January 2003 - Packet Writing - openSUSE Mailing Lists

2.5.xx / 2.6
by Some Bloke 02 Feb '03

02 Feb '03

Hello all, I'm trying to find patches for the newest 2.5.59 kernel. Any pointers? Interestingly, I saw something here http://kernelnewbies.org/status/Status-22-Jan-2003.html (look for UDF), that suggests new packet-cd patches have been merged into the main kernel code. However, when I downloaded 2.5.59 and looked for packet-cd options, I couldn't find any. What's going on? Any help appreciated. SB. _________________________________________________________________ Use MSN Messenger to send music and pics to your friends http://messenger.msn.co.uk

2 1

Bootable CD in UDF
by Jacob Lahr 27 Jan '03

27 Jan '03

Hi, is it possible to create a bootable cd-rw that is packet-writeable? My aim is a eval distribution, that has a part of the cd left free in order to save some informations like preferences and modified documents. kind regards, Jacob -- http://www.jacoblahr.de http://www.h-mpeg.de -> ganz neu

6 9

UDF-fs: writing on write-once media (DVD+R)
by Sergio Ballestrero 26 Jan '03

26 Jan '03

Dear all, maybe this mail is somewhat off-topic, but I tried to send it to the Linux-UDF mailing list, and that list seems dead quiet. So, I try to post my question here, before resorting to linux-kernel. Many thanks to anybody who will answer, or simply address me to the proper list or person. Begin forwarded message: From: Sergio Ballestrero <s.ballestrero(a)firenze.linux.it> Date: Fri Jan 24, 2003 12:26:23 Europe/Rome To: linux_udf(a)hpesjro.fc.hp.com Subject: writing on write-once media Dear all, as I unfortunately need to write log files on write-once media (for an Italian fiscal rules-compliant application), I was investigating writing UDF on DVD+/-R media. Looking on the most recent sources from SourceForge CVS, I can read "WORM support" in the TODO list, and I have not really found much code related to ACCESS_TYPE_WRITE_ONCE or to StrategyType 4096. The questions: - do I understand correctly that writing on (any kind of) WO media is NOT supported at the moment ? - could anybody estimate how much effort would be required to support it ? The company I work for might consider dedicating a bit of time and/or money for this. I've been browsing the ECMA and OSTA documentation, and I'd have quite a few more technical questions, but I'll leave them for later. Thanks and cheers, Sergio Ballestrero PS: my apologies if this question has been answered before - I have not found an archive of the list.

3 2

Re: UDFtools stopped working
by Bringfried Stecklum 21 Jan '03

21 Jan '03

quick update: Got my hands on the CD-RW which I UDF-formatted first. Surprise, surprise. I was able to mount it via /dev/pktdvd without problems. Thus the kernel driver is not to blame. Then I looked into the cdrwtool source and figured out where the programs stops with the "set_options" comment. Sorry for being ruthless but I just did /* if ((ret = cdrom_open_check(fd))) { fprintf(stderr, "set_options\n"); cdrom_close(fd); return ret; } */ and compiled. And (not quite unexpectedly) cdrwtool is working again! This is certainly not a true fix but useful for me (I will _never_ eject a CD-RW until cdrwtool is finished). regards, B.St. -- Bringfried Stecklum Th"uringer Landessternwarte Tautenburg Sternwarte 5, D - 07778 Tautenburg, Germany Phone: +49-36427-863-54, FAX: +49-36427-863-29

1 0

udf/iso9660 on dvd-r read trouble
by Andreas Plesch 21 Jan '03

21 Jan '03

I am running 2.4.18-5 with udf-0.9.6. The udf file system works great on a firewire dvd-ram and I can exchange data on dvd-ram with a Win2000 PC. Now I want to read a dvd-r which I created on the windows PC with the UDF packetwriting software instantwrite by vob. Instantwrite lets you finalize the dvd and this puts a iso9660-3 "shell" around the existing udf filesystem, as far as I understand. I do not think that "finalize" closes the disk because you are allowed to continue to write to it even after the "finalize". When I mount this dvd-r in my linux box I can do it only with -t iso9660 although this seems to be reserved for CD-RW. Accordingly, the kernel prints a messages that the cruft mount option is used (there is a >1GByte and a >2GByte file on the DVD, data not multimedia content) and files are shortened. So I think I am supposed to use mount -t udf. However, this leads to long drive activity during which some I/O-errors are printed by the kernel. After some time then mount returns silently but nothing is mounted. Jan 20 23:23:17 garvin kernel: scsi0: ERROR on channel 0, id 0, lun 0, CDB: Request Sense 00 00 00 40 00 Jan 20 23:23:17 garvin kernel: Current sd0b:00: sense key Medium Error Jan 20 23:23:17 garvin kernel: Additional sense indicates No seek complete Jan 20 23:23:17 garvin kernel: I/O error: dev 0b:00, sector 9193980 ... Jan 20 23:26:14 garvin kernel: UDF-fs: No partition found (1) [I use ide-scsi for a reason I have forgotten but was important at some point :)] I also installed cdfs but mounting the dvd with -t cdfs only gives total 0 -r--r--r-- 1 root root 0 Dec 31 1969 videocd-1.mpeg Is there a way I could read the DVD-R ? Is iso9660-3 somehow supported ? Is there a way to get rid of the cruft mount option ? Does a udf filesystem on a dvd-r have to be closed in order to be read ? Any help or partial answers appreciated, Andreas

