Hard disk I/O randomly freezes when hald is running and optical drive is empty

[Keenan Pepper]

Binary package hint: hal

When I upgraded the kernel on my System76 Gazelle (basically a ASUS Z62FP without the Microsoft tax) from 2.6.17 to 2.6.20, the hard disk began freezing for 30 seconds every few minutes whenever the CD/DVD drive was empty. When there is a disk in the optical drive, the freezes occur much less often, but I'm sure there's been at least one even with a CD in.

The relevant part of the dmesg is:

[ 188.960000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 188.960000] ata1.01: cmd a0/00:00:00:00:20/00:00:00:00:00/b0 tag 0 cdb 0x0 data 0
[ 188.960000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask 0x4 (timeout)
[ 195.964000] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 218.980000] ata1: port failed to respond (30 secs, Status 0xd0)
[ 218.980000] ata1: soft resetting port
[ 219.332000] ata1.00: configured for UDMA/100
[ 219.516000] ata1.01: configured for UDMA/33
[ 219.516000] ata1: EH complete
[ 219.532000] SCSI device sda: 78140160 512-byte hdwr sectors (40008 MB)
[ 219.540000] sda: Write Protect is off
[ 219.540000] sda: Mode Sense: 00 3a 00 00
[ 219.900000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 219.904000] SCSI device sda: 78140160 512-byte hdwr sectors (40008 MB)
[ 219.904000] sda: Write Protect is off
[ 219.904000] sda: Mode Sense: 00 3a 00 00
[ 219.908000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

I'm filing this bug against HAL because kernel developer Tejun Heo says HAL is poking the CD/DVD drive and confusing it, and indeed when I kill hald the problem goes away. On the other hand, I only noticed the problem after upgrading the kernel, and when I force the old ide_generic driver to be used by blacklisting ata_piix, the problem also goes away, so maybe it should be filed against the kernel package instead.

dmesg and lspci -vvx attached

[Keenan Pepper (keenanpepper)]

I tried running hald in verbose mode with "hald --daemon=yes --verbose=yes --use-syslog", but no messages from hald appear in /var/log/syslog at the same time as the freeze. There are plenty of messages when hald starts up, but then it's silent when the freeze actually happens.

[Keenan Pepper (keenanpepper)]

Here's the output of "lshal" though, that might be useful.

[didier (did447-deactivatedaccount)]

Hi,
there's a process hald-addon-storage which, surprise, polls removable media. You can try to kill the one associated with your CD/DVD drive. But as you said it's surely a kernel bug and hald only triggers it.

[Keenan Pepper (keenanpepper)]

Oh, forgot to link to the LKML threads:

http://www.ussg.iu.edu/hypermail/linux/kernel/0702.0/0930.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0702.1/0657.html

[Kevin P (kevin-cybercolloids)]

I have seen a similar issue with my up to date feisty installation. AMD64 and two SATA drives. the problem appears to be with a new 250GB Maxtor. An older 80GB Samsung Spinpoint seems OK.

uname = Linux kryton 2.6.20-5-generic #2 SMP Sat Jan 6 09:44:32 UTC 2007 x86_64 GNU/Linux

dmesg = [ 1410.976979] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 1419.402905] ata1.00: limiting speed to PIO0
[ 1419.402910] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 1419.402917] ata1.00: cmd 20/00:68:86:20:6f/00:00:00:00:00/e2 tag 0 cdb 0x0 data 53248 in
[ 1419.402919] res 50/01:01:01:00:00/01:00:00:00:00/00 Emask 0x202 (HSM violation)

I also get
[ 236.863319] ata1.00: limiting speed to UDMA/100
[ 236.863323] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 236.863580] ata1.00: cmd ca/00:08:76:3a:b0/00:00:00:00:00/e3 tag 0 cdb 0x0 data 4096 out
[ 236.863582] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 236.863909] ata1: soft resetting port
[ 237.035354] ata1.00: configured for UDMA/100

[Florian Schmid (annaeus)]

Some here on feisty after upgrading to latest Kernel 2.6.20-13-generic:

