From: Elod K. <kir...@gr...> - 2008-01-30 14:15:53
|
Hi! I'm trying to use PULL data mode in Player 2.0.5 using Stage 2.0.4. I send the PULL mode request message to the server. Then I set the replace rule for all data messages to have them replaced on the message queue by newer messages as described at: http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html Everything seems to work fine, I don't get any "*push onto full message queue error*" text. The only problem is, that when I send a PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No data message arrives to me, although I'm subscribed to the bumber device on my robot (simulated in stage). Any clues what I'm doing wrong or how I could debug this (perhaps enable console debug logging in player/stage to see what is actually pushed onto the message queue)? Thanks, Elod |
From: Elod K. <kir...@gr...> - 2008-02-05 09:18:04
|
Any ideo on this problem please? Elod > Hi! > > I'm trying to use PULL data mode in Player 2.0.5 using Stage 2.0.4. I > send the PULL mode request message to the server. Then I set the replace > rule for all data messages to have them replaced on the message queue by > newer messages as described at: > > http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html > > Everything seems to work fine, I don't get any "*push onto full message > queue error*" text. The only problem is, that when I send a > PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No data > message arrives to me, although I'm subscribed to the bumber device on > my robot (simulated in stage). Any clues what I'm doing wrong or how I > could debug this (perhaps enable console debug logging in player/stage > to see what is actually pushed onto the message queue)? > > Thanks, > > Elod > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users > |
From: Toby C. <tco...@pl...> - 2008-02-06 02:40:26
|
Did you get bumper data in PUSH mode? Toby Elod Kironsky wrote: > Any ideo on this problem please? > > Elod > >> Hi! >> >> I'm trying to use PULL data mode in Player 2.0.5 using Stage 2.0.4. I >> send the PULL mode request message to the server. Then I set the replace >> rule for all data messages to have them replaced on the message queue by >> newer messages as described at: >> >> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >> >> Everything seems to work fine, I don't get any "*push onto full message >> queue error*" text. The only problem is, that when I send a >> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No data >> message arrives to me, although I'm subscribed to the bumber device on >> my robot (simulated in stage). Any clues what I'm doing wrong or how I >> could debug this (perhaps enable console debug logging in player/stage >> to see what is actually pushed onto the message queue)? >> >> Thanks, >> >> Elod >> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >> _______________________________________________ >> Playerstage-users mailing list >> Pla...@li... >> https://lists.sourceforge.net/lists/listinfo/playerstage-users >> >> > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users > > |
From: Elod K. <elo...@av...> - 2008-02-06 12:29:10
|
Yes, I get the bumber data without any problems. Also in PULL mode, I get the data when requesting for it, but as soon as I set the replace rule I get only SYNCH messages. Is there any setting in cfg file that I could use to have the bumber print debug messages? Is the order of commands sending to the server significant? I mean the order of actions: subscribe to the bumber, set data mode, set replace mode? Elod > Did you get bumper data in PUSH mode? > > Toby > > Elod Kironsky wrote: > >> Any ideo on this problem please? >> >> Elod >> >> >>> Hi! >>> >>> I'm trying to use PULL data mode in Player 2.0.5 using Stage 2.0.4. I >>> send the PULL mode request message to the server. Then I set the replace >>> rule for all data messages to have them replaced on the message queue by >>> newer messages as described at: >>> >>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >>> >>> Everything seems to work fine, I don't get any "*push onto full message >>> queue error*" text. The only problem is, that when I send a >>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No data >>> message arrives to me, although I'm subscribed to the bumber device on >>> my robot (simulated in stage). Any clues what I'm doing wrong or how I >>> could debug this (perhaps enable console debug logging in player/stage >>> to see what is actually pushed onto the message queue)? >>> >>> Thanks, >>> >>> Elod >>> >>> ------------------------------------------------------------------------- >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _______________________________________________ >>> Playerstage-users mailing list >>> Pla...@li... >>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>> >>> >>> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >> _______________________________________________ >> Playerstage-users mailing list >> Pla...@li... >> https://lists.sourceforge.net/lists/listinfo/playerstage-users >> >> >> > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users > -- Elod Kironsky Developer AVG Technologies Email: elo...@av... www.avg.com www.avg.cz |
From: Kevin B. <ba...@gm...> - 2008-02-06 17:44:13
|
> order of actions: subscribe to the bumber, set data mode, set replace mode? I would try changing the order. I'm not sure it matters, but when I've done it, I always did set data mode first, then subscribe. Below is some code that I've used (It's in MATLAB not C so don't copy/paste, but the syntax is close so you'll get the idea) client = player('client_create', config.player_host, config.player_port); player('client_connect', client); player('client_datamode', client, PLAYER_DATAMODE_PULL); player('client_set_replace_rule', client, -1, -1, PLAYER_MSGTYPE_DATA, -1, 1); camera = player('camera_create', client, config.player_camera_num); player('camera_subscribe', camera, PLAYER_OPEN_MODE); On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> wrote: > Yes, I get the bumber data without any problems. Also in PULL mode, I > get the data when requesting for it, but > as soon as I set the replace rule I get only SYNCH messages. Is there > any setting in cfg file that I could use to > have the bumber print debug messages? Is the order of commands sending > to the server significant? I mean the > order of actions: subscribe to the bumber, set data mode, set replace mode? > > Elod > > > Did you get bumper data in PUSH mode? > > > > Toby > > > > Elod Kironsky wrote: > > > >> Any ideo on this problem please? > >> > >> Elod > >> > >> > >>> Hi! > >>> > >>> I'm trying to use PULL data mode in Player 2.0.5 using Stage 2.0.4. I > >>> send the PULL mode request message to the server. Then I set the replace > >>> rule for all data messages to have them replaced on the message queue by > >>> newer messages as described at: > >>> > >>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html > >>> > >>> Everything seems to work fine, I don't get any "*push onto full message > >>> queue error*" text. The only problem is, that when I send a > >>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No data > >>> message arrives to me, although I'm subscribed to the bumber device on > >>> my robot (simulated in stage). Any clues what I'm doing wrong or how I > >>> could debug this (perhaps enable console debug logging in player/stage > >>> to see what is actually pushed onto the message queue)? > >>> > >>> Thanks, > >>> > >>> Elod > >>> > >>> ------------------------------------------------------------------------- > >>> This SF.net email is sponsored by: Microsoft > >>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>> _______________________________________________ > >>> Playerstage-users mailing list > >>> Pla...