Ubuntu
linux package

2.6.27-7.14-generic problems booting

Bug #289754 reported by Angel Carpintero on 2008-10-27

This bug report is a duplicate of: Bug #296500: Update to 2.6.27.5 stable kernel. Edit Remove

Affects		Status	Importance	Assigned to	Milestone
	linux (Ubuntu)	Incomplete	Undecided	Unassigned

Bug Description

Binary package hint: linux-image-2.6.27-7-generic

No way to boot with kernel 2.6.27 any version i tried had same effect , error on ATA and fall to shell :

ALERT! /dev/disk/by-uuid/c44cc51d-88f9-4ca5-b133-7c1fb7772fff does not exist. Dropping to a shell!

, short log :

[ 5.627683] ata3: hard resetting link
[ 6.100104] ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[ 6.124560] ata3.00: configured for UDMA/100
[ 6.124646] ata3: EH complete
[ 6.124824] sd 2:0:0:0: [sda] 312581808 512-byte hardware sectors (160042 MB)
[ 6.125085] ata3: EH in SWNCQ mode,QC:qc_active 0x1 sactive 0x1
[ 6.125162] ata3: SWNCQ:qc_active 0x1 defer_bits 0x0 last_issue_tag 0x0
[ 6.125164] dhfis 0x0 dmafis 0x0 sdbfis 0x0
[ 6.125307] ata3: ATA_REG 0x51 ERR_REG 0x4
[ 6.125379] ata3: tag : dhfis dmafis sdbfis sacitve
[ 6.125451] ata3: tag 0x0: 0 0 0 1
[ 6.125533] ata3.00: NCQ disabled due to excessive errors
[ 6.125607] ata3.00: exception Emask 0x1 SAct 0x1 SErr 0x0 action 0x6 frozen
[ 6.125682] ata3.00: Ata error. fis:0x41
[ 6.125758] ata3.00: cmd 60/08:00:00:00:00/00:00:00:00:00/40 tag 0 ncq 4096 in
[ 6.125759] res 51/04:08:00:00:00/04:00:00:00:00/40 Emask 0x1 (device error)
[ 6.125928] ata3.00: status: { DRDY ERR }
[ 6.125999] ata3.00: error: { ABRT }

Attached initramfs debug , dmesg and /proc/modules.

Let me know if it's a known issue or do you need some more test , boot with some parameters , etc ...

Thanks !

- Angel

Revision history for this message

Angel Carpintero (sack) wrote on 2008-10-27:

log.tar.gz Edit (21.9 KiB, application/x-tar)

Revision history for this message

Leann Ogasawara (leannogasawara) wrote on 2008-10-28:

Hi Angel,

I've seen slightly similar bugs of being dropped to a busybox, bug 288790 and bug 288388 . However, they both comment that if they exit the busybox shell they are able to continue booting successfully. I'm not sure this is the case for you based on the errors you posted though. The error messages you've posted indicate an actual hw issue. Are you able to successfully boot a different kernel, say a 2.6.24 kernel from Hardy?

Changed in linux:
status:	New → Incomplete

Revision history for this message

Angel Carpintero (sack) wrote on 2008-10-28:

2.6.24-21-generic.tar.gz Edit (14.0 KiB, application/x-tar)

I checked bug 288790 and bug 288388 i don't think my issues is like those.
I can boot without issue with 2.6.24-21-generic , problem looks started since 2.6.26.x
I'll try to get latest working kernel.

Attached some logs .

Revision history for this message

Angel Carpintero (sack) wrote on 2008-11-08:

boot-kernel-2.6.28-rc3.log Edit (43.3 KiB, text/plain)

No way to boot in intrepid with any 2.6.27 .... so i can only boot with 2.6.24.
I've verified that is a bug introduce in kernel ... looks fixed in 2.6.28-rc3.

Attached boot message with 2.6.28 , where sata_nv works as expected.

Cheers,

Revision history for this message

Angel Carpintero (sack) wrote on 2008-11-10:

boot-kernel-2.6.28-rc4.log Edit (97.6 KiB, text/plain)

I've tested linux-image-2.6.27-8-generic same result i got with linux-image-2.6.27-7-generic.

Also i have tested 2.6.28-rc4 , that worked as 2.6.28-rc3.

Attached boot messages for 2.6.24-1 and 2.6.28-rc4, so looks that since 2.6.28-rc3 sata_nv with Nforce
controller was fixed ( i would say rather regression fix ). Problem appeared in 2.6.26 , not sure if also in 2.6.25

Cheers,

Revision history for this message

Leann Ogasawara (leannogasawara) wrote on 2008-11-11:

Thanks Angel,

The upcoming Jaunty kernel should hopefully resolve this issue for you. However, I think this would also qualify for an Intrepid SRU (Stable Release Update) assuming the patch that should be backported is not too invasive. I don't supposed you'd be able to help narrow down this specific patch? I'd think you should be able to do a git bisect to isolate the fix. The following wiki describes how to perform a git bisect, you'll just need to reverse the meanings of good and bad since you want to find the patch that fixes the problem http://www.kernel.org/doc/local/git-quick.html#bisect . Let me know if you have any questions or need any help and I can try to walk you through it. Thanks.

Revision history for this message

Angel Carpintero (sack) wrote on 2008-11-12:

boot-kernel-2.6.27.5.log Edit (38.7 KiB, text/plain)

Leann , i have tested current stable kernel ... and my issue is fixed as most of patches from 2.6.28-rc3 and rc4 related to sata_nv / sata and others have been backported to 2.6.27.5 .

I've attached a boot message using 2.6.27.5 , i think that bisect now is useless ...

So as you said , next Intrepid kernel will solve sata_nv and other issues , many regressions :-/

Cheers,

Revision history for this message

Leann Ogasawara (leannogasawara) wrote on 2008-11-19:

Hi Angel,

You'll want to keep an eye on bug 296500 then. It's tracking the 2.6.27.5 stable patches that will be applied to the Intrepid kernel as a Stable Release Update. It's still making it's way into the intrepid-proposed repository for testing. Actually, I think I'll just mark this bug as a duplicate of bug 296500 so you'll automatically be notified when you can test. Thanks.

Revision history for this message

Angel Carpintero (sack) wrote on 2008-11-20:

Leann , ok perfect i will add futher comments to bug 296500 , i did some new test with stable kernel 2.6.27.6 and also 2.6.28-rc5 both
worked fine and not issues.

Cheers,