[ 5249.792000] SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
[ 5249.792000] sda: Write Protect is off
[ 5249.792000] sda: Mode Sense: 00 3a 00 00
[ 5249.792000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 5519.372000] ata1.01: qc timeout (cmd 0xa0)
[ 5519.372000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 5519.372000] ata1.01: cmd a0/00:00:00:00:20/00:00:00:00:00/b0 tag 0 cdb 0x0 data 0
[ 5519.372000] res 51/20:03:00:00:00/00:00:00:00:00/b0 Emask 0x5 (timeout)
[ 5526.372000] ata1: port is slow to respond, please be patient (Status 0xd1)
[ 5549.388000] ata1: port failed to respond (30 secs, Status 0xd1)
[ 5549.388000] ata1: soft resetting port
[ 5549.740000] ata1.00: configured for UDMA/33
[ 5549.920000] ata1.01: configured for UDMA/33
[ 5549.920000] ata1: EH complete

[Kevin P (kevin-cybercolloids)]

My problem began after a Feisty update as well.

Try booting into recovery mode and running e2fsck on each partition.
Then add the kernel options acpi=off pci=bios That seems to have got my
system working again. I did some tests last night and could boot with
acpi=off but without the option I had problems. I have also noticed some
hard disk corruption now as well. The same disk worked with no problems
when I booted using an old knoppix disk. So I conclude there is a
problem somewhere between the libata driver and the hardware.

Florian Schmid wrote:
> Some here on feisty after upgrading to latest Kernel 2.6.20-13-generic:
>
> [ 5249.792000] SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
> [ 5249.792000] sda: Write Protect is off
> [ 5249.792000] sda: Mode Sense: 00 3a 00 00
> [ 5249.792000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
> [ 5519.372000] ata1.01: qc timeout (cmd 0xa0)
> [ 5519.372000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
> [ 5519.372000] ata1.01: cmd a0/00:00:00:00:20/00:00:00:00:00/b0 tag 0 cdb 0x0 data 0
> [ 5519.372000] res 51/20:03:00:00:00/00:00:00:00:00/b0 Emask 0x5 (timeout)
> [ 5526.372000] ata1: port is slow to respond, please be patient (Status 0xd1)
> [ 5549.388000] ata1: port failed to respond (30 secs, Status 0xd1)
> [ 5549.388000] ata1: soft resetting port
> [ 5549.740000] ata1.00: configured for UDMA/33
> [ 5549.920000] ata1.01: configured for UDMA/33
> [ 5549.920000] ata1: EH complete
>
>

[Kevin P (kevin-cybercolloids)]

Ran some tests last night:

1. Adding acpi=off as a boot parameter appears to make the system more stable. I tried various bios and kernel parameter options but the best combination was leaving the bios in all default settings and adding the acpi=off to the kernel. This could be a red herring as the problem seems to be intermittent anyway.

2. I only get problems with a 250GB Maxtor maxline iii drive. An 80Gb Samsung Spinpoint works with no issues.

3. Booting the system from an old Knoppix disk I got no issues. The issue appears to lie with how recent kernels interact with the maxline disk.

4. I am seeing some corruption and had to e2fsck my root partition.

[ 1565.181909] ata4.00: speed down requested but no transfer mode left
[ 1565.181914] ata4.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 1565.181920] ata4.00: cmd 20/00:80:46:a4:76/00:00:00:00:00/e2 tag 0 cdb 0x0 data 65536 in
[ 1565.181922] res 50/01:01:01:00:00/01:00:00:00:00/00 Emask 0x202 (HSM violation)
[ 1565.181932] ata4: soft resetting port
[ 1565.266976] ATA: abnormal status 0x7F on port 0xD007
[ 1565.272358] ATA: abnormal status 0x7F on port 0xD007
[ 1565.285001] ata4.00: configured for PIO0
[ 1565.285008] ata4: EH complete
[ 1565.306696] SCSI device sda: 490234752 512-byte hdwr sectors (251000 MB)
[ 1565.307302] sda: Write Protect is off
[ 1565.307304] sda: Mode Sense: 00 3a 00 00
[ 1565.308249] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

[Joey Stanford (joey)]

confirming this bug. It's happening on my laptop as well. I passed this to Kyle.

