You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@qpid.apache.org by Zhemzhitsky Sergey <Se...@troika.ru> on 2012/07/30 12:55:45 UTC
What's the difference between Receiver::get and Receiver::fetch
Hi qpid gurus,
Could you shed some light on what is the difference between Receiver::get and Receiver::fetch when using qpid messaging API?
According to the documentation Receiver::get retrieves a message from this receivers local queue and Receiver::fetch will check with the server that there is no message.
What is receiver’s local queue and how it should be used?
What’s is the best way from the point of view of performance to create a queue subscriber that will poll the queue continuously?
Best Regards,
Sergey
_______________________________________________________
The information contained in this message may be privileged and conf idential and protected from disclosure. If you are not the original intended recipient, you are hereby notified that any review, retransmission, dissemination, or other use of, or taking of any action in reliance upon, this information is prohibited. If you have received this communication in error, please notify the sender immediately by replying to this message and delete it from your computer. Thank you for your cooperation. Troika Dialog, Russia.
If you need assistance please contact our Contact Center (+7495) 258 0500 or go to www.troika.ru/eng/Contacts/system.wbp
RE: What's the difference between Receiver::get and Receiver::fetch
Posted by Zhemzhitsky Sergey <Se...@troika.ru>.
Thanks a lot!
Best Regards,
Sergey
-----Original Message-----
From: Gordon Sim [mailto:gsim@redhat.com]
Sent: Monday, July 30, 2012 3:59 PM
To: users@qpid.apache.org
Subject: Re: What's the difference between Receiver::get and Receiver::fetch
On 07/30/2012 12:52 PM, Zhemzhitsky Sergey wrote:
> So If I set the receiver's capacity to N, the call to the fetch()
> method will retrieve no more than N messages and any subsequent call
> to the fetch() will do the round trip to the server only if there are
> no more messages in the local receiver's queue?
Correct.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For additional commands, e-mail: users-help@qpid.apache.org
_______________________________________________________
The information contained in this message may be privileged and conf idential and protected from disclosure. If you are not the original intended recipient, you are hereby notified that any review, retransmission, dissemination, or other use of, or taking of any action in reliance upon, this information is prohibited. If you have received this communication in error, please notify the sender immediately by replying to this message and delete it from your computer. Thank you for your cooperation. Troika Dialog, Russia.
If you need assistance please contact our Contact Center (+7495) 258 0500 or go to www.troika.ru/eng/Contacts/system.wbp
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
RE: What's the difference between Receiver::get and Receiver::fetch
Posted by Zhemzhitsky Sergey <Se...@troika.ru>.
Gordon, thanks a lot!
Not it's clear enough.
Best Regards,
Sergey
-----Original Message-----
From: Gordon Sim [mailto:gsim@redhat.com]
Sent: Tuesday, July 31, 2012 1:57 PM
To: users@qpid.apache.org
Subject: Re: What's the difference between Receiver::get and Receiver::fetch
On 07/31/2012 10:33 AM, Zhemzhitsky Sergey wrote:
> So there is no difference between get() and fetch() if the capacity is specified and it is greater than zero?
There is, but it is quite subtle and likely not that relevant in most cases, at least where the timeout is also non-zero.
If you have a non-zero capacity and after the timeout ends there is no prefetched message available, get() will give up return without a message. There could however be a message 'on the way'. just pushed out by the server but not arrived at the client at that point. In the same circumstance fetch() would do one final check on the broker. So with
fetch() you would know that at the time the timeout expired there really was no message on the queue.
In most cases this isn't that interesting, but to try and illustrate the difference, imagine a rather contrived example where you do a synchronous send() to a queue, then immediately call get() for a receiver on that queue (whose capacity is greater than zero).
Whether get() returns a message will depend on the timeout specified and the time it takes the broker to actually push out the message. If you called fetch() instead, it would always return the message (assuming no one else sneaked in to take it), regardless of the timeout or the time taken for the broker to actually push it out.