@li... > >>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>> > >>> > >>> > >> ------------------------------------------------------------------------- > >> This SF.net email is sponsored by: Microsoft > >> Defy all challenges. Microsoft(R) Visual Studio 2008. > >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >> _______________________________________________ > >> Playerstage-users mailing list > >> Pla...@li... > >> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >> > >> > >> > > > > > > ------------------------------------------------------------------------- > > This SF.net email is sponsored by: Microsoft > > Defy all challenges. Microsoft(R) Visual Studio 2008. > > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > _______________________________________________ > > Playerstage-users mailing list > > Pla...@li... > > https://lists.sourceforge.net/lists/listinfo/playerstage-users > > > > > -- > Elod Kironsky > Developer > AVG Technologies > Email: elo...@av... > www.avg.com > www.avg.cz > > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users > |
From: Brian G. <br...@ge...> - 2008-02-06 21:03:31
|
Changing the order, as Kevin suggests, is definitely worth a try. I suspect that we're not properly handling the case when data is already on the socket when the data mode and/or replacement rules are changed by the client. So, set everything up, *then* subscribe. Here's a relevant post from a while back: https://sourceforge.net/mailarchive/message.php?msg_name=14825455.post%40talk.nabble.com brian. Kevin Barry wrote: >> order of actions: subscribe to the bumber, set data mode, set replace mode? > > I would try changing the order. I'm not sure it matters, but when I've > done it, I always did set data mode first, then subscribe. Below is > some code that I've used (It's in MATLAB not C so don't copy/paste, > but the syntax is close so you'll get the idea) > > > client = player('client_create', config.player_host, config.player_port); > player('client_connect', client); > > player('client_datamode', client, PLAYER_DATAMODE_PULL); > player('client_set_replace_rule', client, -1, -1, PLAYER_MSGTYPE_DATA, -1, 1); > > camera = player('camera_create', client, config.player_camera_num); > player('camera_subscribe', camera, PLAYER_OPEN_MODE); > > > > On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> wrote: >> Yes, I get the bumber data without any problems. Also in PULL mode, I >> get the data when requesting for it, but >> as soon as I set the replace rule I get only SYNCH messages. Is there >> any setting in cfg file that I could use to >> have the bumber print debug messages? Is the order of commands sending >> to the server significant? I mean the >> order of actions: subscribe to the bumber, set data mode, set replace mode? >> >> Elod >> >>> Did you get bumper data in PUSH mode? >>> >>> Toby >>> >>> Elod Kironsky wrote: >>> >>>> Any ideo on this problem please? >>>> >>>> Elod >>>> >>>> >>>>> Hi! >>>>> >>>>> I'm trying to use PULL data mode in Player 2.0.5 using Stage 2.0.4. I >>>>> send the PULL mode request message to the server. Then I set the replace >>>>> rule for all data messages to have them replaced on the message queue by >>>>> newer messages as described at: >>>>> >>>>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >>>>> >>>>> Everything seems to work fine, I don't get any "*push onto full message >>>>> queue error*" text. The only problem is, that when I send a >>>>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No data >>>>> message arrives to me, although I'm subscribed to the bumber device on >>>>> my robot (simulated in stage). Any clues what I'm doing wrong or how I >>>>> could debug this (perhaps enable console debug logging in player/stage >>>>> to see what is actually pushed onto the message queue)? >>>>> >>>>> Thanks, >>>>> >>>>> Elod >>>>> >>>>> ------------------------------------------------------------------------- >>>>> This SF.net email is sponsored by: Microsoft >>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>> _______________________________________________ >>>>> Playerstage-users mailing list >>>>> Pla...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>> >>>>> >>>>> >>>> ------------------------------------------------------------------------- >>>> This SF.net email is sponsored by: Microsoft >>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>> _______________________________________________ >>>> Playerstage-users mailing list >>>> Pla...@li... >>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>> >>>> >>>> >>> >>> ------------------------------------------------------------------------- >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _______________________________________________ >>> Playerstage-users mailing list >>> Pla...@li... >>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>> >> >> -- >> Elod Kironsky >> Developer >> AVG Technologies >> Email: elo...@av... >> www.avg.com >> www.avg.cz >> >> >> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >> _______________________________________________ >> Playerstage-users mailing list >> Pla...@li... >> https://lists.sourceforge.net/lists/listinfo/playerstage-users >> > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users |
From: Elod K. <elo...@av...> - 2008-02-07 12:11:59
|
I tried to change the order to (1) set data mode to PULL, (2) set replace rule, (3) subscribe to bumpers. It did not work. However I figured out, that if I pull the packages in a shorter interval, some data occasionally is sent from the server. The shorter the interval the more data arrives. Originally I was pulling data every 5 secs, just to test if everything is fine, but I only got SYNCH messages. If I set the pull interval to 0.5 sec, I started to receive data packages, but *not* for every data request. Is there a timeout for data packages waiting on the queue for an interface? If yes, how can I change it? Elod > Changing the order, as Kevin suggests, is definitely worth a try. I > suspect that we're not properly handling the case when data is already > on the socket when the data mode and/or replacement rules are changed by > the client. So, set everything up, *then* subscribe. > > Here's a relevant post from a while back: > > https://sourceforge.net/mailarchive/message.php?msg_name=14825455.post%40talk.nabble.com > > brian. > > Kevin Barry wrote: > >>> order of actions: subscribe to the bumber, set data mode, set replace mode? >>> >> I would try changing the order. I'm not sure it matters, but when I've >> done it, I always did set data mode first, then subscribe. Below is >> some code that I've used (It's in MATLAB not C so don't copy/paste, >> but the syntax is close so you'll get the idea) >> >> >> client = player('client_create', config.player_host, config.player_port); >> player('client_connect', client); >> >> player('client_datamode', client, PLAYER_DATAMODE_PULL); >> player('client_set_replace_rule', client, -1, -1, PLAYER_MSGTYPE_DATA, -1, 1); >> >> camera = player('camera_create', client, config.player_camera_num); >> player('camera_subscribe', camera, PLAYER_OPEN_MODE); >> >> >> >> On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> wrote: >> >>> Yes, I get the bumber data without any problems. Also in PULL mode, I >>> get the data when requesting for it, but >>> as soon as I set the replace rule I get only SYNCH messages. Is there >>> any setting in cfg file that I could use to >>> have the bumber print debug