[Ben Collins (ben-collins)]

Please let the devs confirm bugs, or read the policies at https://wiki.ubuntu.com/KernelTeamBugPolicies for how to correctly handle kernel bugs.

Thanks

[Marc Tardif (cr3)]

I am experiencing the same problem on a System76 Darter Z35F running Feisty beta installed from the alternate CD. I am attaching information from the machine below.

[Kevin P (kevin-cybercolloids)]

After the previous editors comment about optical disks I have run my machine with a CD in both CD drives. However I triggered another 30s halt by plugging in a USB camera. Here is the dmseg.

[ 123.506818] usb 4-1: new full speed USB device using uhci_hcd and address 2
[ 123.691727] usb 4-1: configuration #1 chosen from 1 choice
[ 141.849608] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 141.849642] ata1.00: cmd c8/00:10:fe:1b:76/00:00:00:00:00/e2 tag 0 cdb 0x0 data 8192 in
[ 141.849644] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 145.349321] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 156.865767] ata1: port failed to respond (30 secs, Status 0xd0)
[ 156.865797] ata1: soft resetting port
[ 156.949132] ATA: abnormal status 0x7F on port 0x000000000001d007
[ 156.954547] ATA: abnormal status 0x7F on port 0x000000000001d007
[ 156.966548] ata1.00: configured for UDMA/133
[ 156.966555] ata1: EH complete
[ 156.983921] SCSI device sda: 490234752 512-byte hdwr sectors (251000 MB)
[ 156.983974] sda: Write Protect is off
[ 156.983976] sda: Mode Sense: 00 3a 00 00
[ 156.985481] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

[Curtis Hovey (sinzui)]

Well I can see I'm in good company here. My system76 Pangolin began throwing 'exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen' in the log, followed by a 'soft resetting port' message a few weeks ago. This last upgrade appears to have made the problem worse

[Curtis Hovey (sinzui)]

Linux autumn.annrky-sinzui.local 2.6.20-13-generic #2 SMP Sun Mar 25 00:21:25 UTC 2007 i686 GNU/Linux

[Kevin P (kevin-cybercolloids)]

Some notes from my testing.

DMESG after a reboot - The system locks 6 times, here are the lock ups plus what immediately precedes them.

[ 42.595211] input: USB HID v1.00 Mouse [Microsoft Microsoft Wheel Mouse Optical®] on usb-0000:00:10.1-2
[ 42.595225] usbcore: registered new interface driver usbhid
[ 42.595228] drivers/usb/input/hid-core.c: v2.6:USB HID core driver
[ 68.032891] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

[ 106.221210] skge eth0: enabling interface
[ 107.888688] skge eth0: Link is up at 100 Mbps, half duplex, flow control none
[ 136.306699] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

[ 170.722289] lp0: using parport0 (polling).
[ 170.763665] ieee1394: Initialized config rom entry `ip1394'
[ 201.006966] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

[ 201.260754] EXT3 FS on sda3, internal journal
[ 231.478370] ata1.00: limiting speed to UDMA/100:PIO4
[ 231.478375] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

[10889.413852] ISO 9660 Extensions: Microsoft Joliet Level 3
[10889.415648] ISO 9660 Extensions: RRIP_1991A
[10917.763403] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

[10947.987814] SCSI device sda: 490234752 512-byte hdwr sectors (251000 MB)
[10948.003947] sda: Write Protect is off
[10948.003950] sda: Mode Sense: 00 3a 00 00
[10948.024838] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[10983.654841] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

[Alan Ferrier (alan-ferrier)]

Workaround is to kill the hald-addon-storage process

[Scott Henson (scotth)]

I'd like to add a nice little me too to this bug. I have a system76 gazelle as well. I can say that 2.6.20-11-generic worked fine, while -12 and -13 exhibit the behavior above. Though I believe -11 uses the older non-libata driver. For now I have -11 set to be my primary kernel and I boot into the latest whenever I see an update to check if its fixed.

I also tried the cd in drive thing and it seemed to work for me. Once I took the cd out I got a freeze within a few minutes.

