|
From: Steve N. <sr...@co...> - 2016-08-31 13:41:58
|
Dear Schily -- Many thanks for your kind response. I also tried it with a CAM device argument, and that failed, too. Session attached herewith. Steve On 08/31/2016 07:37 AM, Joerg Schilling wrote: > Steve Newcomb <sr...@co...> wrote: > >> Dear Schily, >> >> I just bought a Yokkao USB-3 DVD drive from Amazon/China >> <https://www.amazon.com/gp/product/B01EMXQBLM>. It failed to burn a >> DVD+R with cdrecord 3.01 -- a version I have used countless times with >> other DVD drives. It also failed with 3.02. >> >> However, the hardware worked perfectly with the xfburn tool that >> evidently comes with Xubuntu 16.04, and I verified the resulting DVD; >> it really did work. >> >> Obviously, I would prefer to use cdrecord rather than xfburn. I'm >> attaching the output of the cdrecord 3.02 session FYI. > Your problem is not a problem with cderecord but either caused by your usage or > caused by another (hostile) program on Linux. > > Because Linux comes with competing drivers for CD-ROMs that partially do not > implement working or useful DMA, it is a really bad idea to force cdrecord to > use the unsupported dev=/dev/* syntax. Either fully remove the dev= parameter > (in case you have only one CD-ROM drive in the system) or use the SCSI CAM > standard conforming dev=b,t,l syntax. This allows libscg to select the best > driver which is usually /dev/sg*, so make sure that your system has /dev/sg* > available. > > The error message is either caused by non-working DMA or by a hostile program > that disregards the rules for accessing CD-ROM drives that may be written to. > > The first hostile program that was known to interrupt CD-ROM writing is "hald" > but later the same defective code has been moved to "udev" or to "systemd". > > If using the officially supported dev=b,t,l syntax does not help, you need to > kill the program that incorporates the defective and hostile code. > > As you might not use "hald", you would need to get help from other people to > learn which program you need to kill. > > http://www.fokus.fraunhofer.de/usr/schilling ftp://ftp.berlios.de/pub/schily |