messages? Is the order of commands sending >>> to the server significant? I mean the >>> order of actions: subscribe to the bumber, set data mode, set replace mode? >>> >>> Elod >>> >>> >>>> Did you get bumper data in PUSH mode? >>>> >>>> Toby >>>> >>>> Elod Kironsky wrote: >>>> >>>> >>>>> Any ideo on this problem please? >>>>> >>>>> Elod >>>>> >>>>> >>>>> >>>>>> Hi! >>>>>> >>>>>> I'm trying to use PULL data mode in Player 2.0.5 using Stage 2.0.4. I >>>>>> send the PULL mode request message to the server. Then I set the replace >>>>>> rule for all data messages to have them replaced on the message queue by >>>>>> newer messages as described at: >>>>>> >>>>>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >>>>>> >>>>>> Everything seems to work fine, I don't get any "*push onto full message >>>>>> queue error*" text. The only problem is, that when I send a >>>>>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No data >>>>>> message arrives to me, although I'm subscribed to the bumber device on >>>>>> my robot (simulated in stage). Any clues what I'm doing wrong or how I >>>>>> could debug this (perhaps enable console debug logging in player/stage >>>>>> to see what is actually pushed onto the message queue)? >>>>>> >>>>>> Thanks, >>>>>> >>>>>> Elod >>>>>> >>>>>> ------------------------------------------------------------------------- >>>>>> This SF.net email is sponsored by: Microsoft >>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>> _______________________________________________ >>>>>> Playerstage-users mailing list >>>>>> Pla...@li... >>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>> >>>>>> >>>>>> >>>>>> >>>>> ------------------------------------------------------------------------- >>>>> This SF.net email is sponsored by: Microsoft >>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>> _______________________________________________ >>>>> Playerstage-users mailing list >>>>> Pla...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>> >>>>> >>>>> >>>>> >>>> ------------------------------------------------------------------------- >>>> This SF.net email is sponsored by: Microsoft >>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>> _______________________________________________ >>>> Playerstage-users mailing list >>>> Pla...@li... >>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>> >>>> >>> -- >>> Elod Kironsky >>> Developer >>> AVG Technologies >>> Email: elo...@av... >>> www.avg.com >>> www.avg.cz >>> >>> >>> >>> ------------------------------------------------------------------------- >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _______________________________________________ >>> Playerstage-users mailing list >>> Pla...@li... >>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>> >>> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >> _______________________________________________ >> Playerstage-users mailing list >> Pla...@li... >> https://lists.sourceforge.net/lists/listinfo/playerstage-users >> > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users > -- Elod Kironsky Developer AVG Technologies Email: elo...@av... www.avg.com www.avg.cz |
From: Brian G. <br...@ge...> - 2008-03-03 21:13:14
|
How do you know that you're not getting data for the bumpers? Have you tried looking at the playerc_bumper_t structure? It gets filled in as a side-effect of reading (this is actually libplayerc's default callback in action). You can test for fresh data in any libplayerc device proxy by examining the 'info.fresh' field. brian. On Feb 7, 2008, at 4:11 AM, Elod Kironsky wrote: > I tried to change the order to (1) set data mode to PULL, (2) set > replace rule, (3) subscribe to bumpers. It did > not work. However I figured out, that if I pull the packages in a > shorter interval, some data occasionally is sent > from the server. The shorter the interval the more data arrives. > Originally I was pulling data every 5 secs, just to > test if everything is fine, but I only got SYNCH messages. If I set > the > pull interval to 0.5 sec, I started to receive > data packages, but *not* for every data request. Is there a timeout > for > data packages waiting on the queue for an > interface? If yes, how can I change it? > > Elod >> Changing the order, as Kevin suggests, is definitely worth a try. I >> suspect that we're not properly handling the case when data is >> already >> on the socket when the data mode and/or replacement rules are >> changed by >> the client. So, set everything up, *then* subscribe. >> >> Here's a relevant post from a while back: >> >> https://sourceforge.net/mailarchive/message.php?msg_name=14825455.post%40talk.nabble.com >> >> brian. >> >> Kevin Barry wrote: >> >>>> order of actions: subscribe to the bumber, set data mode, set >>>> replace mode? >>>> >>> I would try changing the order. I'm not sure it matters, but when >>> I've >>> done it, I always did set data mode first, then subscribe. Below is >>> some code that I've used (It's in MATLAB not C so don't copy/paste, >>> but the syntax is close so you'll get the idea) >>> >>> >>> client = player('client_create', config.player_host, >>> config.player_port); >>> player('client_connect', client); >>> >>> player('client_datamode', client, PLAYER_DATAMODE_PULL); >>> player('client_set_replace_rule', client, -1, -1, >>> PLAYER_MSGTYPE_DATA, -1, 1); >>> >>> camera = player('camera_create', client, config.player_camera_num); >>> player('camera_subscribe', camera, PLAYER_OPEN_MODE); >>> >>> >>> >>> On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> wrote: >>> >>>> Yes, I get the bumber data without any problems. Also in PULL >>>> mode, I >>>> get the data when requesting for it, but >>>> as soon as I set the replace rule I get only SYNCH messages. Is >>>> there >>>> any setting in cfg file that I could use to >>>> have the bumber print debug messages? Is the order of commands >>>> sending >>>> to the server significant? I mean the >>>> order of actions: subscribe to the bumber, set data mode, set >>>> replace mode? >>>> >>>> Elod >>>> >>>> >>>>> Did you get bumper data in PUSH mode? >>>>> >>>>> Toby >>>>> >>>>> Elod Kironsky wrote: >>>>> >>>>> >>>>>> Any ideo on this problem please? >>>>>> >>>>>> Elod >>>>>> >>>>>> >>>>>> >>>>>>> Hi! >>>>>>> >>>>>>> I'm trying to use PULL data mode in Player 2.0.5 using Stage >>>>>>> 2.0.4. I >>>>>>> send the PULL mode request message to the server. Then I set >>>>>>> the replace >>>>>>> rule for all data messages to have them replaced on the >>>>>>> message queue by >>>>>>> newer messages as described at: >>>>>>> >>>>>>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >>>>>>> >>>>>>> Everything seems to work fine, I don't get any "*push onto >>>>>>> full message >>>>>>> queue error*" text. The only problem is, that when I send a >>>>>>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No >>>>>>> data >>>>>>> message arrives to me, although I'm subscribed to the bumber >>>>>>> device on >>>>>>> my robot (simulated in stage). Any clues what I'm doing wrong >>>>>>> or how I >>>>>>> could debug this (perhaps enable console debug logging in >>>>>>> player/stage >>>>>>> to see what is actually pushed onto the message queue)? >>>>>>> >>>>>>> Thanks, >>>>>>> >>>>>>> Elod >>>>>>> >>>>>>> ------------------------------------------------------------------------- >>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>> _______________________________________________ >>>>>>> Playerstage-users mailing list >>>>>>> Pla...