Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
On Thu, Jan 9, 2014 at 8:24 AM, crocket crockabisc...@gmail.com wrote: Nowadays, almost every city is covered with WiFi hotspots. And, almost every modern building and meetup is covered well by WiFi hotspots. Most houses have a hotspot or two. Why don't we take advantage of existing WiFi hotspots? It would make things so much easier while we devise a way to connect phones directly or wait for 802.11s. You can, of course. There are some aspects to take into account: * It works perfectly e.g. in the home or office where people naturally put devices on their access point. * Many public hotspots, especially in the US, block client-to-client traffic. * Public hotspots are trivial to tap, if you care about anonymity. * There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] I need some advice please for StreamQ-Proxy
On Thu, Jan 9, 2014 at 7:50 AM, Laurent Alebarde l.aleba...@free.fr wrote: And a 4th one: have the security protocol (CURVE or PARANO) build on top of libzmq or czmq. They would simply use the libzmq default mechanism NULL. That is what libcurve does. BTW, what are the pros/cons to have a security protocol built has a libzmq mechanism (as CURVE today), or built on top of libzmq ? it's obviously far more work to make and use libcurve than to use libzmq. It does not work in other languages except C. However you can do proxying. -Pieter ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] I need some advice please for StreamQ-Proxy
Thanks a lot Pieter (I missed your 23:44 message - sorry). Le 09/01/2014 09:08, Pieter Hintjens a écrit : On Thu, Jan 9, 2014 at 7:50 AM, Laurent Alebarde l.aleba...@free.fr wrote: And a 4th one: have the security protocol (CURVE or PARANO) build on top of libzmq or czmq. They would simply use the libzmq default mechanism NULL. That is what libcurve does. BTW, what are the pros/cons to have a security protocol built has a libzmq mechanism (as CURVE today), or built on top of libzmq ? it's obviously far more work to make and use libcurve than to use libzmq. It does not work in other languages except C. However you can do proxying. -Pieter ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] [zmqpp] receiving can only be done to empty messages?
The main reason for not just using push_back was that I was unsure how shifting the message when vectors went over capacity would effect the zmq_msg_t structures. I've since wrapped the whole lot in a frame object that handles it and now does just use push_back. (see https://github.com/benjamg/zmqpp/tree/feature/prepend_pop_message_frames) This branch also has been switched to using a deque over a vector to support push and pop front, however this appears to result in a much less performant binding and so I need to take another look at it. I may branch out the previous commit and issue a pull request for that though as it would solve your need for a new_raw_msg with a size as well. On 8 January 2014 20:42, Lindley French lindl...@gmail.com wrote: A question about the implementation. I notice that the zmqpp::message_t class jumps through some hoops whenever the add() or raw_new_msg() method is called. It appears to be creating a new array of parts, moving all existing parts there, and then swapping them. What's the rationale for this design? It seems like it would require a lot more allocations and copies (thankfully only of pointers, not entire buffers) than simply doing a push_back on the vector. On Wed, Jan 8, 2014 at 2:01 PM, Lindley French lindl...@gmail.com wrote: One additional minor suggestion: In the poller class, the add() and remove() overloads that take simple ints should probably accept them by value rather than const reference. There's no performance penalty, and it would make the type system play a bit nicer in some cases. For instance, I just tried to add a Boost.Asio UDP socket's native_handle() to a zmqpp poller (as a standin until UDP is working in 0MQ), and found that I couldn't do it if I was calling native_handle() from a const reference, but it worked fine with a non-const reference. On Wed, Jan 8, 2014 at 5:59 AM, Ben Gray b...@benjamg.com wrote: Pull request https://github.com/zeromq/zmqpp/pull/29 raised for this part of the issue (and the assert). On 7 January 2014 16:24, Lindley French lindl...@gmail.com wrote: In zmqpp, the socket_t::receive() method throws an exception if you pass it a non-empty message_t. This seems to be contradictory to the philosophy of zmq_recv(), which states that the message passed to it will be deallocated cleanly if necessary. Now, I know zmqpp defines a message as all parts of a multipart message, while libzmq defines it as just one part. So this isn't a direct contradiction of the API semantics. But it seems like this could be fixed without much effort. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] Send HTTP request and get response using ZeroMQ
Just connect to Zurl's REQ-handling endpoint (by default, tcp port 5553) and send a tnetstring-formatted message. You can find a tnetstring PHP library here: https://github.com/phuedx/tnetstring (note, I've not tried this, as I'm not a PHP developer). To make an HTTP request, send a message over the ZeroMQ socket with the following fields: id: a unique id for the request. could be a random number. method: the request method (e.g. GET) uri: the URL to request All of the above fields are string types. Zurl will then respond with a message containing the HTTP response. On 01/08/2014 10:41 PM, Ravir Pandey wrote: I am not familiar with python :( Can you please explain a little bit? On Thu, Jan 9, 2014 at 2:02 AM, Justin Karneges jus...@affinix.com mailto:jus...@affinix.com wrote: Hi Ravir, You shouldn't need a PHP version of Zurl. Just run the daemon and then speak with it via ZeroMQ from any language. For reference, you can see tools/get.py which performs a GET request from Python. Fanout.io uses Zurl on its servers for various tasks, and we decided to open source the code in case others might find it useful. On 01/07/2014 11:33 PM, Ravir Pandey wrote: and one thing i want to know is what's the role of Fanout.io service in this? On Wed, Jan 8, 2014 at 12:45 PM, Ravir Pandey ravir.pan...@commusoft.co.uk mailto:ravir.pan...@commusoft.co.uk mailto:ravir.pan...@commusoft.co.uk mailto:ravir.pan...@commusoft.co.uk wrote: Hi Justin, I have gone through https://github.com/fanout/zurl. Sounds interesting. Is there any PHP version available? On Mon, Jan 6, 2014 at 11:33 PM, Justin Karneges jus...@affinix.com mailto:jus...@affinix.com mailto:jus...@affinix.com mailto:jus...@affinix.com wrote: On 01/05/2014 10:12 PM, Ravir Pandey wrote: Is there any way to send http request and get response from server. For Ex. - i want to send request to my http server http://341.23.43.21, after processing request i need a reply from same server. How can i achieve this? You can try using Zurl. It gateways REQ/REP to HTTP, using libcurl. Justin ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] Send HTTP request and get response using ZeroMQ
