Arch Linux

Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines

Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!
Tasklist

FS#1705 - K3B 0.11.17-1 not working for regular user with kernel 2.6.9

Attached to Project: Arch Linux
Opened by Win Carus (Win) - Saturday, 30 October 2004, 19:51 GMT
Last edited by dorphell (dorphell) - Thursday, 06 January 2005, 06:52 GMT
Task Type Bug Report
Category System
Status Closed
Assigned To Damir Perisa (damir.perisa)
Architecture not specified
Severity Very Low
Priority Normal
Reported Version 0.7 Wombat
Due in Version Undecided
Due Date Undecided
Percent Complete 0%
Votes 0
Private No

Details

After an upgrade from kernel 2.6.8 to 2.6.9, K3B 0.11.17-1 now does not work properly for a regular user (but it does for root). Configuration and all permissions and ownerships are unchanged from before.

This may be a repeat of the same issue seen with the first release of kernel 2.6.8.
This task depends upon

Closed by  Damir Perisa (damir.perisa)
Thursday, 06 January 2005, 11:19 GMT
Reason for closing:  Works for me
Additional comments about closing:  actual kernel is 2.6.10
actual k3b is 0.11.18
Comment by Tobias Powalowski (tpowa) - Saturday, 30 October 2004, 23:20 GMT
what permissions does cdrecord have?
you use custom kernel or stock arch kernel?
what says the error exactly? (command line output)
Comment by Win Carus (Win) - Sunday, 31 October 2004, 17:13 GMT
Permissions/Ownership for cdrecord:

-rws--x--- 1 root cdrom 324188 Sep 14 16:13 /usr/bin/cdrecord

Kernel: stock 2.6.9 SCSI kernel

Command Line Error:

k3b: ***** cdrecord parameters:
k3b: /usr/bin/cdrecord -v gracetime=2 dev=/dev/hdd speed=52 -tao -eject blank=fast -force k3b:
k3b: (K3bCdrecordWriter) error: 12
Comment by Alfonso Pola (alfonso_pola) - Monday, 01 November 2004, 15:23 GMT
I think the problem, is that the cdburning patch that's used in the 2.6.9 package, points to kernel 2.6.8.1 files.

diff -Naur linux-2.6.8.1-orig/drivers/block/scsi_ioctl.c linux-2.6.8.1/drivers/block/scsi_ioctl.c
--- linux-2.6.8.1-orig/drivers/block/scsi_ioctl.c 2004-08-14 03:56:23.000000000 -0700
+++ linux-2.6.8.1/drivers/block/scsi_ioctl.c 2004-08-21 09:38:53.000000000 -0700
@@ -193,8 +193,8 @@
return -EINVAL;
if (copy_from_user(cmd, hdr->cmdp, hdr->cmd_len))
return -EFAULT;
- if (verify_command(file, cmd))
- return -EPERM;
+ /*if (verify_command(file, cmd))
+ return -EPERM;*/

/*
* we'll do that later
Comment by Tobias Powalowski (tpowa) - Monday, 01 November 2004, 15:25 GMT
nope it is patched with -Np1 that means the first thing will be ignored
Comment by Glyn Adgie (glyn) - Thursday, 25 November 2004, 14:12 GMT
Cdrecord does not run properly as a normal user with suid permissions: -rws--x---

You can get rid of the error by setting non-suid permissions:
-rwxr-x--- 1 root optical

This is not perfect, as cdrecord cannot set a high priority. I get the
following messages:

cdrecord: Operation not permitted. WARNING: Cannot set RR-scheduler
cdrecord: Permission denied. WARNING: Cannot set priority using setpriority().
cdrecord: WARNING: This causes a high risk for buffer underruns.

However, there were no actual buffer problems burning a CD on a fairly fast machine.

Note that K3b setup will try to set suid permissions on cdrecord. You can prevent that by unticking the option in the setup window.

There is some interesting discussion by Linus Torvalds and others on the Linux Kernel Mailing List:

http://linux.derkeiler.com/Mailing-Lists/Kernel/2004-08/9154.html

The upshot is that Linus thinks that the problems may have shown up due to changes in the kernel, but it is the user space code that is actually at fault. Some patches to K3b are suggested. I will have a go if I have time.
Comment by Damir Perisa (damir.perisa) - Thursday, 06 January 2005, 09:00 GMT
0.11.18-1 in extra ... is this still an issue?
Comment by Tobias Powalowski (tpowa) - Thursday, 06 January 2005, 09:38 GMT
for me it is not an issue anymore

Loading...