[vdr] VDR not starting if a plugin fails to start?
Hi, I have been running VDR as a daemon started via init script for some years now. I am currently using version 1.6.0-2. The problem is, that every now and then VDR doesn't start when I boot the server. The culprit seems to be lcdproc plugins, which for some reason cannot always connect to the LCDd daemon. In VDR's log it shows up like this: OK: Sep 18 16:31:49 vdrkone2 vdr: [4890] starting plugin: lcdproc Sep 18 16:31:49 vdrkone2 vdr: connection to LCDd at localhost:13666 established. Sep 18 16:31:49 vdrkone2 vdr: LCD output thread started (pid=4890), display size: 2x16 NOT OK: Sep 18 19:05:45 vdrkone2 vdr: [4876] starting plugin: lcdproc Sep 18 19:05:45 vdrkone2 vdr: [4876] stopping plugin: femon Sep 18 19:05:45 vdrkone2 vdr: [4876] stopping plugin: burn Now, the question is, does this really cause VDR to shut down? And if so, should VDR behave in some other way in such a case? Or is it just lcdproc plugin that causes such a behaviour? Is it possible for VDR to retry starting plugin which failed to start? -Petri ___ vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Re: [vdr] VDR not starting if a plugin fails to start?
Petri Helin ha scritto: Hi, I have been running VDR as a daemon started via init script for some years now. I am currently using version 1.6.0-2. The problem is, that every now and then VDR doesn't start when I boot the server. The culprit seems to be lcdproc plugins, which for some reason cannot always connect to the LCDd daemon. In VDR's log it shows up like this: OK: Sep 18 16:31:49 vdrkone2 vdr: [4890] starting plugin: lcdproc Sep 18 16:31:49 vdrkone2 vdr: connection to LCDd at localhost:13666 established. Sep 18 16:31:49 vdrkone2 vdr: LCD output thread started (pid=4890), display size: 2x16 NOT OK: Sep 18 19:05:45 vdrkone2 vdr: [4876] starting plugin: lcdproc Sep 18 19:05:45 vdrkone2 vdr: [4876] stopping plugin: femon Sep 18 19:05:45 vdrkone2 vdr: [4876] stopping plugin: burn Now, the question is, does this really cause VDR to shut down? And if so, should VDR behave in some other way in such a case? Or is it just lcdproc plugin that causes such a behaviour? Is it possible for VDR to retry starting plugin which failed to start? -Petri ___ vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr Hi, what i now there is a patch inside Extension Patch that allow VDR to start even if a plugin fails. I don't remember which it's, just check inside the package. Diego Pierotto -- Member of the Italian VDR Wiki http://vdr.spaghettilinux.org/ ___ vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Re: [vdr] VDR not starting if a plugin fails to start?
Diego Pierotto wrote: Petri Helin ha scritto: Hi, I have been running VDR as a daemon started via init script for some years now. I am currently using version 1.6.0-2. The problem is, that every now and then VDR doesn't start when I boot the server. The culprit seems to be lcdproc plugins, which for some reason cannot always connect to the LCDd daemon. In VDR's log it shows up like this: OK: Sep 18 16:31:49 vdrkone2 vdr: [4890] starting plugin: lcdproc Sep 18 16:31:49 vdrkone2 vdr: connection to LCDd at localhost:13666 established. Sep 18 16:31:49 vdrkone2 vdr: LCD output thread started (pid=4890), display size: 2x16 NOT OK: Sep 18 19:05:45 vdrkone2 vdr: [4876] starting plugin: lcdproc Sep 18 19:05:45 vdrkone2 vdr: [4876] stopping plugin: femon Sep 18 19:05:45 vdrkone2 vdr: [4876] stopping plugin: burn Now, the question is, does this really cause VDR to shut down? And if so, should VDR behave in some other way in such a case? Or is it just lcdproc plugin that causes such a behaviour? Is it possible for VDR to retry starting plugin which failed to start? -Petri ___ vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr Hi, what i now there is a patch inside Extension Patch that allow VDR to start even if a plugin fails. I don't remember which it's, just check inside the package. Diego Pierotto I may have been a bit vague and unclear in what I wrote earlier. I did check the source code and noticed that if a plugin fails to start it really causes VDR to not start also. My aim really was to start some discussion on whether that should be changed. I myself think that VDR should start although some plugin fails to start. I'd hate to find out that some timed recording failed because the lcd display of the PC case malfunctioned. -Petri ___ vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
[vdr] reelbox plugin - help compiling
I have now been working far too long on installing the eHD card. Without this: http://www.vdr-wiki.de/wiki/index.php/OpenSUSE_VDR_DVB-S2_-_Teil2.1:_eHD I would have given up long ago ... My system: Fedora 9 latest updates and kernel 2.6.26.3-29.fc9.i686 vdr 1.6.0 with the h.264 vdr-1.5.18-h264-syncearly-framespersec-audioindexer-fielddetection-speedup.diff Reelbox svn from today Everything up to now installed exactly like in the http link above. Now my last hurdle - I hope - is to get the reelbox plugin compiled. It now stops with this error: In file included from HdTrueColorOsd.c:34: fontsml-iso8859-15.c:1: error: 'tPixelData' in class 'cFont' does not name a type HdTrueColorOsd.c: In member function 'int Reel::HdTrueColorOsd::CacheFont(const cFont)': HdTrueColorOsd.c:147: error: 'FontSml_iso8859_15' was not declared in this scope HdTrueColorOsd.c:182: error: 'NUMCHARS' is not a member of 'cFont' HdTrueColorOsd.c:208: error: 'tCharData' is not a member of 'cFont' HdTrueColorOsd.c:208: error: 'charData' was not declared in this scope HdTrueColorOsd.c:209: error: 'NUMCHARS' is not a member of 'cFont' What is wrong here? Josce _ Explore the seven wonders of the world http://search.msn.com/results.aspx?q=7+wonders+worldmkt=en-USform=QBRE___ vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Re: [vdr] VDR not starting if a plugin fails to start?
Petri Helin wrote: really causes VDR to not start also. My aim really was to start some discussion on whether that should be changed. I myself think that VDR should start although some plugin fails to start. I'd hate to find out that some timed recording failed because the lcd display of the PC case malfunctioned. Why not design plugins that start even if some non-fatal error happens? To me it sounds natural that if some non-recoverable error happens, the whole program stops. yours, Jouni ___ vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