Oh, and the message you send must be prefixed with a capital 'T' character. Byte 0 is a T, and byte 1 is the start of the tnetstring-formatted payload. On 01/09/2014 01:26 AM, Justin Karneges wrote: Just connect to Zurl's REQ-handling endpoint (by default, tcp port 5553) and send a tnetstring-formatted message. You can find a tnetstring PHP library here: https://github.com/phuedx/tnetstring (note, I've not tried this, as I'm not a PHP developer). To make an HTTP request, send a message over the ZeroMQ socket with the following fields: id: a unique id for the request. could be a random number. method: the request method (e.g. GET) uri: the URL to request All of the above fields are string types. Zurl will then respond with a message containing the HTTP response. On 01/08/2014 10:41 PM, Ravir Pandey wrote: I am not familiar with python :( Can you please explain a little bit? On Thu, Jan 9, 2014 at 2:02 AM, Justin Karneges jus...@affinix.com mailto:jus...@affinix.com wrote: Hi Ravir, You shouldn't need a PHP version of Zurl. Just run the daemon and then speak with it via ZeroMQ from any language. For reference, you can see tools/get.py which performs a GET request from Python. Fanout.io uses Zurl on its servers for various tasks, and we decided to open source the code in case others might find it useful. On 01/07/2014 11:33 PM, Ravir Pandey wrote: and one thing i want to know is what's the role of Fanout.io service in this? On Wed, Jan 8, 2014 at 12:45 PM, Ravir Pandey ravir.pan...@commusoft.co.uk mailto:ravir.pan...@commusoft.co.uk mailto:ravir.pan...@commusoft.co.uk mailto:ravir.pan...@commusoft.co.uk wrote: Hi Justin, I have gone through https://github.com/fanout/zurl. Sounds interesting. Is there any PHP version available? On Mon, Jan 6, 2014 at 11:33 PM, Justin Karneges jus...@affinix.com mailto:jus...@affinix.com mailto:jus...@affinix.com mailto:jus...@affinix.com wrote: On 01/05/2014 10:12 PM, Ravir Pandey wrote: Is there any way to send http request and get response from server. For Ex. - i want to send request to my http server http://341.23.43.21, after processing request i need a reply from same server. How can i achieve this? You can try using Zurl. It gateways REQ/REP to HTTP, using libcurl. Justin ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org mailto:zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
On 01/09/2014 09:06 AM, Pieter Hintjens wrote: * Many public hotspots, especially in the US, block client-to-client traffic. If you are talking about wireless isolation, then they only block multicasting. This means that you cannot discover other devices, but if you happen to know their IP than you can communcate with them. Some systems provide a fall-back solution that will do device discovery via a cloud server. Qualcomm's AllJoyn/AllSeen is an example of that. Not very EdgeNetish though. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
On Jan 9, 2014, at 10:52, Bjorn Reese bre...@mail1.stofanet.dk wrote: On 01/09/2014 09:06 AM, Pieter Hintjens wrote: * Many public hotspots, especially in the US, block client-to-client traffic. If you are talking about wireless isolation, then they only block multicasting. This means that you cannot discover other devices, but if you happen to know their IP than you can communcate with them. I think Pieter was quite clear with his sentence, and explains with much more detail on the book. I’ll add that in Europe it’s also typical to block client-to-client on the mobile network, including traffic initiated from the internet into the public IP of the mobile or data card. signature.asc Description: Message signed with OpenPGP using GPGMail ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
On Thu, Jan 9, 2014 at 12:19 PM, Bruno D. Rodrigues bruno.rodrig...@litux.org wrote: On Jan 9, 2014, at 10:52, Bjorn Reese bre...@mail1.stofanet.dk wrote: If you are talking about wireless isolation, then they only block multicasting. This means that you cannot discover other devices, but if you happen to know their IP than you can communcate with them. I think Pieter was quite clear with his sentence, and explains with much more detail on the book. From testing, it varies. On some hotspots, UDP broadcast and/or multicast is blocked. On other hotspots, client to client access is blocked, even if you know the IP address. One test case is bulk connecting, i.e. issue zmq_connect () to every other IP address on the block. That fails on perhaps 20-50% of hotspots, even in offices. It's a bit of a mess. Since we're not in control of the hardware we cannot be very ambitious. Hence my preference for the smartphone hotspot. -Pieter ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] I need some advice please for StreamQ-Proxy
No problem. The project you want is here: https://github.com/zeromq/libcurve On Thu, Jan 9, 2014 at 9:27 AM, Laurent Alebarde l.aleba...@free.fr wrote: Thanks a lot Pieter (I missed your 23:44 message - sorry). Le 09/01/2014 09:08, Pieter Hintjens a écrit : On Thu, Jan 9, 2014 at 7:50 AM, Laurent Alebarde l.aleba...@free.fr wrote: And a 4th one: have the security protocol (CURVE or PARANO) build on top of libzmq or czmq. They would simply use the libzmq default mechanism NULL. That is what libcurve does. BTW, what are the pros/cons to have a security protocol built has a libzmq mechanism (as CURVE today), or built on top of libzmq ? it's obviously far more work to make and use libcurve than to use libzmq. It does not work in other languages except C. However you can do proxying. -Pieter ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
[zeromq-dev] Majordomo C example slow
I am a new user. I just compiled the c Majordomo example and run it in my development laptop. It ran successfully but was extremely slow. ~/gitzguide/zguide/examples/C$ time ./mdclient 10 requests/replies processed real1m8.093s user0m8.224s sys0m8.872s Has anyone experienced this issue before? Am I doing something seriously wrong? -- Thanks !! Sridhar ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
