I would not blame python, nor owfs, but fuse. IMHO: owfs+fuse is very handy for debugging, but it makes really no sense in a production environment. The way to go is owserver with its protocol, and the appropriate client libraries.
Shell clients can access 1-wire data with owshell (owdir, owread, owwrite) while python clients can bind with the ow or ownet python libraries. Please consider also the newer pyownet, documented at http://pyownet.readthedocs.org/en/latest/ <http://pyownet.readthedocs.org/en/latest/> and pyowfs documented at http://priesch.co.at/pyowfs <http://priesch.co.at/pyowfs> Both packages are available on pypi, so they can be installed with pip install pyowfs pip install pyownet for easy testing. Stefano. PS: I’m the author of pyownet, which I wrote when transitioning from perl to python to overcome some of the shortcomings of ownet, and for the fun of it. And for sure I think it is really bad idea to exec() an external program from inside a python script. Minimal instruction for use are >>> from pyownet import protocol >>> owproxy = protocol.proxy(host="server.example.com", port=4304) >>> owproxy.dir() [u'/10.A7F1D92A82C8/', u'/05.D8FE434D9855/', u'/26.8CE2B3471711/'] >>> owproxy.read('/10.A7F1D92A82C8/temperature') ' 6.68422' The only thing you should be aware of is that creating an owproxy object is quite time consuming, so usually you have a single owproxy instance, shared globally. > On 04 Nov 2015, at 12:32, Martin Patzak (GMX) <martin.pat...@gmx.de> wrote: > > Well, this is interesting: > > while further investigating the problem it looks like this: > > the value of sensed.BYTE toggles between 55 (00110111) and 183 (10110111) for > quite a while - so its only flipping one bit. > > When I increase the toggling speed, I can provoke my 553 error from > sensed.BYTE from once a day to several times a minute. > > It looks like reading the owfs-file sensed.BYTE in python 2.7 is not an > atomic function!?! > > Any thoughts? > > Cheers, > > Martin > > > > > > > On 10/30/2015 08:23 AM, Martin Patzak (GMX) wrote: >> Hello All, >> >> I am experiencing 'strange' readings from a DS2408. >> >> Every once in a while (that is about once a day, with reading the >> sensed.BYTE folders once a second - so its rather rare) the value read from >> one of the DS2408 is 553 where it should be between 0 and 255. >> >> Is this some 'special' value indicating an error or something the like? >> Anybody else saw this phenomena before? >> >> I throw this value away and re-read again and everything is fine! >> >> >> I am running solely owfs 2.9p8 on a BananaPi running Debian 8 'jessie' (no >> owserver or owhttp is running). >> >> There are 25 temperature sensors and 2 DS2408 on a powered bus with an >> USB-Link as bus-master. >> >> While there is some perl-code writing to the PIO.<bit> of the DS2408, it is >> Python-code reading the sensed.BYTE of the 2408 to display the state in an >> HMI on a webpage. >> >> >> Cheers >> >> Martin >> >> >> >> ------------------------------------------------------------------------------ >> >> >> _______________________________________________ >> Owfs-developers mailing list >> Owfs-developers@lists.sourceforge.net >> <mailto:Owfs-developers@lists.sourceforge.net> >> https://lists.sourceforge.net/lists/listinfo/owfs-developers >> <https://lists.sourceforge.net/lists/listinfo/owfs-developers> > > ------------------------------------------------------------------------------ > _______________________________________________ > Owfs-developers mailing list > Owfs-developers@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/owfs-developers
------------------------------------------------------------------------------
_______________________________________________ Owfs-developers mailing list Owfs-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/owfs-developers