Generally I would say use fetch() and if you want prefetch for greater efficiency set the capacity to a non-zero value. It may have been a mistake on my part to expose get().
> I'm able to get messages without any calls to fetch(), i.e.
>
> string
> addr("test-queue/test-subject;{create:always,node:{type:queue}}");
>
> Connection connection("amqp:tcp:localhost:5672", "{username: guest,
> password: guest}"); connection.open();
>
> Session session = connection.createSession("sessionId");
> Sender sender = session.createSender(addr); sender.send(Message("Hello
> World!"), true); session.sync(true);
>
> Receiver receiver = session.createReceiver(addr);
> receiver.setCapacity(1);
>
> Message message;
> Duration duration(1000);
> if(receiver.get(message, duration)) {
> cout << "Get: " << message.getContent() << endl;
> session.acknowledge(true);
> }
>
>
> Best Regards,
> Sergey
>
>
> -----Original Message-----
> From: Gordon Sim [mailto:gsim@redhat.com]
> Sent: Tuesday, July 31, 2012 1:18 PM
> To: users@qpid.apache.org
> Subject: Re: What's the difference between Receiver::get and
> Receiver::fetch
>
> On 07/31/2012 09:32 AM, Zhemzhitsky Sergey wrote:
>> I have recently found that Receiver::get does not return anything if the receiver's capacity is not specified (i.e. zero) and if the receiver's capacity is greater than 0, then Receiver::get retrieves messages from the broker.
>> At the same time Receiver::fetch always retrieves messages from the broker even if the receiver's capacity is not specified.
>>
>> Is this an expected behavior?
>
> If the capacity is zero there is no prefetch, and get() currently simply returns any prefetched messages (if they exist). On th eother hand, every fetch() call will pull off a message if there is no prefetch buffer to take one from.
>
> Does that make sense?
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For
> additional commands, e-mail: users-help@qpid.apache.org
>
>
> _______________________________________________________
>
> The information contained in this message may be privileged and conf idential and protected from disclosure. If you are not the original intended recipient, you are hereby notified that any review, retransmission, dissemination, or other use of, or taking of any action in reliance upon, this information is prohibited. If you have received this communication in error, please notify the sender immediately by replying to this message and delete it from your computer. Thank you for your cooperation. Troika Dialog, Russia.
> If you need assistance please contact our Contact Center (+7495) 258
> 0500 or go to www.troika.ru/eng/Contacts/system.wbp
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For
> additional commands, e-mail: users-help@qpid.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For additional commands, e-mail: users-help@qpid.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
Re: What's the difference between Receiver::get and Receiver::fetch
Posted by Gordon Sim <gs...@redhat.com>.
On 07/31/2012 10:33 AM, Zhemzhitsky Sergey wrote:
> So there is no difference between get() and fetch() if the capacity is specified and it is greater than zero?
There is, but it is quite subtle and likely not that relevant in most
cases, at least where the timeout is also non-zero.
If you have a non-zero capacity and after the timeout ends there is no
prefetched message available, get() will give up return without a
message. There could however be a message 'on the way'. just pushed out
by the server but not arrived at the client at that point. In the same
circumstance fetch() would do one final check on the broker. So with
fetch() you would know that at the time the timeout expired there really
was no message on the queue.
In most cases this isn't that interesting, but to try and illustrate the
difference, imagine a rather contrived example where you do a
synchronous send() to a queue, then immediately call get() for a
receiver on that queue (whose capacity is greater than zero).
Whether get() returns a message will depend on the timeout specified and
the time it takes the broker to actually push out the message. If you
called fetch() instead, it would always return the message (assuming no
one else sneaked in to take it), regardless of the timeout or the time
taken for the broker to actually push it out.
Generally I would say use fetch() and if you want prefetch for greater
efficiency set the capacity to a non-zero value. It may have been a
mistake on my part to expose get().