Can manual assistance be effective in case of anonymous/pseudonomous broadcast used in async message apps? On Jan 9, 2014 3:05 PM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 1:32 AM, crocket crockabisc...@gmail.com wrote: Until there comes the edgenet mailing list, we can talk about it here. Sure. By manual assistance, do you mean people turn on hotspots and connect to them manually on site? Ideally, from within applications, e.g. inviting someone into your document would switch on a hotspot and display a PIN. Other people could then Search for invitations (scan for matching SSIDs) and enter the PIN to join. Is there any better manual method than I described above? On Jan 9, 2014 8:28 AM, Pieter Hintjens p...@imatix.com wrote: (a) I think we need to start an edgenet mailing list, as this is both such an interesting discussion, and yet so different from what people are expecting here (b) wifidirect is a not quite mesh 1-to-1 link and in fact very close to what I was thinking of with edgenet in the beginning. There's not a lot of choice, if you don't have 802.11s mesh working in the firmware, which few devices have. (c) you don't need root access to enable AP mode on most phones afaik. Not Android, anyhow (the API isn't documented but you can dig it out.) (d) we tried, and failed (in a previous closed project) to create and discover cells automatically. Best approach seems to be manual assistance. -Pieter On Wed, Jan 8, 2014 at 4:58 PM, Arnaud Loonstra arn...@sphaero.org wrote: On 01/08/2014 03:46 PM, Bruno D. Rodrigues wrote: I don’t know too many details about FON, but we have an ISP in Portugal that provides FON over all their clients (if they accept) and as soon as I connect to one FON hotstpot, the device automatically connects to any other. No idea if it’s just because it’s the same name, or if there is any other value that is forced to be the same, but somehow it is possible. On Jan 8, 2014, at 12:03, crocket crockabisc...@gmail.com mailto:crockabisc...@gmail.com wrote: People may be able to devise an algorithm for automatically turning on hotspots in several smartphones among many. But, even with hotspots turned on, how do phones connect to each other if smartphones simply refuse to connect to randomly named hotspots? Should we forcefully rename hotspots to a predetermined name(e.g., edgenet) and agree to connect to hotspots named edgenet? You could follow a roaming setup. Setup an hotspot with any name and make sure it is setup for roaming. Usually the authentication is done using radius. Which would be very easy to implement as radius is very simple protocol. Even just MAC authentication would suffice but you could go on into EAP/TTLS or PEAP setups. To setup an hotspot on a phone you'll probably need root access. WifiDirect would be better but I don't any devices that support it. It will probably be stopped by vendors ;) Just like PTT. Rg, Arnaud -- w: http://www.sphaero.org t: http://twitter.com/sphaero g: http://github.com/sphaero i: freenode: sphaero_z25 ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
[zeromq-dev] How many OS threads does ZeroMQ use internally?
Hey there, We are currently doing research for our project to figure out the best way to do inter-thread communication with a message queue. As we are using zmq for networking purposes already, it was logical to also research it for the above purpose. Now, we will have a number of threads (say 5-10) communicating with each other to send commands, synchronize data, etc. Each of the threads will have to communicate with a certain number of other threads. I figured that for this setup, making each thread a publisher as well as a subscriber to the other threads it wants to listen to would be the best bet. However, I do not know how many internal zmq threads that would yield. Imagine this socket situation for thread A (all threads use the same context, only inproc is used): Socket for publishing Socket for subscribing to thread B Socket for subscribing to thread F Socket for subscribing to thread J How many internal zmq OS threads would be spawned for this? If it is just one (per context), it would be fine. If it is one additional internal thread per our thread (or per publishing socket), it would probably be acceptable. If it was per socket (any kind of socket), however, that would be really bad as it would mean we had roughly +5 threads per our own thread. Even if those thread do not do much, they would still take away too many pieces of the cake. Hope someone can answer that question :) Regards, Jan ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
My idea was to make it like a game: by finding and peering with other devices you gain points. It is very close to deliberate peering with people you actually know. The two cases may overlap. On Thu, Jan 9, 2014 at 2:13 PM, crocket crockabisc...@gmail.com wrote: Can manual assistance be effective in case of anonymous/pseudonomous broadcast used in async message apps? On Jan 9, 2014 3:05 PM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 1:32 AM, crocket crockabisc...@gmail.com wrote: Until there comes the edgenet mailing list, we can talk about it here. Sure. By manual assistance, do you mean people turn on hotspots and connect to them manually on site? Ideally, from within applications, e.g. inviting someone into your document would switch on a hotspot and display a PIN. Other people could then Search for invitations (scan for matching SSIDs) and enter the PIN to join. Is there any better manual method than I described above? On Jan 9, 2014 8:28 AM, Pieter Hintjens p...@imatix.com wrote: (a) I think we need to start an edgenet mailing list, as this is both such an interesting discussion, and yet so different from what people are expecting here (b) wifidirect is a not quite mesh 1-to-1 link and in fact very close to what I was thinking of with edgenet in the beginning. There's not a lot of choice, if you don't have 802.11s mesh working in the firmware, which few devices have. (c) you don't need root access to enable AP mode on most phones afaik. Not Android, anyhow (the API isn't documented but you can dig it out.) (d) we tried, and failed (in a previous closed project) to create and discover cells automatically. Best approach seems to be manual assistance. -Pieter On Wed, Jan 8, 2014 at 4:58 PM, Arnaud Loonstra arn...@sphaero.org wrote: On 01/08/2014 03:46 PM, Bruno D. Rodrigues wrote: I don’t know too many details about FON, but we have an ISP in Portugal that provides FON over all their clients (if they accept) and as soon as I connect to one FON hotstpot, the device automatically connects to any other. No idea if it’s just because it’s the same name, or if there is any other value that is forced to be the same, but somehow it is possible. On Jan 8, 2014, at 12:03, crocket crockabisc...@gmail.com mailto:crockabisc...@gmail.com wrote: People may be able to devise an algorithm for automatically turning on hotspots in several smartphones among many. But, even with hotspots turned on, how do phones connect to each other if smartphones simply refuse to connect to randomly named hotspots? Should we forcefully rename hotspots to a predetermined name(e.g., edgenet) and agree to connect to hotspots named edgenet? You could follow a roaming setup. Setup an hotspot with any name and make sure it is setup for roaming. Usually the authentication is done using radius. Which would be very easy to implement as radius is very simple protocol. Even just MAC authentication would suffice but you could go on into EAP/TTLS or PEAP setups. To setup an hotspot on a phone you'll probably need root access. WifiDirect would be better but I don't any devices that support it. It will probably be stopped by vendors ;) Just like PTT. Rg, Arnaud -- w: http://www.sphaero.org t: http://twitter.com/sphaero g: http://github.com/sphaero i: freenode: sphaero_z25 ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How many OS threads does ZeroMQ use internally?
