Felix Miata composed on 2015-02-02 17:23 (UTC-0500):
I just spent most of the past several days getting this Intel 2 core E6700 P965/ICH8 RAID1 system switched from 11.4 to 13.1 and 13.2 in multiboot via Grub Legacy on primary partition. To do so, I used a single HD Intel 2 core E7600 mule/testing system with 2 new disks added, using running 13.1's yast to build all RAID devices on the new pair prior to initializing 13.1 installation. I eventually got both 13.1 and 13.2 booting and running satisfactorily, restored an rsync backup of user data to the RAID, then shut down this system and moved the RAID pair to it.
13.1 worked satisfactorily immediately, including NIC config retaining fixed IP on eth0.
13.2 spewed trap/segfault errors in a neverending stream virtually instantly in trying to boot from some cmdline config attempts. Many of the others ended in trap data, blinking keyboard LEDs and frozen screen claiming reboot would occur in 90 seconds.
The failsafe stanza provided boot, but during init many errors were spewed on tty1 among normal messages. Before I even thought to try failsafe, I chrooted from 13.1 to zypper up and update kernel from 3.16.6 to 3.16.7, then when that didn't help, TW's 3.18.3, also no help. I booted failsafe taking one more option off cmdline each time about 8 times until the only rescue options left were noresume and apm=off, so apm=off definitely functions to get a
I was exhausted when I wrote OP. nosmp is what's required, not apm=off.
working boot, yet not without seeing the error stream on tty1. The last attempt without apm=off produced the frozen screen and keyboard condition.
I looked in the mailing list archives and Bugzilla, but didn't find anything I recognized as similar. Is there something journalctl can capture that can help someone help me figure out if this is a 13.2 bug that may have escaped discovery before now?
I followed up with a TW installation on a third md device. Its 3.18.3, and 3.18.6 from BS Stable, essentially misbehave exactly like 13.2's kernels, requiring nosmp to boot at all, and lots of error/segfault/panic messages before boot finishes. I see no such messages booting 13.1.
FWIW, installing new kernels via chroot produces 13.2 initrds 50% bigger than those happening on account of zypper up processing. Here out of habit, zypper up/dup excludes adding a new kernel. I zypper al kernel* first, up, unlock, then repeat, so that the new kernel's initrd is guaranteed to be built once only.
Also FWIW, 13.1 (like 11.4 did) requires clock=hpet on cmdline to prevent clock apparently overpowering ntp. While asleep last night it gained something like 10 minutes in 6 hours. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation)
Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org