@li... >>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>> ------------------------------------------------------------------------- >>>>>> This SF.net email is sponsored by: Microsoft >>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>> _______________________________________________ >>>>>> Playerstage-users mailing list >>>>>> Pla...@li... >>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>> >>>>>> >>>>>> >>>>>> >>>>> ------------------------------------------------------------------------- >>>>> This SF.net email is sponsored by: Microsoft >>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>> _______________________________________________ >>>>> Playerstage-users mailing list >>>>> Pla...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>> >>>>> >>>> -- >>>> Elod Kironsky >>>> Developer >>>> AVG Technologies >>>> Email: elo...@av... >>>> www.avg.com >>>> www.avg.cz >>>> >>>> >>>> >>>> ------------------------------------------------------------------------- >>>> This SF.net email is sponsored by: Microsoft >>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>> _______________________________________________ >>>> Playerstage-users mailing list >>>> Pla...@li... >>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>> >>>> >>> ------------------------------------------------------------------------- >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _______________________________________________ >>> Playerstage-users mailing list >>> Pla...@li... >>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>> >> >> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >> _______________________________________________ >> Playerstage-users mailing list >> Pla...@li... >> https://lists.sourceforge.net/lists/listinfo/playerstage-users >> > > > -- > Elod Kironsky > Developer > AVG Technologies > Email: elo...@av... > www.avg.com > www.avg.cz > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users |
From: Elod K. <elo...@av...> - 2008-03-03 22:48:24
|
Well, I'm printing every message from the server to the console (read header, read length number of bytes and print it). If I use shorter intervals, I get some bumper data messages but mostly only SYNCH messages. So that is why I suspect, that the data messages are simply timed out somewhere on the queue. I have my client written in Smalltalk, so I cannot examine the playerc_bumper_t structure, I use my own decoding methods that work with a bunch of bytes :-) Elod > How do you know that you're not getting data for the bumpers? Have > you tried looking at the playerc_bumper_t structure? It gets filled > in as a side-effect of reading (this is actually libplayerc's default > callback in action). You can test for fresh data in any libplayerc > device proxy by examining the 'info.fresh' field. > > brian. > > On Feb 7, 2008, at 4:11 AM, Elod Kironsky wrote: > >> I tried to change the order to (1) set data mode to PULL, (2) set >> replace rule, (3) subscribe to bumpers. It did >> not work. However I figured out, that if I pull the packages in a >> shorter interval, some data occasionally is sent >> from the server. The shorter the interval the more data arrives. >> Originally I was pulling data every 5 secs, just to >> test if everything is fine, but I only got SYNCH messages. If I set the >> pull interval to 0.5 sec, I started to receive >> data packages, but *not* for every data request. Is there a timeout for >> data packages waiting on the queue for an >> interface? If yes, how can I change it? >> >> Elod >>> Changing the order, as Kevin suggests, is definitely worth a try. I >>> suspect that we're not properly handling the case when data is already >>> on the socket when the data mode and/or replacement rules are >>> changed by >>> the client. So, set everything up, *then* subscribe. >>> >>> Here's a relevant post from a while back: >>> >>> https://sourceforge.net/mailarchive/message.php?msg_name=14825455.post%40talk.nabble.com >>> >>> >>> brian. >>> >>> Kevin Barry wrote: >>> >>>>> order of actions: subscribe to the bumber, set data mode, set >>>>> replace mode? >>>>> >>>> I would try changing the order. I'm not sure it matters, but when I've >>>> done it, I always did set data mode first, then subscribe. Below is >>>> some code that I've used (It's in MATLAB not C so don't copy/paste, >>>> but the syntax is close so you'll get the idea) >>>> >>>> >>>> client = player('client_create', config.player_host, >>>> config.player_port); >>>> player('client_connect', client); >>>> >>>> player('client_datamode', client, PLAYER_DATAMODE_PULL); >>>> player('client_set_replace_rule', client, -1, -1, >>>> PLAYER_MSGTYPE_DATA, -1, 1); >>>> >>>> camera = player('camera_create', client, config.player_camera_num); >>>> player('camera_subscribe', camera, PLAYER_OPEN_MODE); >>>> >>>> >>>> >>>> On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> wrote: >>>> >>>>> Yes, I get the bumber data without any problems. Also in PULL mode, I >>>>> get the data when requesting for it, but >>>>> as soon as I set the replace rule I get only SYNCH messages. Is there >>>>> any setting in cfg file that I could use to >>>>> have the bumber print debug messages? Is the order of commands >>>>> sending >>>>> to the server significant? I mean the >>>>> order of actions: subscribe to the bumber, set data mode, set >>>>> replace mode? >>>>> >>>>> Elod >>>>> >>>>> >>>>>> Did you get bumper data in PUSH mode? >>>>>> >>>>>> Toby >>>>>> >>>>>> Elod Kironsky wrote: >>>>>> >>>>>> >>>>>>> Any ideo on this problem please? >>>>>>> >>>>>>> Elod >>>>>>> >>>>>>> >>>>>>> >>>>>>>> Hi! >>>>>>>> >>>>>>>> I'm trying to use PULL data mode in Player 2.0.5 using Stage >>>>>>>> 2.0.4. I >>>>>>>> send the PULL mode request message to the server. Then I set >>>>>>>> the replace >>>>>>>> rule for all data messages to have them replaced on the message >>>>>>>> queue by >>>>>>>> newer messages as described at: >>>>>>>> >>>>>>>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >>>>>>>> >>>>>>>> >>>>>>>> Everything seems to work fine, I don't get any "*push onto full >>>>>>>> message >>>>>>>> queue error*" text. The only problem is, that when I send a >>>>>>>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No >>>>>>>> data >>>>>>>> message arrives to me, although I'm subscribed to the bumber >>>>>>>> device on >>>>>>>> my robot (simulated in stage). Any clues what I'm doing wrong >>>>>>>> or how I >>>>>>>> could debug this (perhaps enable console debug logging in >>>>>>>> player/stage >>>>>>>> to see what is actually pushed onto the message queue)? >>>>>>>> >>>>>>>> Thanks, >>>>>>>> >>>>>>>> Elod >>>>>>>> >>>>>>>> ------------------------------------------------------------------------- >>>>>>>> >>>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>>> _______________________________________________ >>>>>>>> Playerstage-users mailing list >>>>>>>> Pla...@li... >>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>> ------------------------------------------------------------------------- >>>>>>> >>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>> _______________________________________________ >>>>>>> Playerstage-users mailing list >>>>>>> Pla...@li... >>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>> ------------------------------------------------------------------------- >>>>>> >>>>>> This SF.net email is sponsored by: Microsoft >>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>> _______________________________________________ >>>>>> Playerstage-users mailing list >>>>>> Pla...@li... >>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>> >>>>>> >>>>> -- >>>>> Elod Kironsky >>>>> Developer >>>>> AVG Technologies >>>>> Email: elo...