The documentation recommends creating one single context per process, and when creating it you specify how many zeromq threads to use. -- Bruno Rodrigues Sent from my iPhone No dia 09/01/2014, às 13:33, Jan Drabner j...@jdrabner.eu escreveu: Hey there, We are currently doing research for our project to figure out the best way to do inter-thread communication with a message queue. As we are using zmq for networking purposes already, it was logical to also research it for the above purpose. Now, we will have a number of threads (say 5-10) communicating with each other to send commands, synchronize data, etc. Each of the threads will have to communicate with a certain number of other threads. I figured that for this setup, making each thread a publisher as well as a subscriber to the other threads it wants to listen to would be the best bet. However, I do not know how many internal zmq threads that would yield. Imagine this socket situation for thread A (all threads use the same context, only inproc is used): Socket for publishing Socket for subscribing to thread B Socket for subscribing to thread F Socket for subscribing to thread J How many internal zmq OS threads would be spawned for this? If it is just one (per context), it would be fine. If it is one additional internal thread per our thread (or per publishing socket), it would probably be acceptable. If it was per socket (any kind of socket), however, that would be really bad as it would mean we had roughly +5 threads per our own thread. Even if those thread do not do much, they would still take away too many pieces of the cake. Hope someone can answer that question :) Regards, Jan ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
Bluetooth works well enough as a standard network interface bnep0 if the messages you're sending are small. (It really is pretty slow, though.) The hard part is programmatically configuring the stack to use it that way, and (preferably) programmatically pairing without user interaction. On Thu, Jan 9, 2014 at 1:02 AM, Pieter Hintjens p...@imatix.com wrote: Bluetooth is not designed for general app-to-app messaging, rather as a wireless connector for low-rate devices. You can plausibly use it to set up WiFi connections, a hybrid solution. AllJoyn is made by Qualcomm, who have the expertise to make BlueTooth work. For most people it's not practical. The ideal would be 802.11s mesh (as used in the OLPC), which no-one seems to want to build into consumer mobile devices. My assumption is that useful apps running on WiFi can push the market towards better connectivity. On Thu, Jan 9, 2014 at 1:31 AM, Lindley French lindl...@gmail.com wrote: Wifi isn't the only answer here. Bluetooth can provide similar functionality. I don't know the details but the Alljoyn people claim to be able to automatically set up Bluetooth links with phones in range. On Jan 8, 2014, at 6:27 PM, Pieter Hintjens p...@imatix.com wrote: (a) I think we need to start an edgenet mailing list, as this is both such an interesting discussion, and yet so different from what people are expecting here (b) wifidirect is a not quite mesh 1-to-1 link and in fact very close to what I was thinking of with edgenet in the beginning. There's not a lot of choice, if you don't have 802.11s mesh working in the firmware, which few devices have. (c) you don't need root access to enable AP mode on most phones afaik. Not Android, anyhow (the API isn't documented but you can dig it out.) (d) we tried, and failed (in a previous closed project) to create and discover cells automatically. Best approach seems to be manual assistance. -Pieter On Wed, Jan 8, 2014 at 4:58 PM, Arnaud Loonstra arn...@sphaero.org wrote: On 01/08/2014 03:46 PM, Bruno D. Rodrigues wrote: I don’t know too many details about FON, but we have an ISP in Portugal that provides FON over all their clients (if they accept) and as soon as I connect to one FON hotstpot, the device automatically connects to any other. No idea if it’s just because it’s the same name, or if there is any other value that is forced to be the same, but somehow it is possible. On Jan 8, 2014, at 12:03, crocket crockabisc...@gmail.com mailto:crockabisc...@gmail.com wrote: People may be able to devise an algorithm for automatically turning on hotspots in several smartphones among many. But, even with hotspots turned on, how do phones connect to each other if smartphones simply refuse to connect to randomly named hotspots? Should we forcefully rename hotspots to a predetermined name(e.g., edgenet) and agree to connect to hotspots named edgenet? You could follow a roaming setup. Setup an hotspot with any name and make sure it is setup for roaming. Usually the authentication is done using radius. Which would be very easy to implement as radius is very simple protocol. Even just MAC authentication would suffice but you could go on into EAP/TTLS or PEAP setups. To setup an hotspot on a phone you'll probably need root access. WifiDirect would be better but I don't any devices that support it. It will probably be stopped by vendors ;) Just like PTT. Rg, Arnaud -- w: http://www.sphaero.org t: http://twitter.com/sphaero g: http://github.com/sphaero i: freenode: sphaero_z25 ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. Battery life may be more of an issue than you realize. Wifi radios use far less power than 3G radios, but they still *do* use power, and the pattern of network activity can have a strong effect on battery life. The most important thing to realize is that people expect their smartphones not to use much power while they're idle. If you're doing network activity all the time in the background, that isn't the case, and people will become unhappy with your software and uninstall it. (Note the reactions to the first release of Google Now, which committed a similar sin with location services. GPS receivers use a lot of power too.) To some extent you can be smart enough to reduce network activity significantly while the phone is idle compared to active. Beacon less often, include a parameter in the beacon letting other phones know you'll only be listening for a short window after a beacon, etc. It's possible to give the radio a break. My concern is that once a phone is trying to host an access point, it gets a lot harder to be selective about when the wifi radio is active. It has to be on pretty much all the time. The exception might be ad-hoc mode-once one node creates an ad-hoc network (which is surprisingly tricky on Android), others can sustain the SSID without it. However, if the anchoring node was providing DHCP services, then obviously that won't work anymore once it goes offline. On Thu, Jan 9, 2014 at 3:06 AM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 8:24 AM, crocket crockabisc...@gmail.com wrote: Nowadays, almost every city is covered with WiFi hotspots. And, almost every modern building and meetup is covered well by WiFi hotspots. Most houses have a hotspot or two. Why don't we take advantage of existing WiFi hotspots? It would make things so much easier while we devise a way to connect phones directly or wait for 802.11s. You can, of course. There are some aspects to take into account: * It works perfectly e.g. in the home or office where people naturally put devices on their access point. * Many public hotspots, especially in the US, block client-to-client traffic. * Public hotspots are trivial to tap, if you care about anonymity. * There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How many OS threads does ZeroMQ use internally?