[Felix Heinonen (fheinonen)]

Another me too on a Fujitsu-Siemens Amilo M7440G
Dmesg:
[ 2817.976000] ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 2817.976000] ata3.00: cmd c8/00:80:d7:58:44/00:00:00:00:00/e8 tag 0 cdb 0x0 data 65536 in
[ 2817.976000] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 2824.976000] ata3: port is slow to respond, please be patient (Status 0xd0)
[ 2847.992000] ata3: port failed to respond (30 secs, Status 0xd0)
[ 2847.992000] ata3: soft resetting port
[ 2848.172000] ata3.00: configured for UDMA/100
[ 2848.172000] ata3: EH complete
[ 2848.604000] SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
[ 2848.604000] sda: Write Protect is off
[ 2848.604000] sda: Mode Sense: 00 3a 00 00
[ 2848.604000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

[Alan Ferrier (alan-ferrier)]

It would appear that this bug still exists in kernel 2.6.20-14-generic

Apr 5 18:56:35 localhost kernel: [ 2880.464000] ata1.01: qc timeout (cmd 0xa0)
Apr 5 18:56:35 localhost kernel: [ 2880.464000] res 51/20:03:00:00:00/00:00:00:00:00/b0 Emask 0x5 (timeout)
Apr 5 18:56:42 localhost kernel: [ 2887.468000] ata1: port is slow to respond, please be patient (Status 0xd0)
Apr 5 18:57:05 localhost kernel: [ 2910.484000] ata1: soft resetting port
Apr 5 18:57:06 localhost kernel: [ 2910.844000] ata1.00: configured for UDMA/100
Apr 5 18:57:06 localhost kernel: [ 2911.024000] ata1.01: configured for UDMA/33
Apr 5 18:57:06 localhost kernel: [ 2911.024000] ata1: EH complete
Apr 5 18:57:06 localhost kernel: [ 2911.032000] SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
Apr 5 18:57:06 localhost kernel: [ 2911.044000] sda: Write Protect is off
Apr 5 18:57:06 localhost kernel: [ 2911.064000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Apr 5 18:57:06 localhost kernel: [ 2911.084000] SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
Apr 5 18:57:06 localhost kernel: [ 2911.084000] sda: Write Protect is off
Apr 5 18:57:06 localhost kernel: [ 2911.088000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

[Scott Henson (scotth)]

2.6.20-15-generic exhibits the same errors

[ 283.180000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 283.180000] ata1.01: cmd a0/00:00:00:00:20/00:00:00:00:00/b0 tag 0 cdb 0x1e data 0
[ 283.180000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask 0x4 (timeout)
[ 290.184000] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 313.200000] ata1: port failed to respond (30 secs, Status 0xd0)
[ 313.200000] ata1: soft resetting port
[ 313.544000] ata1.00: ata_hpa_resize 1: sectors = 78140160, hpa_sectors = 78140160
[ 313.552000] ata1.00: ata_hpa_resize 1: sectors = 78140160, hpa_sectors = 78140160
[ 313.552000] ata1.00: configured for UDMA/100
[ 313.732000] ata1.01: configured for UDMA/33
[ 313.732000] ata1: EH complete
[ 313.740000] SCSI device sda: 78140160 512-byte hdwr sectors (40008 MB)
[ 313.740000] sda: Write Protect is off
[ 313.740000] sda: Mode Sense: 00 3a 00 00
[ 314.156000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 314.160000] SCSI device sda: 78140160 512-byte hdwr sectors (40008 MB)
[ 314.160000] sda: Write Protect is off
[ 314.160000] sda: Mode Sense: 00 3a 00 00
[ 314.160000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

[Marc Tardif (cr3)]

The same problem still occurs after installing 20070415.

[Marc Tardif (cr3)]

Here are steps to workaround the problem during the installation process:

  1. Add break=top to the kernel cmdline
  2. At the prompt, run: modprobe piix
  3. Then exit

And, here steps to make the workaround permanent after an installation:

  echo blacklist ata_piix | sudo tee -a /etc/modprobe.d/blacklist-ata
  echo piix | sudo tee -a /etc/initramfs-tools/modules
  sudo update-initramfs -u
  sudo reboot

The problem should be fixed for the first SRU release of Feisty.