@av... >>>>> www.avg.com >>>>> www.avg.cz >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------------- >>>>> >>>>> This SF.net email is sponsored by: Microsoft >>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>> _______________________________________________ >>>>> Playerstage-users mailing list >>>>> Pla...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>> >>>>> >>>> ------------------------------------------------------------------------- >>>> >>>> This SF.net email is sponsored by: Microsoft >>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>> _______________________________________________ >>>> Playerstage-users mailing list >>>> Pla...@li... >>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>> >>> >>> >>> ------------------------------------------------------------------------- >>> >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _______________________________________________ >>> Playerstage-users mailing list >>> Pla...@li... >>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>> >> >> >> -- >> Elod Kironsky >> Developer >> AVG Technologies >> Email: elo...@av... >> www.avg.com >> www.avg.cz >> >> >> ------------------------------------------------------------------------- >> >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >> _______________________________________________ >> Playerstage-users mailing list >> Pla...@li... >> https://lists.sourceforge.net/lists/listinfo/playerstage-users > -- Elod Kironsky Developer AVG Technologies Email: elo...@av... www.avg.com www.avg.cz |
From: Brian G. <br...@ge...> - 2008-03-03 22:56:19
|
On Mar 3, 2008, at 2:48 PM, Elod Kironsky wrote: > Well, I'm printing every message from the server to the console (read > header, read length number of bytes and print it). If I use shorter > intervals, I get some bumper data messages but mostly only SYNCH > messages. So that is why I suspect, that the data messages are simply > timed out somewhere on the queue. I have my client written in > Smalltalk, > so I cannot examine the playerc_bumper_t structure, I use my own > decoding methods that work with a bunch of bytes :-) Ah. Well, there's no timeout in the queues on the server side. Once a message is put onto a client's outgoing queue, it will either be delivered to the client, or it will be overwritten by a subsequent message with the same signature (replacement rule depending). In PULL mode with replacement, if you poll at a rate less than the rate at which the underlying driver publishes the data, then you should get a new data message (followed by a SYNCH) every cycle. Of course, if you read faster, then sometimes there won't be new data for you, and so you'll just get the SYNCH. If you're using Stage, then the data publication rate should be about 10Hz. brian. >> How do you know that you're not getting data for the bumpers? Have >> you tried looking at the playerc_bumper_t structure? It gets filled >> in as a side-effect of reading (this is actually libplayerc's default >> callback in action). You can test for fresh data in any libplayerc >> device proxy by examining the 'info.fresh' field. >> >> brian. >> >> On Feb 7, 2008, at 4:11 AM, Elod Kironsky wrote: >> >>> I tried to change the order to (1) set data mode to PULL, (2) set >>> replace rule, (3) subscribe to bumpers. It did >>> not work. However I figured out, that if I pull the packages in a >>> shorter interval, some data occasionally is sent >>> from the server. The shorter the interval the more data arrives. >>> Originally I was pulling data every 5 secs, just to >>> test if everything is fine, but I only got SYNCH messages. If I >>> set the >>> pull interval to 0.5 sec, I started to receive >>> data packages, but *not* for every data request. Is there a >>> timeout for >>> data packages waiting on the queue for an >>> interface? If yes, how can I change it? >>> >>> Elod >>>> Changing the order, as Kevin suggests, is definitely worth a >>>> try. I >>>> suspect that we're not properly handling the case when data is >>>> already >>>> on the socket when the data mode and/or replacement rules are >>>> changed by >>>> the client. So, set everything up, *then* subscribe. >>>> >>>> Here's a relevant post from a while back: >>>> >>>> https://sourceforge.net/mailarchive/message.php?msg_name=14825455.post%40talk.nabble.com >>>> >>>> >>>> brian. >>>> >>>> Kevin Barry wrote: >>>> >>>>>> order of actions: subscribe to the bumber, set data mode, set >>>>>> replace mode? >>>>>> >>>>> I would try changing the order. I'm not sure it matters, but >>>>> when I've >>>>> done it, I always did set data mode first, then subscribe. Below >>>>> is >>>>> some code that I've used (It's in MATLAB not C so don't copy/ >>>>> paste, >>>>> but the syntax is close so you'll get the idea) >>>>> >>>>> >>>>> client = player('client_create', config.player_host, >>>>> config.player_port); >>>>> player('client_connect', client); >>>>> >>>>> player('client_datamode', client, PLAYER_DATAMODE_PULL); >>>>> player('client_set_replace_rule', client, -1, -1, >>>>> PLAYER_MSGTYPE_DATA, -1, 1); >>>>> >>>>> camera = player('camera_create', client, >>>>> config.player_camera_num); >>>>> player('camera_subscribe', camera, PLAYER_OPEN_MODE); >>>>> >>>>> >>>>> >>>>> On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> >>>>> wrote: >>>>> >>>>>> Yes, I get the bumber data without any problems. Also in PULL >>>>>> mode, I >>>>>> get the data when requesting for it, but >>>>>> as soon as I set the replace rule I get only SYNCH messages. Is >>>>>> there >>>>>> any setting in cfg file that I could use to >>>>>> have the bumber print debug messages? Is the order of commands >>>>>> sending >>>>>> to the server significant? I mean the >>>>>> order of actions: subscribe to the bumber, set data mode, set >>>>>> replace mode? >>>>>> >>>>>> Elod >>>>>> >>>>>> >>>>>>> Did you get bumper data in PUSH mode? >>>>>>> >>>>>>> Toby >>>>>>> >>>>>>> Elod Kironsky wrote: >>>>>>> >>>>>>> >>>>>>>> Any ideo on this problem please? >>>>>>>> >>>>>>>> Elod >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>>> Hi! >>>>>>>>> >>>>>>>>> I'm trying to use PULL data mode in Player 2.0.5 using Stage >>>>>>>>> 2.0.4. I >>>>>>>>> send the PULL mode request message to the server. Then I set >>>>>>>>> the replace >>>>>>>>> rule for all data messages to have them replaced on the >>>>>>>>> message >>>>>>>>> queue by >>>>>>>>> newer messages as described at: >>>>>>>>> >>>>>>>>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >>>>>>>>> >>>>>>>>> >>>>>>>>> Everything seems to work fine, I don't get any "*push onto >>>>>>>>> full >>>>>>>>> message >>>>>>>>> queue error*" text. The only problem is, that when I send a >>>>>>>>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No >>>>>>>>> data >>>>>>>>> message arrives to me, although I'm subscribed to the bumber >>>>>>>>> device on >>>>>>>>> my robot (simulated in stage). Any clues what I'm doing wrong >>>>>>>>> or how I >>>>>>>>> could debug this (perhaps enable console debug logging in >>>>>>>>> player/stage >>>>>>>>> to see what is actually pushed onto the message queue)? >>>>>>>>> >>>>>>>>> Thanks, >>>>>>>>> >>>>>>>>> Elod >>>>>>>>> >>>>>>>>> ------------------------------------------------------------------------- >>>>>>>>> >>>>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>>>> _______________________________________________ >>>>>>>>> Playerstage-users mailing list >>>>>>>>> Pla...@li... >>>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>> ------------------------------------------------------------------------- >>>>>>>> >>>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>>> _______________________________________________ >>>>>>>> Playerstage-users mailing list >>>>>>>> Pla...