For inproc you don't need any I/O threads at all. The default will be 1 additional thread per context, so as Bruno says, per process. On Thu, Jan 9, 2014 at 2:33 PM, Jan Drabner j...@jdrabner.eu wrote: Hey there, We are currently doing research for our project to figure out the best way to do inter-thread communication with a message queue. As we are using zmq for networking purposes already, it was logical to also research it for the above purpose. Now, we will have a number of threads (say 5-10) communicating with each other to send commands, synchronize data, etc. Each of the threads will have to communicate with a certain number of other threads. I figured that for this setup, making each thread a publisher as well as a subscriber to the other threads it wants to listen to would be the best bet. However, I do not know how many internal zmq threads that would yield. Imagine this socket situation for thread A (all threads use the same context, only inproc is used): Socket for publishing Socket for subscribing to thread B Socket for subscribing to thread F Socket for subscribing to thread J How many internal zmq OS threads would be spawned for this? If it is just one (per context), it would be fine. If it is one additional internal thread per our thread (or per publishing socket), it would probably be acceptable. If it was per socket (any kind of socket), however, that would be really bad as it would mean we had roughly +5 threads per our own thread. Even if those thread do not do much, they would still take away too many pieces of the cake. Hope someone can answer that question :) Regards, Jan ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
Ad-hoc mode is too poorly supported to depend on. And it's slow. In any case a phone cannot act as a hotspot for any length of time. On some firmwares enabling AP mode will switch off 3G. This is still fine for us. My idea was to put this decision at the user level, initially. The application chooses to host or to join some activity. It's a natural pattern in a group of people. The host of a meeting is less likely to walk out the door, and more likely to have power. On Thu, Jan 9, 2014 at 3:14 PM, Lindley French lindl...@gmail.com wrote: There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. Battery life may be more of an issue than you realize. Wifi radios use far less power than 3G radios, but they still *do* use power, and the pattern of network activity can have a strong effect on battery life. The most important thing to realize is that people expect their smartphones not to use much power while they're idle. If you're doing network activity all the time in the background, that isn't the case, and people will become unhappy with your software and uninstall it. (Note the reactions to the first release of Google Now, which committed a similar sin with location services. GPS receivers use a lot of power too.) To some extent you can be smart enough to reduce network activity significantly while the phone is idle compared to active. Beacon less often, include a parameter in the beacon letting other phones know you'll only be listening for a short window after a beacon, etc. It's possible to give the radio a break. My concern is that once a phone is trying to host an access point, it gets a lot harder to be selective about when the wifi radio is active. It has to be on pretty much all the time. The exception might be ad-hoc mode-once one node creates an ad-hoc network (which is surprisingly tricky on Android), others can sustain the SSID without it. However, if the anchoring node was providing DHCP services, then obviously that won't work anymore once it goes offline. On Thu, Jan 9, 2014 at 3:06 AM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 8:24 AM, crocket crockabisc...@gmail.com wrote: Nowadays, almost every city is covered with WiFi hotspots. And, almost every modern building and meetup is covered well by WiFi hotspots. Most houses have a hotspot or two. Why don't we take advantage of existing WiFi hotspots? It would make things so much easier while we devise a way to connect phones directly or wait for 802.11s. You can, of course. There are some aspects to take into account: * It works perfectly e.g. in the home or office where people naturally put devices on their access point. * Many public hotspots, especially in the US, block client-to-client traffic. * Public hotspots are trivial to tap, if you care about anonymity. * There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] Router socket to force peer disconnect
If you don't trust your peers you should probably be using ZeroMQ 4.x and security. Then you will only see authenticated peers. On Wed, Jan 8, 2014 at 12:32 PM, Dmitriy Vsekhvalnov dvsekhval...@gmail.com wrote: Hi All, continue my little evaluation of zeromq and wanted to ask next question: is it possible to force disconnect of client from router side? (e.g. disconnect req or dealer). If it is not possible are those clients who connected to router socket consume resources? A lot? Even if no messages send/received from them? We are in environment where we don't trust peers until they complete handshake with router socket. And would like to disconnect them if handshake failed. (i know we can simply ignore them and keep connected). Thanks. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