[bonsiware (bonsiware-deactivatedaccount)]

tryed Marc's workaround, but now I can't boot...

[Carl Richell (carlrichell)]

VERY IMPORTANT: The workaround at the bottom of the report is only for NEW INSTALLATIONS. If the work around is used on an existing install your system will not boot!

[bonsiware (bonsiware-deactivatedaccount)]

OK... I'm very interested in this workaround, but I don't understand when to do what...

What does it mean "during the installation process"? Do I have to boot from the live cd? and follow the first three steps?

And what about the remaining four steps? when do I have to follow them?

Thanks for the patience

[Keenan Pepper (keenanpepper)]

Carl, I did pretty much the same thing on my existing installation and it works fine. It's just using the old piix driver instead of the new ata-piix driver. What's supposed to be the problem?

[holycow (mik-mars)]

confirmed, i posted in the wrong bug report.

same issue, kernel 2.6.20-15-generic on an asus z96f laptop.

i won't be trying the workarounds, i'll wait for a fix.

thx for the heads up on this bug.

[Goldenear (goldenear)]

Same bug here on my asus a8jc. (feisty kernel 2.6.20-15.27)

[220073.844000] SCSI device sda: 195371568 512-byte hdwr sectors (100030 MB)
[220073.844000] sda: Write Protect is off
[220073.844000] sda: Mode Sense: 00 3a 00 00
[220073.848000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[220643.280000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[220643.280000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0 cdb 0x25 data 8 in
[220643.280000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask 0x4 (timeout)
[220650.284000] ata1: port is slow to respond, please be patient (Status 0xd0)
[220673.300000] ata1: port failed to respond (30 secs, Status 0xd0)
[220673.300000] ata1: soft resetting port
[220673.648000] ata1.00: ata_hpa_resize 1: sectors = 195371568, hpa_sectors = 195371568
[220673.656000] ata1.00: ata_hpa_resize 1: sectors = 195371568, hpa_sectors = 195371568
[220673.660000] ata1.00: configured for UDMA/100
[220673.844000] ata1.01: configured for UDMA/25
[220673.844000] ata1: EH complete
[220673.848000] SCSI device sda: 195371568 512-byte hdwr sectors (100030 MB)
[220673.848000] sda: Write Protect is off
[220673.848000] sda: Mode Sense: 00 3a 00 00
[220673.848000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

[Carl Richell (carlrichell)]

More of a precaution than anything else. We wanted to test across more machines before customers started using the fix (we had mixed reports). We're putting together a .deb to automatically apply the fix and will push it down through our repo.

[Scott Henson (scotth)]

I'm curious where this deb is. I consider myself to be a reasonably clueful user/developer and Id love to provide some testing of the deb if at all possible.

[holycow (mik-mars)]

same here.

how do i access the deb to test this out?

[kshitiz (kshitiz-saxena)]

I am also facing same issue on connoi laptop. Hopefully we will get the patch soon.

[jafa (jafa)]

A word of warning if you are running a software RAID array - this bug can freeze for long enough that the RAID system think the drive has died and drops the drive.
The kernel will do a soft-reset and the drive will be alive again but it is no longer part of the array.

If you don't notice the problem then your RAID array will die the next time the error occurs.

In my case I got another error while rebuilding the array! Fortunately this time the kernel recovered before the RAID system noticed.

Moderators - please consider upgrading the importance of this bug. For people without software RAID the problem is an annoyance. For people with software RAID the problem is critical.

Nick

[adam0509 (adam050986)]

http://ubuntuforums.org/showthread.php?t=415892

Duplicate bug :

https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/64587

https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/107417

[Carl Richell (carlrichell)]

For System76 customers a fix has been released with the latest System76 Driver version 2.0.2. If this bug is effecting your machine first check Synaptic to make sure you have the latest 2.0.2 version then go to System > Administration > System76 Driver. Click the Install Drivers tab. Click Install and the reboot when prompted. Running the driver will not effect those of you that have already applied the fix.