@li... >>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>> ------------------------------------------------------------------------- >>>>>>> >>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>> _______________________________________________ >>>>>>> Playerstage-users mailing list >>>>>>> Pla...@li... >>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>> >>>>>>> >>>>>> -- >>>>>> Elod Kironsky >>>>>> Developer >>>>>> AVG Technologies >>>>>> Email: elo...@av... >>>>>> www.avg.com >>>>>> www.avg.cz >>>>>> >>>>>> >>>>>> >>>>>> ------------------------------------------------------------------------- >>>>>> >>>>>> This SF.net email is sponsored by: Microsoft >>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>> _______________________________________________ >>>>>> Playerstage-users mailing list >>>>>> Pla...@li... >>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>> >>>>>> >>>>> ------------------------------------------------------------------------- >>>>> >>>>> This SF.net email is sponsored by: Microsoft >>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>> _______________________________________________ >>>>> Playerstage-users mailing list >>>>> Pla...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>> >>>> >>>> >>>> ------------------------------------------------------------------------- >>>> >>>> This SF.net email is sponsored by: Microsoft >>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>> _______________________________________________ >>>> Playerstage-users mailing list >>>> Pla...@li... >>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>> >>> >>> >>> -- >>> Elod Kironsky >>> Developer >>> AVG Technologies >>> Email: elo...@av... >>> www.avg.com >>> www.avg.cz >>> >>> >>> ------------------------------------------------------------------------- >>> >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _______________________________________________ >>> Playerstage-users mailing list >>> Pla...@li... >>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >> > > > -- > Elod Kironsky > Developer > AVG Technologies > Email: elo...@av... > www.avg.com > www.avg.cz > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users |
From: Elod K. <elo...@av...> - 2008-03-04 10:55:11
|
Brian Gerkey wrote: > > On Mar 3, 2008, at 2:48 PM, Elod Kironsky wrote: > >> Well, I'm printing every message from the server to the console (read >> header, read length number of bytes and print it). If I use shorter >> intervals, I get some bumper data messages but mostly only SYNCH >> messages. So that is why I suspect, that the data messages are simply >> timed out somewhere on the queue. I have my client written in Smalltalk, >> so I cannot examine the playerc_bumper_t structure, I use my own >> decoding methods that work with a bunch of bytes :-) > > Ah. Well, there's no timeout in the queues on the server side. Once > a message is put onto a client's outgoing queue, it will either be > delivered to the client, or it will be overwritten by a subsequent > message with the same signature (replacement rule depending). > > In PULL mode with replacement, if you poll at a rate less than the > rate at which the underlying driver publishes the data, then you > should get a new data message (followed by a SYNCH) every cycle. Of > course, if you read faster, then sometimes there won't be new data for > you, and so you'll just get the SYNCH. If you're using Stage, then > the data publication rate should be about 10Hz. > > brian. > This however means, that using Stage (this is my case), with a poll rate every 5 secs and with the replacement rule set, I should get data message for every read. Is there any way to turn some kind of debug logging on onto the console, so I could verify if the replacement rule was set correctly and mainly if a new data message is in the bumper's queue (with time stamp diplayed)? I could modify the driver and recompile it then, but I don't know where to start or what to search for. Or is there a global debugging switch in the configuration files? Elod >>> How do you know that you're not getting data for the bumpers? Have >>> you tried looking at the playerc_bumper_t structure? It gets filled >>> in as a side-effect of reading (this is actually libplayerc's default >>> callback in action). You can test for fresh data in any libplayerc >>> device proxy by examining the 'info.fresh' field. >>> >>> brian. >>> >>> On Feb 7, 2008, at 4:11 AM, Elod Kironsky wrote: >>> >>>> I tried to change the order to (1) set data mode to PULL, (2) set >>>> replace rule, (3) subscribe to bumpers. It did >>>> not work. However I figured out, that if I pull the packages in a >>>> shorter interval, some data occasionally is sent >>>> from the server. The shorter the interval the more data arrives. >>>> Originally I was pulling data every 5 secs, just to >>>> test if everything is fine, but I only got SYNCH messages. If I set >>>> the >>>> pull interval to 0.5 sec, I started to receive >>>> data packages, but *not* for every data request. Is there a timeout >>>> for >>>> data packages waiting on the queue for an >>>> interface? If yes, how can I change it? >>>> >>>> Elod >>>>> Changing the order, as Kevin suggests, is definitely worth a try. I >>>>> suspect that we're not properly handling the case when data is >>>>> already >>>>> on the socket when the data mode and/or replacement rules are >>>>> changed by >>>>> the client. So, set everything up, *then* subscribe. >>>>> >>>>> Here's a relevant post from a while back: >>>>> >>>>> https://sourceforge.net/mailarchive/message.php?msg_name=14825455.post%40talk.nabble.com >>>>> >>>>> >>>>> >>>>> brian. >>>>> >>>>> Kevin Barry wrote: >>>>> >>>>>>> order of actions: subscribe to the bumber, set data mode, set >>>>>>> replace mode? >>>>>>> >>>>>> I would try changing the order. I'm not sure it matters, but when >>>>>> I've >>>>>> done it, I always did set data mode first, then subscribe. Below is >>>>>> some code that I've used (It's in MATLAB not C so don't copy/paste, >>>>>> but the syntax is close so you'll get the idea) >>>>>> >>>>>> >>>>>> client = player('client_create', config.player_host, >>>>>> config.player_port); >>>>>> player('client_connect', client); >>>>>> >>>>>> player('client_datamode', client, PLAYER_DATAMODE_PULL); >>>>>> player('client_set_replace_rule', client, -1, -1, >>>>>> PLAYER_MSGTYPE_DATA, -1, 1); >>>>>> >>>>>> camera = player('camera_create', client, config.player_camera_num); >>>>>> player('camera_subscribe', camera, PLAYER_OPEN_MODE); >>>>>> >>>>>> >>>>>> >>>>>> On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> wrote: >>>>>> >>>>>>> Yes, I get the bumber data without any problems. Also in PULL >>>>>>> mode, I >>>>>>> get the data when requesting for it, but >>>>>>> as soon as I set the replace rule I get only SYNCH messages. Is >>>>>>> there >>>>>>> any setting in cfg file that I could use to >>>>>>> have the bumber print debug messages? Is the order of commands >>>>>>> sending >>>>>>> to the server significant? I mean the >>>>>>> order of actions: subscribe to the bumber, set data mode, set >>>>>>> replace mode? >>>>>>> >>>>>>> Elod >>>>>>> >>>>>>> >>>>>>>> Did you get bumper data in PUSH mode? >>>>>>>> >>>>>>>> Toby >>>>>>>> >>>>>>>> Elod Kironsky wrote: >>>>>>>> >>>>>>>> >>>>>>>>> Any ideo on this problem please? >>>>>>>>> >>>>>>>>> Elod >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>>> Hi! >>>>>>>>>> >>>>>>>>>> I'm trying to use PULL data mode in Player 2.0.5 using Stage >>>>>>>>>> 2.0.4. I >>>>>>>>>> send the PULL mode request message to the server. Then I set >>>>>>>>>> the replace >>>>>>>>>> rule for all data messages to have them replaced on the message >>>>>>>>>> queue by >>>>>>>>>> newer messages as described at: >>>>>>>>>> >>>>>>>>>> http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Everything seems to work fine, I don't get any "*push onto full >>>>>>>>>> message >>>>>>>>>> queue error*" text. The only problem is, that when I send a >>>>>>>>>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No >>>>>>>>>> data >>>>>>>>>> message arrives to me, although I'm subscribed to the bumber >>>>>>>>>> device on >>>>>>>>>> my robot (simulated in stage). Any clues what I'm doing wrong >>>>>>>>>> or how I >>>>>>>>>> could debug this (perhaps enable console debug logging in >>>>>>>>>> player/stage >>>>>>>>>> to see what is actually pushed onto the message queue)? >>>>>>>>>> >>>>>>>>>> Thanks, >>>>>>>>>> >>>>>>>>>> Elod >>>>>>>>>> >>>>>>>>>> ------------------------------------------------------------------------- >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>>>>> _______________________________________________ >>>>>>>>>> Playerstage-users mailing list >>>>>>>>>> Pla...