Having no experience on the matter, why not use bluetooth as an interface that searches for peers and wifi for real network traffic. Wifi will be switched on only when there is need for data to be sent. That could also help the wifi automatically switch to the best access point. Bluetooth has a range of 50m. That is good enough for peer discovery and peer coordination. From the wikipedia, both of them are available on smartphones. http://en.wikipedia.org/wiki/Bluetooth_low_energy 2014/1/9 Pieter Hintjens p...@imatix.com Ad-hoc mode is too poorly supported to depend on. And it's slow. In any case a phone cannot act as a hotspot for any length of time. On some firmwares enabling AP mode will switch off 3G. This is still fine for us. My idea was to put this decision at the user level, initially. The application chooses to host or to join some activity. It's a natural pattern in a group of people. The host of a meeting is less likely to walk out the door, and more likely to have power. On Thu, Jan 9, 2014 at 3:14 PM, Lindley French lindl...@gmail.com wrote: There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. Battery life may be more of an issue than you realize. Wifi radios use far less power than 3G radios, but they still *do* use power, and the pattern of network activity can have a strong effect on battery life. The most important thing to realize is that people expect their smartphones not to use much power while they're idle. If you're doing network activity all the time in the background, that isn't the case, and people will become unhappy with your software and uninstall it. (Note the reactions to the first release of Google Now, which committed a similar sin with location services. GPS receivers use a lot of power too.) To some extent you can be smart enough to reduce network activity significantly while the phone is idle compared to active. Beacon less often, include a parameter in the beacon letting other phones know you'll only be listening for a short window after a beacon, etc. It's possible to give the radio a break. My concern is that once a phone is trying to host an access point, it gets a lot harder to be selective about when the wifi radio is active. It has to be on pretty much all the time. The exception might be ad-hoc mode-once one node creates an ad-hoc network (which is surprisingly tricky on Android), others can sustain the SSID without it. However, if the anchoring node was providing DHCP services, then obviously that won't work anymore once it goes offline. On Thu, Jan 9, 2014 at 3:06 AM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 8:24 AM, crocket crockabisc...@gmail.com wrote: Nowadays, almost every city is covered with WiFi hotspots. And, almost every modern building and meetup is covered well by WiFi hotspots. Most houses have a hotspot or two. Why don't we take advantage of existing WiFi hotspots? It would make things so much easier while we devise a way to connect phones directly or wait for 802.11s. You can, of course. There are some aspects to take into account: * It works perfectly e.g. in the home or office where people naturally put devices on their access point. * Many public hotspots, especially in the US, block client-to-client traffic. * Public hotspots are trivial to tap, if you care about anonymity. * There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev -- Sincerely yours, Apostolis Xekoukoulotakis ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] [zmqpp] receiving can only be done to empty messages?
I imagine you've figured it out by now, but just in case When a vector method needs to reallocate the underlying array, existing elements are move-constructed onto the new memory. If no move constructor is defined or if you are not using C++11 mode, they will be copy-constructed instead. (Placement new is used.) Then the objects in the original memory will be destructed. This usually provides the expected behavior for both C and C++ types. My one concern is that the vector class does contain some optimizations to allow it to use memcpy() if the contained type is plain-old-data. This is usually detected correctly, but since zmq.h does some trickery to make the zmq_msg_t type opaque, I can't be certain the right thing will happen without more study. Your solution in the above thread of defining a frame move constructor that calls zmq_msg_move() should be safe enough. This branch also has been switched to using a deque over a vector to support push and pop front, however this appears to result in a much less performant binding and so I need to take another look at it. Strange. A deque will be slower than a vector in some cases, but it should be a small constant multiplier. If you figure out what's going on I'd be curious to know what it is. Pragmatically speaking, a vector should usually be fine even if you want to support a push/pop_front capabilityzmq messages will rarely have *that* many parts, and your current master branch shows that zmq_msg_move()ing all of them when a part is added isn't unacceptably slow anyway. Another way of looking at the situation is that it's pretty rare for people to access message parts in other than linear order, so something like a std::list could be made to work relatively well. The worst-case promises would be a little odd though. On Thu, Jan 9, 2014 at 4:22 AM, Ben Gray b...@benjamg.com wrote: The main reason for not just using push_back was that I was unsure how shifting the message when vectors went over capacity would effect the zmq_msg_t structures. I've since wrapped the whole lot in a frame object that handles it and now does just use push_back. (see https://github.com/benjamg/zmqpp/tree/feature/prepend_pop_message_frames) This branch also has been switched to using a deque over a vector to support push and pop front, however this appears to result in a much less performant binding and so I need to take another look at it. I may branch out the previous commit and issue a pull request for that though as it would solve your need for a new_raw_msg with a size as well. On 8 January 2014 20:42, Lindley French lindl...@gmail.com wrote: A question about the implementation. I notice that the zmqpp::message_t class jumps through some hoops whenever the add() or raw_new_msg() method is called. It appears to be creating a new array of parts, moving all existing parts there, and then swapping them. What's the rationale for this design? It seems like it would require a lot more allocations and copies (thankfully only of pointers, not entire buffers) than simply doing a push_back on the vector. On Wed, Jan 8, 2014 at 2:01 PM, Lindley French lindl...