> I'm able to get messages without any calls to fetch(), i.e.
>
> string addr("test-queue/test-subject;{create:always,node:{type:queue}}");
>
> Connection connection("amqp:tcp:localhost:5672", "{username: guest, password: guest}");
> connection.open();
>
> Session session = connection.createSession("sessionId");
> Sender sender = session.createSender(addr);
> sender.send(Message("Hello World!"), true);
> session.sync(true);
>
> Receiver receiver = session.createReceiver(addr);
> receiver.setCapacity(1);
>
> Message message;
> Duration duration(1000);
> if(receiver.get(message, duration)) {
> cout << "Get: " << message.getContent() << endl;
> session.acknowledge(true);
> }
>
>
> Best Regards,
> Sergey
>
>
> -----Original Message-----
> From: Gordon Sim [mailto:gsim@redhat.com]
> Sent: Tuesday, July 31, 2012 1:18 PM
> To: users@qpid.apache.org
> Subject: Re: What's the difference between Receiver::get and Receiver::fetch
>
> On 07/31/2012 09:32 AM, Zhemzhitsky Sergey wrote:
>> I have recently found that Receiver::get does not return anything if the receiver's capacity is not specified (i.e. zero) and if the receiver's capacity is greater than 0, then Receiver::get retrieves messages from the broker.
>> At the same time Receiver::fetch always retrieves messages from the broker even if the receiver's capacity is not specified.
>>
>> Is this an expected behavior?
>
> If the capacity is zero there is no prefetch, and get() currently simply returns any prefetched messages (if they exist). On th eother hand, every fetch() call will pull off a message if there is no prefetch buffer to take one from.
>
> Does that make sense?
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For additional commands, e-mail: users-help@qpid.apache.org
>
>
> _______________________________________________________
>
> The information contained in this message may be privileged and conf idential and protected from disclosure. If you are not the original intended recipient, you are hereby notified that any review, retransmission, dissemination, or other use of, or taking of any action in reliance upon, this information is prohibited. If you have received this communication in error, please notify the sender immediately by replying to this message and delete it from your computer. Thank you for your cooperation. Troika Dialog, Russia.
> If you need assistance please contact our Contact Center (+7495) 258 0500 or go to www.troika.ru/eng/Contacts/system.wbp
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
> For additional commands, e-mail: users-help@qpid.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
RE: What's the difference between Receiver::get and Receiver::fetch
Posted by Zhemzhitsky Sergey <Se...@troika.ru>.
So there is no difference between get() and fetch() if the capacity is specified and it is greater than zero?
I'm able to get messages without any calls to fetch(), i.e.
string addr("test-queue/test-subject;{create:always,node:{type:queue}}");
Connection connection("amqp:tcp:localhost:5672", "{username: guest, password: guest}");
connection.open();
Session session = connection.createSession("sessionId");
Sender sender = session.createSender(addr);
sender.send(Message("Hello World!"), true);
session.sync(true);
Receiver receiver = session.createReceiver(addr);
receiver.setCapacity(1);
Message message;
Duration duration(1000);
if(receiver.get(message, duration)) {
cout << "Get: " << message.getContent() << endl;
session.acknowledge(true);
}
Best Regards,
Sergey
-----Original Message-----
From: Gordon Sim [mailto:gsim@redhat.com]
Sent: Tuesday, July 31, 2012 1:18 PM
To: users@qpid.apache.org
Subject: Re: What's the difference between Receiver::get and Receiver::fetch
On 07/31/2012 09:32 AM, Zhemzhitsky Sergey wrote:
> I have recently found that Receiver::get does not return anything if the receiver's capacity is not specified (i.e. zero) and if the receiver's capacity is greater than 0, then Receiver::get retrieves messages from the broker.
> At the same time Receiver::fetch always retrieves messages from the broker even if the receiver's capacity is not specified.
>
> Is this an expected behavior?