@li... >>>>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>> ------------------------------------------------------------------------- >>>>>>>>> >>>>>>>>> >>>>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>>>> _______________________________________________ >>>>>>>>> Playerstage-users mailing list >>>>>>>>> Pla...@li... >>>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>> ------------------------------------------------------------------------- >>>>>>>> >>>>>>>> >>>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>>> _______________________________________________ >>>>>>>> Playerstage-users mailing list >>>>>>>> Pla...@li... >>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>>> >>>>>>>> >>>>>>> -- >>>>>>> Elod Kironsky >>>>>>> Developer >>>>>>> AVG Technologies >>>>>>> Email: elo...@av... >>>>>>> www.avg.com >>>>>>> www.avg.cz >>>>>>> >>>>>>> >>>>>>> >>>>>>> ------------------------------------------------------------------------- >>>>>>> >>>>>>> >>>>>>> This SF.net email is sponsored by: Microsoft >>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>>> _______________________________________________ >>>>>>> Playerstage-users mailing list >>>>>>> Pla...@li... >>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>>> >>>>>>> >>>>>> ------------------------------------------------------------------------- >>>>>> >>>>>> >>>>>> This SF.net email is sponsored by: Microsoft >>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>>> _______________________________________________ >>>>>> Playerstage-users mailing list >>>>>> Pla...@li... >>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------------- >>>>> >>>>> >>>>> This SF.net email is sponsored by: Microsoft >>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>>> _______________________________________________ >>>>> Playerstage-users mailing list >>>>> Pla...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>>>> >>>> >>>> >>>> -- >>>> Elod Kironsky >>>> Developer >>>> AVG Technologies >>>> Email: elo...@av... >>>> www.avg.com >>>> www.avg.cz >>>> >>>> >>>> ------------------------------------------------------------------------- >>>> >>>> >>>> This SF.net email is sponsored by: Microsoft >>>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>>> _______________________________________________ >>>> Playerstage-users mailing list >>>> Pla...@li... >>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users >>> >> >> >> -- >> Elod Kironsky >> Developer >> AVG Technologies >> Email: elo...@av... >> www.avg.com >> www.avg.cz >> >> >> ------------------------------------------------------------------------- >> >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >> _______________________________________________ >> Playerstage-users mailing list >> Pla...@li... >> https://lists.sourceforge.net/lists/listinfo/playerstage-users > -- Elod Kironsky Developer AVG Technologies Email: elo...@av... www.avg.com www.avg.cz |
From: Toby C. <tco...@pl...> - 2008-03-04 18:30:36
|
There is not a global message debug that I know of, but if you use the passthrough driver to access the bumper interface and then run player with -d 2 I think you get a message to the console with every message... alternatively you could subscribe with another client, such as player print and see the frequency it is getting data. Toby On 04/03/2008, Elod Kironsky <elo...@av...> wrote: > > Brian Gerkey wrote: > > > > On Mar 3, 2008, at 2:48 PM, Elod Kironsky wrote: > > > >> Well, I'm printing every message from the server to the console (read > >> header, read length number of bytes and print it). If I use shorter > >> intervals, I get some bumper data messages but mostly only SYNCH > >> messages. So that is why I suspect, that the data messages are simply > >> timed out somewhere on the queue. I have my client written in > Smalltalk, > >> so I cannot examine the playerc_bumper_t structure, I use my own > >> decoding methods that work with a bunch of bytes :-) > > > > Ah. Well, there's no timeout in the queues on the server side. Once > > a message is put onto a client's outgoing queue, it will either be > > delivered to the client, or it will be overwritten by a subsequent > > message with the same signature (replacement rule depending). > > > > In PULL mode with replacement, if you poll at a rate less than the > > rate at which the underlying driver publishes the data, then you > > should get a new data message (followed by a SYNCH) every cycle. Of > > course, if you read faster, then sometimes there won't be new data for > > you, and so you'll just get the SYNCH. If you're using Stage, then > > the data publication rate should be about 10Hz. > > > > brian. > > > > This however means, that using Stage (this is my case), with a poll rate > every 5 secs and with the replacement rule set, I should get data > message for every read. Is there any way to turn some kind of debug > logging on onto the console, so I could verify if the replacement rule > was set correctly and mainly if a new data message is in the bumper's > queue (with time stamp diplayed)? I could modify the driver and > recompile it then, but I don't know where to start or what to search > for. Or is there a global debugging switch in the configuration files? > > > Elod > > >>> How do you know that you're not getting data for the bumpers? Have > >>> you tried looking at the playerc_bumper_t structure? It gets filled > >>> in as a side-effect of reading (this is actually libplayerc's default > >>> callback in action). You can test for fresh data in any libplayerc > >>> device proxy by examining the 'info.fresh' field. > >>> > >>> brian. > >>> > >>> On Feb 7, 2008, at 4:11 AM, Elod Kironsky wrote: > >>> > >>>> I tried to change the order to (1) set data mode to PULL, (2) set > >>>> replace rule, (3) subscribe to bumpers. It did > >>>> not work. However I figured out, that if I pull the packages in a > >>>> shorter interval, some data occasionally is sent > >>>> from the server. The shorter the interval the more data arrives. > >>>> Originally I was pulling data every 5 secs, just to > >>>> test if everything is fine, but I only got SYNCH messages. If I set > >>>> the > >>>> pull interval to 0.5 sec, I started to receive > >>>> data packages, but *not* for every data request. Is there a timeout > >>>> for > >>>> data packages waiting on the queue for an > >>>> interface? If yes, how can I change it? > >>>> > >>>> Elod > >>>>> Changing the order, as Kevin suggests, is definitely worth a try. I > >>>>> suspect that we're not properly handling the case when data is > >>>>> already > >>>>> on the socket when the data mode and/or