@gmail.comwrote: One additional minor suggestion: In the poller class, the add() and remove() overloads that take simple ints should probably accept them by value rather than const reference. There's no performance penalty, and it would make the type system play a bit nicer in some cases. For instance, I just tried to add a Boost.Asio UDP socket's native_handle() to a zmqpp poller (as a standin until UDP is working in 0MQ), and found that I couldn't do it if I was calling native_handle() from a const reference, but it worked fine with a non-const reference. On Wed, Jan 8, 2014 at 5:59 AM, Ben Gray b...@benjamg.com wrote: Pull request https://github.com/zeromq/zmqpp/pull/29 raised for this part of the issue (and the assert). On 7 January 2014 16:24, Lindley French lindl...@gmail.com wrote: In zmqpp, the socket_t::receive() method throws an exception if you pass it a non-empty message_t. This seems to be contradictory to the philosophy of zmq_recv(), which states that the message passed to it will be deallocated cleanly if necessary. Now, I know zmqpp defines a message as all parts of a multipart message, while libzmq defines it as just one part. So this isn't a direct contradiction of the API semantics. But it seems like this could be fixed without much effort. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
In the short term that may be the only solution that's workable without root access on at least some of the phones. I think the full benefits of EdgeNet won't be realized until arbitrary phones can serve as routers or data mules without the user knowing or caring, though. I still think WiFi Direct has a place in the solution, when it's available. According to Wikipedia, only one phone of a group has to support WiFi Direct for the entire group to use it. On Thu, Jan 9, 2014 at 9:19 AM, Pieter Hintjens p...@imatix.com wrote: Ad-hoc mode is too poorly supported to depend on. And it's slow. In any case a phone cannot act as a hotspot for any length of time. On some firmwares enabling AP mode will switch off 3G. This is still fine for us. My idea was to put this decision at the user level, initially. The application chooses to host or to join some activity. It's a natural pattern in a group of people. The host of a meeting is less likely to walk out the door, and more likely to have power. On Thu, Jan 9, 2014 at 3:14 PM, Lindley French lindl...@gmail.com wrote: There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. Battery life may be more of an issue than you realize. Wifi radios use far less power than 3G radios, but they still *do* use power, and the pattern of network activity can have a strong effect on battery life. The most important thing to realize is that people expect their smartphones not to use much power while they're idle. If you're doing network activity all the time in the background, that isn't the case, and people will become unhappy with your software and uninstall it. (Note the reactions to the first release of Google Now, which committed a similar sin with location services. GPS receivers use a lot of power too.) To some extent you can be smart enough to reduce network activity significantly while the phone is idle compared to active. Beacon less often, include a parameter in the beacon letting other phones know you'll only be listening for a short window after a beacon, etc. It's possible to give the radio a break. My concern is that once a phone is trying to host an access point, it gets a lot harder to be selective about when the wifi radio is active. It has to be on pretty much all the time. The exception might be ad-hoc mode-once one node creates an ad-hoc network (which is surprisingly tricky on Android), others can sustain the SSID without it. However, if the anchoring node was providing DHCP services, then obviously that won't work anymore once it goes offline. On Thu, Jan 9, 2014 at 3:06 AM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 8:24 AM, crocket crockabisc...@gmail.com wrote: Nowadays, almost every city is covered with WiFi hotspots. And, almost every modern building and meetup is covered well by WiFi hotspots. Most houses have a hotspot or two. Why don't we take advantage of existing WiFi hotspots? It would make things so much easier while we devise a way to connect phones directly or wait for 802.11s. You can, of course. There are some aspects to take into account: * It works perfectly e.g. in the home or office where people naturally put devices on their access point. * Many public hotspots, especially in the US, block client-to-client traffic. * Public hotspots are trivial to tap, if you care about anonymity. * There really is no cost to using a smartphone as hotspot except a little delay and some battery life, while switched on. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] Router socket to force peer disconnect
I'd like to. But it doesn't seem to be possible with Java or CLR at the moment. And we don't have time/budget to invest in support right now. Thanks for answer. On Thu, Jan 9, 2014 at 6:19 PM, Pieter Hintjens p...@imatix.com wrote: If you don't trust your peers you should probably be using ZeroMQ 4.x and security. Then you will only see authenticated peers. On Wed, Jan 8, 2014 at 12:32 PM, Dmitriy Vsekhvalnov dvsekhval...@gmail.com wrote: Hi All, continue my little evaluation of zeromq and wanted to ask next question: is it possible to force disconnect of client from router side? (e.g. disconnect req or dealer). If it is not possible are those clients who connected to router socket consume resources? A lot? Even if no messages send/received from them? We are in environment where we don't trust peers until they complete handshake with router socket. And would like to disconnect them if handshake failed. (i know we can simply ignore them and keep connected). Thanks. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] [zmqpp] [libzmq] Wrapping zmq messages
After much head-banging, valgrinding, I've finally identified the problem.
Since msg stores the zmq_msg_t objects in a vector, and small enough
messages are stored directly in the zmq_msg_t rather than by pointer,
calling raw_new_msg() was actually invalidating some of the pointers I have
previously obtained from zmq_msg_data().
I was able to solve the problem by moving the emplace_back() to a separate
loop.
On Wed, Jan 8, 2014 at 4:44 PM, Lindley French lindl...@gmail.com wrote:
Until we get UDP up and running in 0MQ, I'm using Boost.Asio for my UDP
needs. However, my API uses zmqpp::message_t types, so I need to convert
between these. This is easy enough when sending, since a
boost::asio::buffer only needs a pointer and a size and these can be
obtained from zmq message parts easily enough.