We decided against a separate .deb package. In our testing the above work around has rendered some machines that do not require the fix unbootable. Our driver safely applies the work around only to effected System76 machines.

[Scott Henson (scotth)]

I applied the fix on a Gazel Value and got the attached in my dmesg.

Now it locks up even more regularly and its far more annoying.

[Mathieu Marquer (slasher-fun)]

Same bug for me with an Acer Aspire 1692WLMi and a Samsung Spinpoint M80 120GB IDE (not the original hard disk of the computer).
See bug #104581 for some other reports of this very annoying bug (since this bug makes the hard disk to corrupt data on all the partitions..)

[Tomas Šiaulys (tosi)]

Looks like the bug is caused by the buggy cd/dvd drive firmware. I crossflashed my drive and voila - everything works just great. No freezes, no lockups, everything runs just great. You must use SC03 firmware. If anyone wants to try out flashing the drive - I took instructions from another bug report here
https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/64587
The only thing I did in other way is that I used dos flasher instead of windows flasher, because windows flasher didn't seem to work for me.
Looks like the drive which is poking around is TssTcorp TS-L632D. I presume all of you have it too. Take a look at k3b and you'll find out which model you have.
If you decide to flash your drive - be extremely cautious - bad flashing can damage the drive! Please do it at your own risk.

[Tomas Šiaulys (tosi)]

By the way, my laptop model is Asus Z35F, just for information. This model is used for system76 darter too.

[chantra (chantra)]

Hi Tomas, did you leave it running for long?
I tried this trick a couple of month ago ( see https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.15/+bug/53754/comments/27 )
and did not fixed it. I will try again with the new firmware and hopefully will confirm

[Tomas Šiaulys (tosi)]

Hey,
I'm at work now, but I'll leave it for night when I'll be at home and will let you know how it goes. For now, everything is fine.
By the way, when I used the blacklisting workaround, the drive was freezing after long time anyway (when I left it for all night), though that workaround worked for short term (till battery lasts, approx 3-4 hrs).
I crossflashed my drive just yesterday - as I said, I'll leave it turned on for a longer period of time and let you know about the results.
Cheers,
Tomas

[timo18146 (tgo-serverplusplus)]

I can acknowledge that Crossflashing the drive firmware on my Acer Aspire 9410 solved the problem. The drive is still fully functional.

ThanX.

----- Original Message -----
From: "Tomas Siaulys" <email address hidden>
To: <email address hidden>
Sent: Friday, May 11, 2007 12:47:45 AM (GMT+0100) Europe/Berlin
Subject: [Bug 84603] Re: Hard disk I/O randomly freezes when hald is running and optical drive is empty

Looks like the bug is caused by the buggy cd/dvd drive firmware. I crossflashed my drive and voila - everything works just great. No freezes, no lockups, everything runs just great. You must use SC03 firmware. If anyone wants to try out flashing the drive - I took instructions from another bug report here
https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/64587
The only thing I did in other way is that I used dos flasher instead of windows flasher, because windows flasher didn't seem to work for me.
Looks like the drive which is poking around is TssTcorp TS-L632D. I presume all of you have it too. Take a look at k3b and you'll find out which model you have.
If you decide to flash your drive - be extremely cautious - bad flashing can damage the drive! Please do it at your own risk.

--
Hard disk I/O randomly freezes when hald is running and optical drive is empty
https://bugs.launchpad.net/bugs/84603
You received this bug notification because you are a direct subscriber
of a duplicate bug.

[live_linux_secure (francois-vaningelgom)]

Looks like for me to!
15hrs 25 running and still no visible output in dmesg I grep ata
Hoping so much...

Thanks!

Francois

[chantra (chantra)]

cqn you guys give a link to the flasher software you used? sfdnwin does not seem to recogize the switch -noverify

[Tomas Šiaulys (tosi)]

I have used sfdndos, you can find it here, along with instructions:
http://forum.rpc1.org/viewtopic.php?p=37412#37412
And I've used boot cd for flashing, which you can find here:
http://pioneerdvd.rpc1.org/index.html#BOOTISO
Be sure to have the flasher and the firmware itself on the hdd, NOT on the cd. Cd is used only for booting.
Although if you still wish to use windows to flash it, try "-nocheck" instead of "-noverify", that should work.
By the way, running for long time didn't produce any errors, so I presume that using this firmware is safe.