If the capacity is zero there is no prefetch, and get() currently simply returns any prefetched messages (if they exist). On th eother hand, every fetch() call will pull off a message if there is no prefetch buffer to take one from.
Does that make sense?
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For additional commands, e-mail: users-help@qpid.apache.org
_______________________________________________________
The information contained in this message may be privileged and conf idential and protected from disclosure. If you are not the original intended recipient, you are hereby notified that any review, retransmission, dissemination, or other use of, or taking of any action in reliance upon, this information is prohibited. If you have received this communication in error, please notify the sender immediately by replying to this message and delete it from your computer. Thank you for your cooperation. Troika Dialog, Russia.
If you need assistance please contact our Contact Center (+7495) 258 0500 or go to www.troika.ru/eng/Contacts/system.wbp
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
Re: What's the difference between Receiver::get and Receiver::fetch
Posted by Gordon Sim <gs...@redhat.com>.
On 07/31/2012 09:32 AM, Zhemzhitsky Sergey wrote:
> I have recently found that Receiver::get does not return anything if the receiver's capacity is not specified (i.e. zero) and if the receiver's capacity is greater than 0, then Receiver::get retrieves messages from the broker.
> At the same time Receiver::fetch always retrieves messages from the broker even if the receiver's capacity is not specified.
>
> Is this an expected behavior?
If the capacity is zero there is no prefetch, and get() currently simply
returns any prefetched messages (if they exist). On th eother hand,
every fetch() call will pull off a message if there is no prefetch
buffer to take one from.
Does that make sense?
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
RE: What's the difference between Receiver::get and Receiver::fetch
Posted by Zhemzhitsky Sergey <Se...@troika.ru>.
Hi Gordon,
I have recently found that Receiver::get does not return anything if the receiver's capacity is not specified (i.e. zero) and if the receiver's capacity is greater than 0, then Receiver::get retrieves messages from the broker.
At the same time Receiver::fetch always retrieves messages from the broker even if the receiver's capacity is not specified.
Is this an expected behavior?
Best Regards,
Sergey
-----Original Message-----
From: Gordon Sim [mailto:gsim@redhat.com]
Sent: Monday, July 30, 2012 3:59 PM
To: users@qpid.apache.org
Subject: Re: What's the difference between Receiver::get and Receiver::fetch
On 07/30/2012 12:52 PM, Zhemzhitsky Sergey wrote:
> So If I set the receiver's capacity to N, the call to the fetch()
> method will retrieve no more than N messages and any subsequent call
> to the fetch() will do the round trip to the server only if there are
> no more messages in the local receiver's queue?
Correct.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For additional commands, e-mail: users-help@qpid.apache.org
_______________________________________________________
The information contained in this message may be privileged and conf idential and protected from disclosure. If you are not the original intended recipient, you are hereby notified that any review, retransmission, dissemination, or other use of, or taking of any action in reliance upon, this information is prohibited. If you have received this communication in error, please notify the sender immediately by replying to this message and delete it from your computer. Thank you for your cooperation. Troika Dialog, Russia.
If you need assistance please contact our Contact Center (+7495) 258 0500 or go to www.troika.ru/eng/Contacts/system.wbp
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
Re: What's the difference between Receiver::get and Receiver::fetch
Posted by Gordon Sim <gs...@redhat.com>.
On 07/30/2012 12:52 PM, Zhemzhitsky Sergey wrote:
> So If I set the receiver's capacity to N, the call to the fetch()
> method will retrieve no more than N messages and any subsequent call
> to the fetch() will do the round trip to the server only if there are
> no more messages in the local receiver's queue?
Correct.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
RE: What's the difference between Receiver::get and Receiver::fetch
Posted by Zhemzhitsky Sergey <Se...@troika.ru>.
Hi Gordon,
So If I set the receiver's capacity to N, the call to the fetch() method will retrieve no more than N messages and any subsequent call to the fetch() will do the round trip to the server only if there are no more messages in the local receiver's queue?