On the receiving side, it's less clear to me what to do. zmqpp::message_t
doesn't provide a way to add an uninitalized message part of a specified
size (IMO, raw_new_msg ought to take a size parameter with a default of 0),
and libzmq doesn't provide a simple way to resize and initialized message,
so I'm doing this:
vectormutable_buffer buffers; // ASIO scatter buffers
buffers.reserve(nparts);
// I've determined the number of zmq message parts to expect previously
for (size_t i = 0; i nparts; ++i)
{
zmq_msg_t inplace = msg.raw_new_msg();
{
zmq_msg_t newpart;
zmq_msg_init_size(newpart,sizes[i+1]);
zmq_msg_move(inplace,newpart);
zmq_msg_close(newpart);
}
buffers.emplace_back(zmq_msg_data(inplace),
zmq_msg_size(inplace));
}
asio_udp.receive(buffers);
However, I'm getting memory faults and I suspect this logic is to blame.
Anyone see any problems with it?
___
zeromq-dev mailing list
zeromq-dev@lists.zeromq.org
http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
On Thu, Jan 9, 2014 at 3:55 PM, Lindley French lindl...@gmail.com wrote: In the short term that may be the only solution that's workable without root access on at least some of the phones. I think the full benefits of EdgeNet won't be realized until arbitrary phones can serve as routers or data mules without the user knowing or caring, though. Right. I still think WiFi Direct has a place in the solution, when it's available. According to Wikipedia, only one phone of a group has to support WiFi Direct for the entire group to use it. That works precisely like a hotspot afaik (did not actually try that, though). -Pieter ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] Router socket to force peer disconnect
JZMQ runs on ZeroMQ v4.0 and supports the security layers. On Thu, Jan 9, 2014 at 5:03 PM, Dmitriy Vsekhvalnov dvsekhval...@gmail.com wrote: I'd like to. But it doesn't seem to be possible with Java or CLR at the moment. And we don't have time/budget to invest in support right now. Thanks for answer. On Thu, Jan 9, 2014 at 6:19 PM, Pieter Hintjens p...@imatix.com wrote: If you don't trust your peers you should probably be using ZeroMQ 4.x and security. Then you will only see authenticated peers. On Wed, Jan 8, 2014 at 12:32 PM, Dmitriy Vsekhvalnov dvsekhval...@gmail.com wrote: Hi All, continue my little evaluation of zeromq and wanted to ask next question: is it possible to force disconnect of client from router side? (e.g. disconnect req or dealer). If it is not possible are those clients who connected to router socket consume resources? A lot? Even if no messages send/received from them? We are in environment where we don't trust peers until they complete handshake with router socket. And would like to disconnect them if handshake failed. (i know we can simply ignore them and keep connected). Thanks. ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
[zeromq-dev] vs110 via msbuild vs120 x64 a no go
In case anyone tries this, a recent update from Microsoft has some how blocked this combination and causes a new unique error message: C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V120\Microsoft.Cpp.Platform.targets(58,5): error MSB8032: The Platform or PlatformToolset is not available from a 64bit environment. Consider building from 32bit environment instead. -- Steve-o ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
http://developer.android.com/guide/topics/connectivity/wifip2p.html tells android has WiFi Direct capability since 4.0. This is a good news. I'm not sure about iOS, but iPhones can connect to WiFi-Direct-capable android phones. On Fri, Jan 10, 2014 at 5:31 AM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 3:55 PM, Lindley French lindl...@gmail.com wrote: In the short term that may be the only solution that's workable without root access on at least some of the phones. I think the full benefits of EdgeNet won't be realized until arbitrary phones can serve as routers or data mules without the user knowing or caring, though. Right. I still think WiFi Direct has a place in the solution, when it's available. According to Wikipedia, only one phone of a group has to support WiFi Direct for the entire group to use it. That works precisely like a hotspot afaik (did not actually try that, though). -Pieter ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
Re: [zeromq-dev] How do we activate and connect to WiFi hotspots in edgenets?
Plus, I tested my nexus 5 and 7 with Super Beam app. Super Beam shares files and folders using WiFi Direct. My nexus devices successfully shared files via WiFi Direct. WiFi Direct is confirmed to be working at least on nexus devices. On Fri, Jan 10, 2014 at 5:31 AM, Pieter Hintjens p...@imatix.com wrote: On Thu, Jan 9, 2014 at 3:55 PM, Lindley French lindl...@gmail.com wrote: In the short term that may be the only solution that's workable without root access on at least some of the phones. I think the full benefits of EdgeNet won't be realized until arbitrary phones can serve as routers or data mules without the user knowing or caring, though. Right. I still think WiFi Direct has a place in the solution, when it's available. According to Wikipedia, only one phone of a group has to support WiFi Direct for the entire group to use it. That works precisely like a hotspot afaik (did not actually try that, though). -Pieter ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
[zeromq-dev] STREAM socket not closing.
Dear list, I’m looking at using the STREAM socket to interface with some plain old TCP clients, however, there seems to be two things I’m missing: 1. How do I send a message to a client first? Once the client sends a message I can get the identity frame and use that to reply. How do I send a message before receiving any data? 2. Closing sockets. I see in the docs that a socket can be closed by sending the identity frame followed by a zero length frame. However, this does not seem to be working. I’m expecting to successfully read 0 bytes from the TCP socket meaning that the connection was actively closed, and no error codes. Instead by TCP socket recv blocks forever. The only way I can get this to happen is to close the ZMQ socket, which is not possible when I am using bind and talking to multiple TCP clients. Cheers, Matt ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
[zeromq-dev] Notification of peer disconnection for ZMQ_STREAM sockets.
In CZMQ and PYZMQ you can use the socket event monitor interface to receive socket event notification of things such as connect, disconnect, accepted, etc. I use the zmonitor from CZMQ and zmq.utils.monitor.recv_monitor_message from pyzmq for receiving connection state information in applications where the connection state is important. Perhaps you could check if this would satisfy your needs. Regards, Chris ___ zeromq-dev mailing list zeromq-dev@lists.zeromq.org http://lists.zeromq.org/mailman/listinfo/zeromq-dev
