Hi, me: > > I kill on my machine the processes > > hald-addon-storage chi kwan: > I kill mime and gosh=2C it is burning at 8x as dummy!
So the riddle seems to be solved for now. The following text was written before i learned about your success. (One should read all mails before beginning to write a lengthy answer.) ------------------------------------------------- chi kwan: > dvd+rw-mediainfo /dev/cdrom output: > Mounted Media: 11h, DVD-R Sequential > Disc status: blank > Next Writable Address: 0*2KB This media appears ready for being written beginning at block address 0. > also, for your information, I've been successful in blanking the dvd-rw with > a blank=fast command, (did'nt dare use =full least permanent damage), I > don't know if this means anything. Well, most SCSI commands seem to work well with drive and media. Only the write command is not willing to perform. > so it is not very likely that the system is > continuously checking the drive and causes problems. It is not so clear how hald spoils our burns. It is a known fact, though, that on many systems CD and DVD-R[W] are prone to various errors if hald is allowed to operate on the drive, MMC prescribes that the burn session may not be disturbed by certain other SCSI commands. Obviously hald sends some of them or resets the generic Linux SCSI driver. > > Doesn't "Current: DVD-ROM" mean this DVD-R has already been written to? > Sorry, may be I have the wrong output posted, It would have been quite a blooper if all your tries were made with a written media. > Current: DVD-R sequential recording This looks more like a writable media. (But same failure symptoms as already reported.) > > ... > > 5 30 10 MEDIUM NOT FORMATTED > wow, it looks like from some documents for elites, where can i find it? and Operation of optical media is covered by the MMC specs: http://www.t10.org/ftp/t10/drafts/mmc/mmc5r03c.pdf some general commands are in SPC and SBC http://www.t10.org/ftp/t10/drafts/spc3/spc3r23.pdf http://www.t10.org/ftp/t10/drafts/sbc3/sbc3r17.pdf (The SCSI family applies to all modern burners regardless of their connection cables and controllers.) A compilation of the burn knowledge behind libburn can be found in: http://libburnia-project.org/browser/libburn/trunk/doc/cookbook.txt It has a lot of pointers into MMC and SPC. Your media type is covered by Sequential DVD-R[W] Cookbook > and can i fine tune cdrecord so it won't be > too specific on medias? Before you begin that endeavor you should try growisofs or one of my programs based on libburn (cdrskin, xorriso). As you can see from the cookbook resp. from MMC chapter 4, "Models" the media types are quite variform. If it turns out that a peculiar gesture of cdrecord is indeed to blame, then you may use its option -V to learn about the SCSI commands issued to the drive. Any obvious deviation from the specs should be reported to Joerg Schilling. Nevertheless, cdrecord is not the main suspect for now. > what should i do with hald and O_EXCL ? I assume your successful burns were via growisofs. Afaik it opens the device file with option O_EXCL. See man 2 open for usage. The meaning of O_EXCL on Linux storage device files is fewly documented: A further open(O_EXCL|O_RDWR) will fail as long as there is an open file descriptor with O_EXCL|O_RDWR. This is part of a Linux-specific locking protocol for mount devices. Some versions of hald seem to participate in this protocol. Best is to keep hald from accessing the burner drive while you are burning. I need no auto mounting. So i kill hald-addon-storage and do never restart it. (USB unplug-replug starts it, though.) But that depends much on the version of hald you are running. Each system has its own tricks to curb hald. (It is said to be configurable.) Have a nice day :) Thomas -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
