Funny that it seemed to work with "decode()" as long as I only used C++
applications. I think the reason was that the C++ senders used also
"encode()".
Yes, that would result in an AMQP 0-10 encoded map (with the content
being 'binary' as far as
Am 21.07.2014 13:52, schrieb Gordon Sim:
On 07/21/2014 12:28 PM, Andreas Welchlin wrote:
No, I am not using getContentObject but the decode function:
Ok, that is the problem. You need to use getContentObject() for AMQP
1.0 messages. You can use that for 0-10 also and it works regardless
of
On 07/21/2014 12:28 PM, Andreas Welchlin wrote:
No, I am not using getContentObject but the decode function:
Ok, that is the problem. You need to use getContentObject() for AMQP 1.0
messages. You can use that for 0-10 also and it works regardless of the
type of object sent (e.g. binary or utf
Hi Gordon,
Cool!
I tried it and it works.
Thank you very much!
This workaround helps me very much.
Cheers
Andreas
Am 21.07.2014 13:28, schrieb Andreas Welchlin:
Hi Gordon,
thank you for your answer. I am also using the c++ broker (0.28).
>How are you getting the content using qpid::
Hi Gordon,
thank you for your answer. I am also using the c++ broker (0.28).
>How are you getting the content using qpid::messaging? Are you using
Message::getContentObject()?
No, I am not using getContentObject but the decode function:
MyMessage::MyMessage(const qpid::messaging::Message &q
On 07/21/2014 10:14 AM, Andreas Welchlin wrote:
I found one more strange behaviour between Java and C++ when using AMQP
1.0:
When a java application sets properties they are invisible for the c++
client:
[...]
Do you think this is a bug in QPID or is there something wrong with my
code?
Tha
On 07/18/2014 02:25 PM, Andreas Welchlin wrote:
Hi List,
I used Hello.java from "qpid-java-amqp-1-0-client-jms-0.28.tar.gz" to
send an amqp 1.0 message which contains a map to an exchange.
Java Code:
---
Hi,
Qpid project contains two utilities: qpid-perftest and qpid-latency-test.
You can use these instead of your own program. If you use these you can
share the complete command which you used to start the performance test.
Right now it is not clear for example how big your messages were or what
wa
Thanks for the answer Rob,
actually I expected to get the same logging as in JMS client 0.9 or 0.10
(done with sfl4j).
The protocol logging in JMS client 1.0 is better than nothing but the
previous logging contained more information.
Regards,
Erik
2014-07-20 17:21 GMT+02:00 Rob Godfrey :
>
Hi, everyone
I am using c++ broker on Intel(R) Core(TM) i5-4258U CPU @ 2.40GHz，1G
memory.
My boss wants me to measure the performance of c++ broker. So I wrote a
program to send some messages continues, and note the total cpu time.
This is the result :
total messag
Thank you for the answer Rob,
actually I expected to get the same logging as in JMS client 0.9 or 0.10 (done
with sfl4j).
The protocol logging in JMS client 1.0 is better than nothing but the previous
logging contained much more information.
E.g. correlationId and userId of the received and send
Hi All,
I found one more strange behaviour between Java and C++ when using AMQP 1.0:
When a java application sets properties they are invisible for the c++
client:
*** java sender sends a text message with a property:
MessageProducer messageProducer = producersession.createProducer(topic);
T
21 matches
Mail list logo
