Mailman 3 July 2011 - Twisted

[Twisted-Python] Reconnection to Serialport after connectionLost
by Rene Klootwijk 01 Aug '11

01 Aug '11

I have written several daemons that listen to the serialport. Everything works well, however the serial port connections are actually USB connections over IP (using an USB network hub). This means the connection is lost once in a while which is detected and triggers the connectionLost method of the protocol. I tried numerous things to reconnect to the serial port but without success. When I reconnect to the serial port I doubt what to do with the reactor parameter, when I specify the current reactor, I am able to reconnect to the serial port, the connectionMade method of the protocol is triggered, but it will never trigger dataReceived although I know data is arriving. Does any body have a glue how you can reconnect to the serial port within a running reactor loop? -- Rene

6 10

[Twisted-Python] Adding introspection to xlmrpc server
by Justin Frost 31 Jul '11

31 Jul '11

I'm trying the example found at: http://twistedmatrix.com/documents/current/web/howto/xmlrpc.html#auto3 And it doesn't seem to be working. The following test code fails: import xmlrpclibs = xmlrpclib.ServerProxy("http://localhost:8080")print s.listMethods() The error says that the listMethods method isn't defined. If I create an xmrpc server without twisted and add introspection, it works fine. Is there a problem with the example or am I doing something wrong? Thanks,Justin

3 2

[Twisted-Python] Weekly Bug Summary
by exarkun＠twistedmatrix.com 31 Jul '11

31 Jul '11

1 0

[Twisted-Python] Re-entrancy policy for the reactor
by Itamar Turner-Trauring 28 Jul '11

28 Jul '11

I'm going to merge #5063 next time I have a few minutes when I'm more awake, which will mean my yak stack[1] is empty and I can go back to working on abortConnection(). As explained in http://twistedmatrix.com/trac/ticket/78, abortConnection() is like loseConnection(), except it doesn't wait until buffers are written out, it closes the connection immediately. The question is, how immediately? My current implementation, a bit like one of the half-close code paths, ends up calling connectionLost directly. That means you can have a call stack that looks like this: MyProtocol.dataReceived TCPConnection.abortConnection ... MyProtocol.connectionLost This can lead to re-entrancy bugs. I am tempted to suggest a policy of No Reactor Reentrancy, but that is currently violated by one of the TCP half-close code paths, and producers if I'm not mistaken. Some options: (a) Leave abortConnection() reentrant. (b) Make abortConnection() non-reentrant, don't change any current APIs. (c) Make all reactor APIs non-reentrant. (Since there's a ticket for documenting the reentrancy policy, you will notice I've added another yak to my stack. I'll finish #78 someday!) Thoughts? References: [1] https://secure.wikimedia.org/wiktionary/en/wiki/yak_shaving

3 5

[Twisted-Python] Do you use producers with SSL or TLS?
by Itamar Turner-Trauring 27 Jul '11

27 Jul '11

If so, I'd appreciate if if you could test your application against trunk Twisted (make sure you have a recent version of pyOpenSSL). I just landed some fixes to the new TLS implementation's producer code, and catching any problems before we release 11.1 would be good. Thanks!

2 1

[Twisted-Python] runReactor tests vs. t.i.task.cooperate and other users of the global reactor
by Itamar Turner-Trauring 27 Jul '11

27 Jul '11

Deep within the bowels of the new TLS code, cooperate() is used. Unfortunately, that uses the global reactor... which means it breaks in tests that create a new reactor instance. Should I just change the new TLS code to take in a reactor instance and pass it everywhere? Or, as part of the test framework, have runReactor override the global reactor instance? -Itamar

3 2

[Twisted-Python] A new & improved IResource
by Laurens Van Houtven 25 Jul '11

25 Jul '11

Hey! In relation to my recent getChild problem (specifically, I want to return deferred resources from it), it has become evident that IResource just won't cut it for a number of reasons and we should probably replace the API with something else. The issues we should take care of in this new API: 1. http://tm.tl/3711 (returning deferreds from render methods) 2. http://tm.tl/3621 (returning deferreds from getChild and friends) 3. http://tm.tl/288 (accessing uploaded data while in progress) If there are any missing tickets I'm forgetting about that are only fixable by breaking the interface, now would be a good time to speak up. I have created a new ticket (http://tm.tl/5152) for the new interface using an integration branch that addresses the issues in the individual tickets on a new interface as Glyph proposed. I've also commented on #3711 to say that it just won't work that way and that we need a new interface, referring them to the new ticket number. What should the migration plan look like? At least from the code point of view, exarkun has already partially answered this question: http://twistedmatrix.com/trac/ticket/288#comment:31 cheers lvh

4 8

[Twisted-Python] Weekly Bug Summary
by exarkun＠twistedmatrix.com 24 Jul '11

24 Jul '11

1 0

[Twisted-Python] XMLRPC, how to prevent stdout on post and get?
by hbd666 20 Jul '11

20 Jul '11

I have a simple xmlrpc server written using Twisted. Every time I get a request to it, stdout spits out something like the following: 2011-07-19 00:33:39-0700 [HTTPChannel,332,127.0.0.1] 127.0.0.1 - - [19/Jul/2011:07:33:38 +0000] "POST / HTTP/1.0" 200 128 "-" "xmlrpclib.py/1.0.1 (by www.pythonware.com)" How do I prevent this from happening? I also do not want this going to a log file either. Thanks anybody for your help!

4 5

[Twisted-Python] New TLS Code *Could* Cause Problems
by Reza Lotun 19 Jul '11

19 Jul '11

Hi All, I just wanted to fire a note about our experiences testing Twisted trunk (after _newtls.py landed) on some servers in our cluster, on PyOpenSSL 0.12. In our application (a twisted.web REST api endpoint), we make a lot of https calls to a SimpleDB Amazon AWS endpoint (sdb.amazonaws.com). At some point after a deployed test on some of our machines, we started getting our internel DNS requests to the sdb endpoint throttled (resulting in failures in most of our requests). Since every call to SimpleDB was made to the sdb endpoint, each call would be resolved, bombarding amazon's dns servers with requests. After diving in, I learned all Twisted connect* calls go through socket.gethostbyname, which apparently does not do any DNS caching (much to my surprise) unless something like nscd or a local caching nameserver is running (we ended up using BIND). Now, since we never ran a local caching nameserver before I wondered what changed. Could it have been that using the _newtls code (with PyOpenSSL 0.12) removed some element of DNS caching that OpenSSL was performing behind the scene? Is this possible? If so, then I just wanted to make sure everyone was aware that if you make a lot of DNS calls in a Twisted server running this code, you too could experience problems if you're not running a local caching nameserver. Thanks, Reza -- Reza Lotun mobile: +44 (0)7521 310 763 email: rlotun(a)gmail.com work: rlotun(a)twitter.com @rlotun

3 3