I try to implement another DiscoveryAgent using a kind of database. I find this code and it helps me to understand how the DiscoveryAgent works :
https://github.com/sliard/multicastdb
But I don't know when the method registerService is called. When I changed the code of the multicast DiscoveryAgent to see when it's called, but it isn't when the DiscoveryAgent starts. But if it's not called, the broker can't advertize itself to the database...
Then I don't understand how a broker can communicate with another using the information of a DiscoveryAgent.
So if you can explain me when this method is called and if you have some documentation on the implementation of a DiscoveryAgent, it would be great.
Thanks in advance.
The registerService method is called from the Broker's TransportConnector when it is started which is at a different time from the start of the agent. You should take some time to look at the code of the various agents in ActiveMQ. The multicast version and the HTTP agent are good examples.
Keep in mind that the agents are written for both the Broker and the Client code to use so there is some code that isn't run from the Broker side and some that's not used on the client end. If you want to implement only a client side agent then you don't need to worry about the registration or advertisement bits, but if you want the broker to add itself to your DB implementation you would want to implement those bits so that the broker can add itself and do whatever your advertisement mechanism is.
The source code and test are your best documentation. Look at the code, run the tests and set breakpoints to see what's going on. You can even build from source and add in your own Logging if need be to get a better feel for it.
Related
We are using Akka.Net and in some cases we need actors to communicate reliably while preserving order over a message queue (i.e. Oracle Advanced Queues or WebSphere MQ, but any message queuing system would work such as RabbitMQ).
We have various requirements why we are using the message queue, so the question isn't if we should be using this with Akka, the question is how.
How would we go about connecting the queue up to Akka so that it is as seamless as possible?
Is a a custom Mailbox the route to go down? Do we need to right a custom IMessageQueue implementation? Or maybe we need a custom router? Are there any specific tests we can run to be sure our Mailbox/IMessageQueue works well with Akka.Net?
EDIT:
Should we maybe looking to implement a custom Transport?
Can any pointers be offered on where to start?
In general implementing custom mailbox based on some reliable queue is not feasible solution - actually it has been already done on the Akka JVM side, and it failed all hopes.
One of the basic reasons is usually the misunderstanding of the basic idea - when people are talking about reliable delivery (that MQ-systems offers), what they really mean, is reliable processing. What if your messages has been send with 100% delivery ratio, but ultimately receiving actor/node has crashed while processing them? From the mailbox point of view everything went smooth...
For this reason, usually the way to go is a dedicated actor - or hierarchy of them - working as a gateway to external messaging system. This way you can not only send message them but also mark them as receive after explicit acknowledgement from successfully completed process. One of the examples may be akka-rabbitmq (written in Scala).
I am experimenting with using Reactive Extensions to create a Windows Service.
Essentially what I want is for the Observer to sit on the server, the clients able to create observables and have them pushed to the server, the server informing the client of the progress of the job (not sure how to do this or what mechanism to use to do it), and then when it's done, having the server send the client the return code and output of the program it called. Can this be done? Is it the best way to do what I'm trying to do? If you need any more information, what would you need to know to help me?
This seems back to front. Generally clients know about servers (how to find then and connect). In contrast the Observer pattern (and therefore Rx) is about allowing something to callback to another observer that is does not know about.
In your case I think you simply want to have clients call methods on a server. Potentially these are bound to a single connection/session. The client however maybe an observer of the progress from the server and the final result.
See the Reactive Trader project by the team at Adaptive to see a .NET client server app using Rx.
I'm trying to understand how to persist data in a Twisted application. Let's say I've decided to write a Twisted server that:
Accepts inbound SMTP requests
Sends the message to a 3rd party system for modification
Relays the modified message to its destination
A typical Twisted tutorial would have you build this app using Deferreds and callbacks, roughly:
A Factory handles inbound requests
Each time a full email is received a call is sent to the remote message processor, returning a deferred
Add an errback that substitutes the original message if anything goes wrong in the modify call.
Add a callback to send the message on to the recipient, which again returns a deferred.
A real server would add/include additional call/errbacks to retry or notify the sender or whatnot. Again for simplicity, assume we consider this an acceptable amount of effort and just log errors.
Of course, this persists NO data in the event of a crash/restart/something else. I get that a solution involves a 3rd party persistent datastore (RabbitMQ is often mentioned) and could probably come up with a dozen random ways to achieve the outcome.
However, I imagine there are a few approaches that work best in a Twisted app. What do they look like? How do they store (and restore in the event of a crash) the in-process messages?
If you found this question, you probably already know that Twisted is event-based. It sounds simple, but the "hardest" part of the answer is to get the persistence platform generating the events we need when we need them. Naturally, you can persist the data in a DB or a message queue, but some platforms don't naturally generate events. For example:
ZeroMQ has (or at least had) no callback for new data. It's also relatively poor at persistence.
In other cases, events are easy but reliability is a problem:
pgSQL can be configured to generate events using triggers, but they're one-time things so you can't resume incomplete events
The light at the end of the tunnel seems to be something like RabbitMQ.
RabbitMQ can persist the message in a database to survive a crash
We can use acknowledgements on both legs (incoming SMTP to RabbitMQ and RabbitMQ to outgoing SMTP) to ensure the application is reliable. Importantly, RabbitMQ supports acknowledgements.