replacement rules are > >>>>> changed by > >>>>> the client. So, set everything up, *then* subscribe. > >>>>> > >>>>> Here's a relevant post from a while back: > >>>>> > >>>>> > https://sourceforge.net/mailarchive/message.php?msg_name=14825455.post%40talk.nabble.com > >>>>> > >>>>> > >>>>> > >>>>> brian. > >>>>> > >>>>> Kevin Barry wrote: > >>>>> > >>>>>>> order of actions: subscribe to the bumber, set data mode, set > >>>>>>> replace mode? > >>>>>>> > >>>>>> I would try changing the order. I'm not sure it matters, but when > >>>>>> I've > >>>>>> done it, I always did set data mode first, then subscribe. Below is > >>>>>> some code that I've used (It's in MATLAB not C so don't copy/paste, > >>>>>> but the syntax is close so you'll get the idea) > >>>>>> > >>>>>> > >>>>>> client = player('client_create', config.player_host, > >>>>>> config.player_port); > >>>>>> player('client_connect', client); > >>>>>> > >>>>>> player('client_datamode', client, PLAYER_DATAMODE_PULL); > >>>>>> player('client_set_replace_rule', client, -1, -1, > >>>>>> PLAYER_MSGTYPE_DATA, -1, 1); > >>>>>> > >>>>>> camera = player('camera_create', client, config.player_camera_num); > >>>>>> player('camera_subscribe', camera, PLAYER_OPEN_MODE); > >>>>>> > >>>>>> > >>>>>> > >>>>>> On Feb 6, 2008 7:29 AM, Elod Kironsky <elo...@av...> > wrote: > >>>>>> > >>>>>>> Yes, I get the bumber data without any problems. Also in PULL > >>>>>>> mode, I > >>>>>>> get the data when requesting for it, but > >>>>>>> as soon as I set the replace rule I get only SYNCH messages. Is > >>>>>>> there > >>>>>>> any setting in cfg file that I could use to > >>>>>>> have the bumber print debug messages? Is the order of commands > >>>>>>> sending > >>>>>>> to the server significant? I mean the > >>>>>>> order of actions: subscribe to the bumber, set data mode, set > >>>>>>> replace mode? > >>>>>>> > >>>>>>> Elod > >>>>>>> > >>>>>>> > >>>>>>>> Did you get bumper data in PUSH mode? > >>>>>>>> > >>>>>>>> Toby > >>>>>>>> > >>>>>>>> Elod Kironsky wrote: > >>>>>>>> > >>>>>>>> > >>>>>>>>> Any ideo on this problem please? > >>>>>>>>> > >>>>>>>>> Elod > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>>> Hi! > >>>>>>>>>> > >>>>>>>>>> I'm trying to use PULL data mode in Player 2.0.5 using Stage > >>>>>>>>>> 2.0.4. I > >>>>>>>>>> send the PULL mode request message to the server. Then I set > >>>>>>>>>> the replace > >>>>>>>>>> rule for all data messages to have them replaced on the message > >>>>>>>>>> queue by > >>>>>>>>>> newer messages as described at: > >>>>>>>>>> > >>>>>>>>>> > http://playerstage.sourceforge.net/doc/Player-2.0.0/player/group__libplayerc__datamodes.html > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> Everything seems to work fine, I don't get any "*push onto full > >>>>>>>>>> message > >>>>>>>>>> queue error*" text. The only problem is, that when I send a > >>>>>>>>>> PLAYER_PLAYER_REQ_DATA request, I only get a SYNCH message. No > >>>>>>>>>> data > >>>>>>>>>> message arrives to me, although I'm subscribed to the bumber > >>>>>>>>>> device on > >>>>>>>>>> my robot (simulated in stage). Any clues what I'm doing wrong > >>>>>>>>>> or how I > >>>>>>>>>> could debug this (perhaps enable console debug logging in > >>>>>>>>>> player/stage > >>>>>>>>>> to see what is actually pushed onto the message queue)? > >>>>>>>>>> > >>>>>>>>>> Thanks, > >>>>>>>>>> > >>>>>>>>>> Elod > >>>>>>>>>> > >>>>>>>>>> > ------------------------------------------------------------------------- > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> This SF.net email is sponsored by: Microsoft > >>>>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>>>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>>>>>>>>> _______________________________________________ > >>>>>>>>>> Playerstage-users mailing list > >>>>>>>>>> Pla...@li... > >>>>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>> > ------------------------------------------------------------------------- > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> This SF.net email is sponsored by: Microsoft > >>>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>>>>>>>> _______________________________________________ > >>>>>>>>> Playerstage-users mailing list > >>>>>>>>> Pla...@li... > >>>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>>> > >>>>>>>> > ------------------------------------------------------------------------- > >>>>>>>> > >>>>>>>> > >>>>>>>> This SF.net email is sponsored by: Microsoft > >>>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>>>>>>> _______________________________________________ > >>>>>>>> Playerstage-users mailing list > >>>>>>>> Pla...@li... > >>>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>>>>>>> > >>>>>>>> > >>>>>>> -- > >>>>>>> Elod Kironsky > >>>>>>> Developer > >>>>>>> AVG Technologies > >>>>>>> Email: elo...@av... > >>>>>>> www.avg.com > >>>>>>> www.avg.cz > >>>>>>> > >>>>>>> > >>>>>>> > >>>>>>> > ------------------------------------------------------------------------- > >>>>>>> > >>>>>>> > >>>>>>> This SF.net email is sponsored by: Microsoft > >>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>>>>>> _______________________________________________ > >>>>>>> Playerstage-users mailing list > >>>>>>> Pla...@li... > >>>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>>>>>> > >>>>>>> > >>>>>> > ------------------------------------------------------------------------- > >>>>>> > >>>>>> > >>>>>> This SF.net email is sponsored by: Microsoft > >>>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>>>>> _______________________________________________ > >>>>>> Playerstage-users mailing list > >>>>>> Pla...@li... > >>>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>>>>> > >>>>> > >>>>> > >>>>> > ------------------------------------------------------------------------- > >>>>> > >>>>> > >>>>> This SF.net email is sponsored by: Microsoft > >>>>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>>>> _______________________________________________ > >>>>> Playerstage-users mailing list > >>>>> Pla...@li... > >>>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>>>> > >>>> > >>>> > >>>> -- > >>>> Elod Kironsky > >>>> Developer > >>>> AVG Technologies > >>>> Email: elo...@av... > >>>> www.avg.com > >>>> www.avg.cz > >>>> > >>>> > >>>> > ------------------------------------------------------------------------- > >>>> > >>>> > >>>> This SF.net email is sponsored by: Microsoft > >>>> Defy all challenges. Microsoft(R) Visual Studio 2008. > >>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >>>> _______________________________________________ > >>>> Playerstage-users mailing list > >>>> Pla...@li... > >>>> https://lists.sourceforge.net/lists/listinfo/playerstage-users > >>> > >> > >> > >> -- > >> Elod Kironsky > >> Developer > >> AVG Technologies > >> Email: elo...@av... > >> www.avg.com > >> www.avg.cz > >> > >> > >> > ------------------------------------------------------------------------- > >> > >> This SF.net email is sponsored by: Microsoft > >> Defy all challenges. Microsoft(R) Visual Studio 2008. > >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > >> _______________________________________________ > >> Playerstage-users mailing list > >> Pla...@li... > >> https://lists.sourceforge.net/lists/listinfo/playerstage-users > > > > > -- > Elod Kironsky > Developer > AVG Technologies > Email: elo...@av... > www.avg.com > www.avg.cz > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Playerstage-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/playerstage-users > -- This email is intended for the addressee only and may contain privileged and/or confidential information |