[Bug 581612] New: boot.clock blocked for 3 minutes in udevadm settle --seq-start=724 --seq-end=730
http://bugzilla.novell.com/show_bug.cgi?id=581612 http://bugzilla.novell.com/show_bug.cgi?id=581612#c0 Summary: boot.clock blocked for 3 minutes in udevadm settle --seq-start=724 --seq-end=730 Classification: openSUSE Product: openSUSE 11.2 Version: Final Platform: i586 OS/Version: openSUSE 11.2 Status: NEW Severity: Normal Priority: P5 - None Component: Basesystem AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: phdm@macqel.be QAContact: qa@suse.de Found By: --- Blocker: --- Created an attachment (id=343674) --> (http://bugzilla.novell.com/attachment.cgi?id=343674) boot.msg with added debugging (but udevadm settle debug truncated) User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.7) Gecko/20091222 SUSE/3.5.7-1.1.1 Firefox/3.5.7 every time I boot opensuse 11.2 on my older machine, it blocks for 3 minutes in boot.clock. I have investigated and discovered that udevadm settle --seq-start=724 --seq-end=730 is the culprit : it blocks for 3 minutes. I have then recompiled udev with added debugging and found thay udevadm settle os actually busy waiting with a strange behaviour. Logfile is attached. Reproducible: Always Steps to Reproduce: 1.(re)boot my old system 2. 3. Actual Results: boot process blocked for 3 minutes in boot.clock Expected Results: fast boot sequence -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c1
Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c
Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c2
Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c
yang xiaoyu
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c3
Kay Sievers
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c4
Kay Sievers
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c5
--- Comment #5 from Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c6
Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c7
--- Comment #7 from Kay Sievers
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c8
--- Comment #8 from Kay Sievers
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c9
Kay Sievers
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c
Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c10
--- Comment #10 from Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c11
--- Comment #11 from Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c12
Philippe De Muyter
Fix released upstream and a new udev package is in Factory.
How did you run into this? There is no other report like this for openSUSE 11.2. You compiled your own kernel with a modular rtc_cmos?
I did not compile my kernel myself. It comes from SuSE : diabolo:~ # cat /proc/version Linux version 2.6.31.12-0.1-default (geeko@buildhost) (gcc version 4.4.1 [gcc-4_4-branch revision 150839] (SUSE Linux) ) #1 SMP 2010-01-27 08:20:11 +0100
We usually have rtc_cmos compiled-in, so that the kernel itself can sync the system time at bootup before any userspace runs, and hwclock does not necessarily need to be called: $ dmesg rtc_cmos 00:07: setting system clock to 2010-05-25 11:11:02 UTC
But you're right : rtc_cmos is a module (see attached /proc/config.gz (unzipped)) -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c13
--- Comment #13 from Kay Sievers
I have merged your patch into my verbose version and udevadm settle now finish before the timeout as it should. I cannot tell if it really works as it should, because I notice that the loop is still entered many times. Is that normal ?
Looks all fine. In the logs it loops until event 729 is finished. This is the expected behavior. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c14
Kay Sievers
I did not compile my kernel myself. It comes from SuSE. But you're right : rtc_cmos is a module (see attached /proc/config.gz
Interesting. This bug should trigger for all users then, but I did not see any other report. I hope they don't wait all for 3 minutes. :) Weird, I don't see rtc_cmos as a module in the 11.2 kernel. The kernel you have installed is just from the standard update repo? I don't have a recent 11.2 installation anymore. In the Factory kernel, the one that will be 11.3, it is also compiled in. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c15
Philippe De Muyter
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c16
Kay Sievers
http://bugzilla.novell.com/show_bug.cgi?id=581612
http://bugzilla.novell.com/show_bug.cgi?id=581612#c17
--- Comment #17 from Philippe De Muyter
participants (1)
-
bugzilla_noreply@novell.com