Finally, several of the RabbitMQ clients provide full asynchronous support (see for example pika, txampq, and puka)
It's enough for our purposes that the RabbitMQ client provides us an event-based interface.
At a more theoretical level, however, this need not be the case. In fact, despite the "notification" issue above, ZeroMQ has an event-based client. Even if our software is elegantly event-based, we will run into systems that aren't. In these cases, we have no choice but to fall back on polling. In principle, if not in practice, we just query the message provider for messages. When we exhaust the current queue (and immediately if there are no messages), we use a callLater command to check again in the future. It may feel anti-pattern, but it's (to the best of my knowledge anyway) the right way to handle this particular case.
I have implemented client server program using boost::asio library.
In my implementation there are times when io_service.run() blocks indefinitely. In case I pass another request to io_service, the blocked call begins to execute normally.
Is there any way to see what are the pending requests inside the io_service queue ?
I have not used work object to block the run call!
There are no official ways to query into the io_service to find all pending request. However, there are a few techniques to debug the problem:
Boost 1.47 introduced handler tracking. Simply define BOOST_ASIO_ENABLE_HANDLER_TRACKING and Boost.Asio will write debug output, including timestamps, an identifier, and the operation type, to the standard error stream.
Attach a debugger dig through the layers to find and examine operation queues. This answer covers both understanding handler tracking and using a debugger to examine an operation queue for the epoll_reactor.
Finally, if you believe it is a bug, then it may be worth updating to the latest version or checking the revision history for relevant changes. Regardless, describing the problem in more detail may allow others to help identify the source of the problem and potential solutions.
Now i spent a few hours reading and experimenting (i need more boost::asio functionality for work as well) and it turns out: Kind of.
But it is not as straightforward or readable as one might hope.
Under the hood (well, under the outermost hood) io_service has a bunch of other services registered, which do the work async_ operations of their respective fields require.
These are the "Services" described in the reference.
Now sadly, the services stay registered, wether there is work to do or not. For example if your io_service has a udp socket, it will still have all the corresponding services, even if the socket itself is inactive.
But you can ask your io_service which services it has. Lets say you want to know wether your io_service called m_io_service has an udp datagram_socket_service. Then you can call something like:
if (boost::asio::has_service<boost::asio::datagram_socket_service<boost::asio::ip::udp> >(m_io_service))
{
//Whatever
}
That does not help a lot, because it will be true no matter wether the socket is active or not. But after you know, that you have that service, you can get a ref to it using use_service instead of has_service but with the same elegant amount of <>.
And now you can inspect the service to see what it is up to. Sadly, it will not tell you what the outstanding handlers names are (probably partly because it does not know them) but if it is a socket, you can get its implemention_type and with that check whether it currently is_open or find either the local_endpoint as well as the remote_endpoint.
In case of a deadline_timer_service you can, among other stuff, find out when it expires_at.
See the reference for more information what the service is and is not willing to tell you.
http://www.boost.org/doc/libs/1_54_0/doc/html/boost_asio/reference.html
This information should then hopefully allow you to determine which async_ operation did not return.
And if not, at the very least you can cancel any unexpectedly active services.
I'd like to create a web service that an application server can contact to add itself to a list of servers implementing the application. Clients could then contact the service to get a list of servers. Something similar to how minecraft's heartbeats work for adding your server to the main server list.
I could implement it myself pretty easily, but I'm hoping someone has already created something like this.
Advanced features would be useful. Things like:
Allowing a client to perform queries on application-specific properties like the number of users currently connected to the server
Distributing the server list across more than one machine
Timing out a server's entry in the list if it hasn't sent a heartbeat within some amount of time
Does anyone know of a service like this? I know there are open protocols and servers for doing local-LAN service discovery, but this would be a WAN service.
The protocols I could find that had any relevance to your intended application are these:
XRDS (eXtensible Resource Descriptor Sequence).
XMPP Service Discovery protocol.
The XRDS documentation is obtuse, but you may be able to push service descriptions in XML format. The service type specification might be generic, but I get a headache from trying to decipher committee-speak.
The XMPP Service Discovery protocol (part of the protocol Formerly Known As Jabber) also looked promising, but it seems that even though you could push your service description, they expect it to be one of the services mentioned on this list. Extending it would make it nonstandard.
Finally, I found something called seap (SErvice Announcement Protocol). It's old, it's rickety, the source may be propriety, it's written in C and Perl, it's a kludge, but it seems to do what you want, kind-of.
It seems like pushing a service announcement pulse is such an application-specific and trivial problem, that almost nobody has considered solving the general case.
My advice? Read the protocols and sources mentioned above for inspiration (I'd start with seap), and then write, implement, and publish a generic (probably xml-based) protocol yourself. All the existing ones seem to be either application-specific, incomprehensible, or a kludge.
Basically, you can write it yourself though I am not aware if anyone has one for public (I wrote one over 10 yrs ago, but for a company).
database (TableCols: auto-counter, svr_name, svr_ip, check_in_time, any-other-data)
code to receive heartbeat (http://<you-app.com>?svr_name=XYZ&svr_ip=P.Q.R.S)
code to list out servers within certain check_in_time
code to do some housecleaning once a while (eg: purge old records)
To send a heartbeat out, you only need to send a http:// call, on Linux use wget* with crontab, on windows use wget.exe with task scheduler.
It is application specific, so even if you wrote one yourself, others can't use it without modifying the source code.