From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5) Gecko/20041111 Firefox/1.0 Description of problem: During the fixating phase, cdrecord eats some system resource that blocks almost all other activity. During this period, no processes can be spawned, but those that were running, eg gkrellm, vi, X, continue. A running vi session will do some moving around, but then will lock up until the fixating is done. A new command, eg, ls, will simply block and wait until fixating is over. Gkrellm shows no appreciable use of cpu or disk, but a graph of some unknown parameter in 'proc' rises to a high level. Top stops completely. If a DVD is written with only a small amount of data say, 495 MB out of 4.7 GB, the fixating period lasts a long while - 3:20 m:s and all this time, nothing else can be done on the system. When fixating is done, everything unblocks and goes back to normal behaviour. Version-Release number of selected component (if applicable): cdrecord-2.01.1-5 How reproducible: Always Steps to Reproduce: 1. Record some data (a small part of a DVD): awk -F '\t' '{ print $2 "=" $2 }' $1 | \ /usr/bin/mkisofs -quiet \ -R -D -allow-leading-dots -graft-points -allow-multidot \ -max-iso9660-filenames -relaxed-filenames \ -path-list - | ( /usr/bin/cdrecord -dao -v driveropts=burnfree fs=16m -eject \ dev=$CDDEV tsize=${SIZE}s - 2> /dev/null ) | tee -a $LOG 2. After the data is written, while cdrecord says it is 'fixating', try to do some other work. Run ls, start vi, run ps, etc. 3. Observe that nothing works. No new processes will start. Observe that gkrellm shows substantially no use of cpu or disk, but that 'Proc' shows a graph that rises to an abnormally high value. Actual Results: See 3. above. When fixating is done - more than 3 minutes later - everything unfreezes. All pending keystrokes are processed, as expected. Expected Results: cdrecord should not block other system activity. Additional info: This is at most, inconvenient. There is no lasting harm done. Still, it is not normal and should be fixed. My DVD writer is a PIONEER DVD-RW DVR-105. I don't recall seeing this behaviour before FC3.
What is the output of rpm -q kernel ?
yes, does another kernel change the system behaviour?
I have retained several FC3 kernels: $ rpm -q kernel kernel-2.6.9-1.724_FC3 kernel-2.6.9-1.681_FC3 kernel-2.6.9-1.667 kernel-2.6.10-1.741_FC3 and am using the latest one: $ uname -a Linux datium.datix.lan 2.6.10-1.741_FC3 #1 Thu Jan 13 16:38:22 EST 2005 i686 athlon i386 GNU/Linux I will try earlier kernels and report back.
I have now rebooted to each of the four FC3 kernels listed above and run cdrecord, writing a small collection of data files to a DVD-RW disk. ALL behave the same way. The system freezes during fixating. I suspected two 'foreign' alterations of the kernel - the nvidia driver, and the pptp package. However, not using the nvidia driver and removing pptp did not improve the bad behaviour. I believe the system locks up even during the fixating phase of erasing a DVD-RW disk, but that's pretty brief.
FWIW, the same system lockup occurs when growisofs finishes writing to a DVD-RW. It labels this phase "/dev/dvd: flushing cache", but the freezeup is identical.
Fedora Core 3 is now maintained by the Fedora Legacy project for security updates only. If this problem is a security issue, please reopen and reassign to the Fedora Legacy product. If it is not a security issue and hasn't been resolved in the current FC5 updates or in the FC6 test release, reopen and change the version to match. Thank you!
*** This bug has been marked as a duplicate of 208542 ***
I'ms sorry to say this bug is still with us. I reported it 2005-01-30 for FC3, but now I'm running FC5 with all updates, and it's still the same. I guess this bugzilla should remain open. $ rpm -q cdrecord mkisofs kernel cdrecord-2.01.01.0.a03-3 mkisofs-2.01.01.0.a03-3 kernel-2.6.16-1.2133_FC5 kernel-2.6.17-1.2139_FC5 kernel-2.6.17-1.2145_FC5 kernel-2.6.17-1.2157_FC5 kernel-2.6.17-1.2174_FC5 kernel-2.6.17-1.2187_FC5 <- using this one