From: samuel on
Hi all,

Recently a couple of new machines came to my hands, HP DL 180 G6 with 500Gb
SATA hard disk, and I tried to setup a 2-nodes ganeti cluster with debian
lenny.

Default 2-6-26 kernel does work without any issue.

I've done a fresh installation with nothing more than the basics (vim and
less ;) ) and after a aptitude install ganeti, editing xen-config.xsp and
grub xen options, upon restarting I get the following error on the screen:

Apr 15 13:24:56 virt8 kernel: [ 297.381987] ata3.00: exception Emask 0x0
SAct 0x0 SErr 0x0 action 0x6 frozen
Apr 15 13:24:56 virt8 kernel: [ 297.382163] ata3.00: cmd
a0/00:00:00:24:00/00:00:00:00:00/a0 tag 0 pio 36 in
Apr 15 13:24:56 virt8 kernel: [ 297.382164] cdb 12 00 00 00 24 00
00 00 00 00 00 00 00 00 00 00
Apr 15 13:24:56 virt8 kernel: [ 297.382165] res
40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Apr 15 13:24:56 virt8 kernel: [ 297.382595] ata3.00: status: { DRDY }
Apr 15 13:24:56 virt8 kernel: [ 297.382727] ata3: hard resetting link
Apr 15 13:24:56 virt8 kernel: [ 297.858982] ata3: SATA link up 1.5 Gbps
(SStatus 113 SControl 310)
Apr 15 13:24:56 virt8 kernel: [ 297.914096] ata3.00: configured for PIO0
Apr 15 13:24:56 virt8 kernel: [ 297.914235] ata3: EH complete

The installed versions are:
linux-image-2.6.26-1-xen-686
xen-hypervisor-3.2-1-i386
ganeti 1.2.6-3+lenny2
ganeti-instance-debootstrap 0.5-1
drbd8-utils 2:8.0.14-2+lenny1

I upgraded to kernel 2-6-26-2 to see whether it solved the issue without
success.
I also tried booting with clocksource=acpi_pm. I've also just tried
acpi=off, pci=nosmi, and nocpi as suggests the ata kernel website without
much success.

I've checked through internet and found the following:

https://bugzilla.redhat.com/show_bug.cgi?id=480317

It should be solved because hipervisor version 3.2 is higher than the one
solving the issue (3.0.3) unless redhat's version differ much from the
debian ones in minnor version....which i don't think.


Can anyone shed some light on this issue?

Thank you in advance!!

Samuel