1 0

UDFtools stopped working
by Bringfried Stecklum 21 Jan '03

21 Jan '03

Here is my story. I installed udftools-1.0.0b2 on SuSE8.0 kernel 2.4.19 using the packet-2.4.19 patch yesterday. Everything went fine and I was able to use to the UDF file system on the CD-RW in the TEAC DW-28E combo drive. Later I tried around with the blocksize option since I wanted to use a 700MB CD-RW instead of the 650MB. That was probably my fault. When I interrupted "cdrwtools -d /dev/sr0 -q -m 36000 -t 4" via Ctrl-C the door of the drive was still locked. I issued an eject and the CD-RW came out together with an ioctl error message (which I do not recall, sorry). Since then I cannot use udftools anymore although my drive works well with other programs. The symptoms are as follows: cdrwtools -d /dev/sr0 -q just yields the output >using device /dev/sr0 >set_options and that's it. It is not accessing the drive at all. It just stops without _any_ action (not invoking mkduffs and so on). mount /dev/sr0 /media/dvd -t udf works fine but I cannot mount the DVD via the packet block device /dev/pktcdvd which yields the obviously common "can't read superblock error" (kernel message is pktcdvd: Unknown disc. No track?). I tried everything from rebooting, re-installing udftools and even building kernel/modules with the more recent packet-2.4.19-2 patch but this did not solve the problem. It seems to me that there is a lock file somewhere which prevents udftools from proper working. I really do not want to re-format the system partition to get rid of it. Any help is appreciated. regards, B.St. Bringfried Stecklum stecklum_at_tls-tautenburg.de Th"uringer Landessternwarte Tautenburg Sternwarte 5, D - 07778 Tautenburg, Germany Phone: +49-36427-863-54, FAX: +49-36427-863-29

1 0

Re: Re: Re: ioctl: Inappropriate ioctl for device
by Alimin Bijosono Oei 16 Jan '03

16 Jan '03

Alimin, I get [roy@Neddyseagoon roy]$ ls -al /dev/pktcd* brw-rw-rw- 1 root root 97, 0 Dec 27 10:30 /dev/pktcdvd0 brw-rw-rw- 1 root root 97, 1 Dec 27 10:30 /dev/pktcdvd1 brw-rw-rw- 1 root root 97, 2 Dec 27 10:30 /dev/pktcdvd2 Your permissions look a bit strange. You file appears to be a normal file and not a block device file. You need to use mknod /dev/pktcdvd0 b 97 0 to make the special block device file in /dev. See man mknod for more info. Set the correct permissions after. Regards, Roy Bamford ------------------------------------------------------- Thank you very much Roy. I finally manage to get it done by deleting the normal file '/dev/pktcdvd0' with the strange permission and then applying 'mknod /dev/pktcdvd0 b 97 0'. Regards, Alimin

1 0

ioctl: Inappropriate ioctl for device
by Alimin Bijosono Oei 16 Jan '03

16 Jan '03

Hi all, I am using Redhat 8.0 on my Compaq Presario 1500AP laptop. Recently I compiled and used kernel 2.4.20. I did all the patching procedure in order to enable udf packet writing. Everything seemed ok during the patching, compiling and installation process of the kernel until I rebooted and installed the latest udftools. Pktsetup, mkudffs and cdrwtool were installed succesfully, checking /dev/pktcdvd0 using mknod was also ok but when I tried 'pktsetup /dev/pktcdvd0 /dev/cdrom' and 'pktsetup /dev/pktcdvd0 /dev/scd0' it was not a success with the following message: 'ioctl: Inappropriate ioctl for device' I have checked 'dmesg', there is nothing wrong there. In case anyone is interested, attached is the dmesg produced. Thank you very much for any help and guidance provided. Regards, Alimin

1 0

Re: Re: ioctl: Inappropriate ioctl for device
by Alimin Bijosono Oei 15 Jan '03

15 Jan '03

>Maybe your sistem uses devFS and you have /dev/pktcdvd >auto created (instead of /dev/pktcdvd0). This is in my >case. Thanks for the response. How do I know if this is the case and how do I get it fixed to be able to do packet writing? Checking the presence of '/dev/pktcdvd' has no indication that it is there but '/dev/pktcdvd0' is present. [alimin@compaq alimin]$ ll /dev/pktcdvd ls: /dev/pktcdvd: No such file or directory [alimin@compaq alimin]$ ll /dev/pktcdvd0 --w-r-S--t 1 root root 0 Jan 12 17:10 /dev/pktcdvd0 Regards, Alimin

2 1

ioctl: Inappropriate ioctl for device
by Alimin Bijosono Oei 15 Jan '03

15 Jan '03

Hi all, I am using Redhat 8.0 on my Compaq Presario 1500AP laptop. Recently I compiled and used kernel 2.4.20. I did all the patching procedure in order to enable udf packet writing. Everything seemed ok during the patching, compiling and installation process of the kernel until I rebooted and installed the latest udftools. Pktsetup, mkudffs and cdrwtool were installed succesfully, checking /dev/pktcdvd0 using mknod was also ok but when I tried 'pktsetup /dev/pktcdvd0 /dev/cdrom' and 'pktsetup /dev/pktcdvd0 /dev/scd0' it was not a success with the following message: 'ioctl: Inappropriate ioctl for device' I have checked 'dmesg', there is nothing wrong there. In case anyone is interested, attached is the dmesg produced. Thank you very much for any help and guidance provided. Regards, Alimin

2 1