[chantra (chantra)]

sorry , I post too fast, the switch was -nocheck .

And I also found out that https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.15/+bug/53754/comments/27 was about applying version AC01 of the firmware and not crossflashing.

I am now crossing fingers ;)

By the way, does anybody knows a way to flash without going through windows? (fortunately this time I still have a small windows partition)

[Tomas Šiaulys (tosi)]

As I said earlier, you can use the boot cd. But you will still need a small fat32 partition for that. But no need for windows installation :)
You can use floppy too, but I don't think any laptop ships with floppies nowadays :)

[chantra (chantra)]

HO man that's great. Basically a CD and a usb bar will do :)

[wolfmanjm (morris-wolfman)]

I just upgraded from edgy to feisty and with the kernel at 2.6.20 I still get this problem, even with a cd in the cdrom drive.

The drive that hangs up is my second hard disk drive which is an ata drive with a sil based sata to ide converter SYBA SD-SATA-IDE,
which has been working fine in edgy.

The first drive is a native sata drive and seems to work fine.

If I boot with my previous (edgy) kernel 2.6.17-11-generic it all works fine, as it did under Edgy.

Even though my dvd drive is a TssTcorp, but not the one mentioned above, I believe this bug is in the kernel, and is quite serious.

Although it is odd that if I boot in single user mode, the drive does not give these errors, so it could be something in hald or another daemon (maybe smartd?) is interacting?

I even installed the latest stable kernel too 2.6.21.1 and get the same error.

For now I'll continue to use the edgy kernel, as it seems quite stable.

[ 48.628030] ata2: port is slow to respond, please be patient (Status 0xd0)
[ 71.628106] ata2: port failed to respond (30 secs, Status 0xd0)
[ 71.628112] ata2: soft resetting port
[ 71.793748] ATA: abnormal status 0xD0 on port 0x0001e807
[ 71.804318] ATA: abnormal status 0xD0 on port 0x0001e807
[ 71.814888] ATA: abnormal status 0xD0 on port 0x0001e807
[ 71.825451] ATA: abnormal status 0xD0 on port 0x0001e807
[ 71.836021] ATA: abnormal status 0xD0 on port 0x0001e807
[ 101.819152] ata2.00: qc timeout (cmd 0xec)
[ 101.819159] ata2.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[ 101.819162] ata2.00: revalidation failed (errno=-5)
[ 101.819202] ata2: failed to recover some devices, retrying in 5 secs
[ 113.865307] ata2: port is slow to respond, please be patient (Status 0xd0)
[ 136.840363] ata2: port failed to respond (30 secs, Status 0xd0)
[ 136.840406] ata2: soft resetting port
[ 137.010931] ATA: abnormal status 0xD0 on port 0x0001e807
[ 137.021500] ATA: abnormal status 0xD0 on port 0x0001e807
[ 137.032070] ATA: abnormal status 0xD0 on port 0x0001e807
[ 137.042633] ATA: abnormal status 0xD0 on port 0x0001e807
[ 137.053202] ATA: abnormal status 0xD0 on port 0x0001e807
[ 167.031476] ata2.00: qc timeout (cmd 0xec)
[ 167.031483] ata2.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[ 167.031486] ata2.00: revalidation failed (errno=-5)
[ 167.031523] ata2.00: limiting speed to UDMA/100:PIO3
[ 167.031526] ata2: failed to recover some devices, retrying in 5 secs
[ 179.068950] ata2: port is slow to respond, please be patient (Status 0xd0)

[chantra (chantra)]

it seems to me that the last firmware applied to my cd drive solved the issue.

wolfmanjm:
>>Although it is odd that if I boot in single user mode, the drive does not give these errors, so it could be something in hald or another daemon (maybe smartd?) is interacting?

yes, I believe hal is interacting with the devices and trigger the freeze.

