10.1->10.3/11.0 failing

[xiphos]

Hello, I decided to take the time to upgrade my old 10.1 installation this weekend to 10.3. Execution of freebsd-update as specified here seemed to conclude without an issue up until the first restart. On restarting the system entered a cycle of rebooting with the following error displayed on-screen just before rebooting:

current process  = 0 (thread taskq)
trap number = 9
panic: general protection fault
cpuid = 0
KDB: stack backtrace:
#0 0xffffffff8098e3e0 at kdb_backtrace+0x60
#1 0xffffffff809510b6 at vpanic+0x126
#2 0xffffffff80950f83 at panic+0x43
#3 0xffffffff80d55f8b at trap_fatal+0x36b
#4 0xffffffff80d55c0d at trap+0x77d
#5 0xffffffff80d3b8d2 at calltrap+0x8
#6 0xffffffff80314eb5 at passregister+0x195
#7 0xffffffff802e39a4 at cam_periph_alloc+0x584
#8 0xffffffff80314cd0 at passasync+0xa0
#9 0xffffffff802ed21a at xptsetasyncfunc+0x11a
#10 0xffffffff802ed896 at xptdevicetraverse+0xe6
#11 0xffffffff802ed6ba at xpttargettraverse+0x9a
#12 0xffffffff802ed509 at xptbustraverse+0xa9
#13 0xffffffff802ed09d at xpt_register_async+0x25d
#14 0xffffffff80314c09 at passinit+0x19
#15 0xffffffff802e33fe at periphdriver_init+0x4e
#16 0xffffffff802ecdc6 at xpt_finishconfig_task+0x16
#17 0xffffffff8099f215 at taskqueue_run_locked+0xe5
Uptime: 5s

There's more above this block, but it scrolls by too quickly to capture.
Thinking it was a setting in /boot/loader.conf that didn't like the upgrade I booted from kernel.old and commented out everything, but still no luck.

A quick Google search found a similar, but not identical, issue from about a year ago where it turned out the ZFS pool that had transferred since 8.x had become corrupt. The boot pool is fine to destroy, so I created a USB boot image from the "FreeBSD-10.3-RELEASE-amd64-memstick.img" file and booted off the device with the intent of installing from scratch.

The USB boot image also reboots with the exact same error.

I then tried booting from "FreeBSD-11.0-RELEASE-amd64-memstick.img", and still recieved the error.

Another Google search for my mainboard and FreeBSD indicates support for my mainboard, an MSI H81I, might have been dropped in 10.2? I guess my questions are:

1. How can I determine which, if any specific, component on the mainboard is the root cause so I can disable it?

2. What's the best way to revert to 10.1 and just live in the dark ages until such time as I can afford to upgrade the hardware?

Thank you for your time.

 

[SirDice]

Another Google search for my mainboard and FreeBSD indicates support for my mainboard, an MSI H81I, might have been dropped in 10.2?

I wonder where you got this from. In general support is rarely dropped for hardware. Unless it's really old hardware nobody has or uses anymore. But I very much doubt anything on your mainboard was dropped.

I'd be more inclined to check for hardware errors. Judging by the xpt* messages I'd look for disk and/or controller errors first.

 

[Deleted member 9563]

An LGA1150 socket and DDR3 RAM is not even close to being old. In fact that particular board is probably not even as old as FreeBSD 10.0.

 

[monwarez]

First download a version of FreeBSD 10.1 put it on a memstick and try to boot, if this success you can then downgrading to 10.1.
For that 2 possibilities(the first should work, the second I am not sure):
If you can boot in single mode, you can with another computer download the svn revision of the /usr/src that match the version 10.1 of FreeBSD in a memstick and then mount it , copy files to /usr/src and then follow instruction for rebuilding world and kernel https://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/makeworld.html
If you can't , use the live version of the FreeBSD 10.1 memstick , and mount your hard drive in /mnt , and then put the /usr/src in it , and chroot in /mnt , and follow the same procedure that if you can boot in single mode