Best Regards,
Sergey
-----Original Message-----
From: Gordon Sim [mailto:gsim@redhat.com]
Sent: Monday, July 30, 2012 3:15 PM
To: users@qpid.apache.org
Subject: Re: What's the difference between Receiver::get and Receiver::fetch
On 07/30/2012 11:55 AM, Zhemzhitsky Sergey wrote:
> Could you shed some light on what is the difference between Receiver::get and Receiver::fetch when using qpid messaging API?
The fetch() call is in essence a get() call that will do a roundtrip to the server if the timeout is reached with no message available to return, to verify there really is no message at that point.
The get() call on its own would simply return at that point. If the duration is infinite, there would be no difference between the calls.
> According to the documentation Receiver::get retrieves a message from this receivers local queue and Receiver::fetch will check with the server that there is no message.
>
> What is receiver’s local queue and how it should be used?
Its just the internal queue of message received, awaiting a fetch() call. Other than setting the capacity for the receiver, you don't really use it directly.
> What’s is the best way from the point of view of performance to create a queue subscriber that will poll the queue continuously?
If you have a fetch() in a loop, the the timeout will dictate the frequency of polling. A smaller timeout means more polling. However the call will return after the timeout whether or not a messages if found, hence the need to loop.
In my view though, there is little advantage in general to such polling unless there is some value for application to know whether there is a message available at some frequency/interval. The more you poll, the more load you put on the broker. So unless there is a reason for needing more polling, I would set the timeout to the maximum delay acceptable, i.e. the time at which if no message is available you perhaps want to warn about it and/or give up.
Hope this helps, I'm not entirely sure if I'm answering the question...
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org For additional commands, e-mail: users-help@qpid.apache.org
_______________________________________________________
The information contained in this message may be privileged and conf idential and protected from disclosure. If you are not the original intended recipient, you are hereby notified that any review, retransmission, dissemination, or other use of, or taking of any action in reliance upon, this information is prohibited. If you have received this communication in error, please notify the sender immediately by replying to this message and delete it from your computer. Thank you for your cooperation. Troika Dialog, Russia.
If you need assistance please contact our Contact Center (+7495) 258 0500 or go to www.troika.ru/eng/Contacts/system.wbp
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org
Re: What's the difference between Receiver::get and Receiver::fetch
Posted by Gordon Sim <gs...@redhat.com>.
On 07/30/2012 11:55 AM, Zhemzhitsky Sergey wrote:
> Could you shed some light on what is the difference between Receiver::get and Receiver::fetch when using qpid messaging API?
The fetch() call is in essence a get() call that will do a roundtrip to
the server if the timeout is reached with no message available to
return, to verify there really is no message at that point.
The get() call on its own would simply return at that point. If the
duration is infinite, there would be no difference between the calls.
> According to the documentation Receiver::get retrieves a message from this receivers local queue and Receiver::fetch will check with the server that there is no message.
>
> What is receiver’s local queue and how it should be used?
Its just the internal queue of message received, awaiting a fetch()
call. Other than setting the capacity for the receiver, you don't really
use it directly.
> What’s is the best way from the point of view of performance to create a queue subscriber that will poll the queue continuously?
If you have a fetch() in a loop, the the timeout will dictate the
frequency of polling. A smaller timeout means more polling. However the
call will return after the timeout whether or not a messages if found,
hence the need to loop.
In my view though, there is little advantage in general to such polling
unless there is some value for application to know whether there is a
message available at some frequency/interval. The more you poll, the
more load you put on the broker. So unless there is a reason for needing
more polling, I would set the timeout to the maximum delay acceptable,
i.e. the time at which if no message is available you perhaps want to
warn about it and/or give up.
Hope this helps, I'm not entirely sure if I'm answering the question...
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@qpid.apache.org
For additional commands, e-mail: users-help@qpid.apache.org