[Fred K Huang (fredkhuang-gmail)]

I've been following this bug, and I've had the same kinds of errors as well. In addition, it seems as if having a disk in the CD drive prevents the system from freezing....my question would be however, why does Windows not have this behavior? I'm hesitant to flash my drive if it's a problem specifically in ubuntu....

[Tomas Šiaulys (tosi)]

Well, I think windows has some workaround or something or uses different drive access method. That firmware obviously fixes the problem, and it doesn't affect windows at all - the drive is functioning properly, including burning and reading CDs/DVDs. Well, it's your choice after all. I chose to upgrade firmware and everything works perfect now.

[Keenan Pepper (keenanpepper)]

Okay, here's the procedure I'm about to attempt. Please tell me if there's anything wrong with it.

1. Create a primary FAT partition and filesystem on my hard disk.
2. Get the firmware file from http://www.toshibaer.com/firmware/download.php?TS-L632D/TS-L632D_SC03.zip (unzip it, of course) and the flashing program from http://www.samsungodd.com/KorLib/File/sfdndos.exe and put them in the FAT filesystem.
3. Get the boot CD image from http://pioneerdvd.rpc1.org/boot_iso.zip, unzip it, and burn it on a CD-RW.
4. Boot from the CD-RW.
5. Find the FAT partition I made, navigate to it, and run "sfdndos TS-L632D_SC03.BIN PS -n" (the PS is for Primary Slave)

[wolfmanjm (morris-wolfman)]

Thius is all well and good however I have a TSSTcorp CD/DVDW SH-S183L and it has the latest firmware, so I can;t upgrade.

I think the point is being missed here that a bug in the driver is causing this problem, and flashing the cdrom drive to stop a hard disk from freezing up is a workaround whatever this bug is.

I hope the fact that flashing some cd drives works around the bug isn't stopping this bug reaching the correct developer.

Also it appears that any time the HAL is triggered (ie plugging in a USB flash drive etc) stimulates this bug apparently.

[Fred K Huang (fredkhuang-gmail)]

I concur with wolfmanjm, whether or not crossflashing works is beside the point, there is a real bug somewhere in the driver, and that's what should be fixed. Any developers have any ideas? Do we need to collect more information?

[wolfmanjm (morris-wolfman)]

Another point is if this were not a kernel bug, then how come everything works fine when I boot into kernel 2.6.17-11-generic?

The only difference being the kernel. (I am currently running feisty with kernel 2.6.17-11-generic so I can work!) would imply that HAL and the cd drive are not to blame.

Just my 2c worth.

[Lukáš Zapletal (lzap)]

Same on my machine with LG drive. But my drive has no firmware update and my computer gets rebooted when the error occurs (usually when it finishing with DVD-R writing).

https://bugs.launchpad.net/ubuntu/+bug/115647

[Fabio Povoledo (povvy)]

I've solved the problem editing /etc/initramfs-tools/modules, and adding this lines:

piix
ide_generic
ide_cd
ide_disk

# blacklist bad driver
blacklist ata_piix

# prevent unnecessary modules from being loaded (you don't need to do this)
blacklist ata_generic
blacklist libata
blacklist scsi_mod

after editing:

sudo update-initramfs -u

after reboot I have my hard disks with the old fashioned /dev/hd*, without bugs!!!:)

[live_linux_secure (francois-vaningelgom)]

If i do this trick my hard drive is also recognised as the old fashioned "/dev/hdaX" but DMA is off and it is impossible to turn it on!
I would like to have the denomination /dev/hdaX because when the system recognise my disk as a SATA it make a weird noise when it turn off!
That's why i would like to have it recognised as a ATA drive WITH the DMA!

Thanks François

[jasampler (jasampler)]

Hi, I have the same problem reported at the end of https://answers.launchpad.net/ubuntu/+question/5537
Apart of reinstalling Ubuntu 7.04, I also installed Debian to see if the problem could be solved that way and the computer does the same. Searching here about this I found a lot of similar bugs linked. These are the ones I've collected:

https://bugs.launchpad.net/bugs/106512 ATA Abnormal status

https://bugs.launchpad.net/bugs/53754 ata_piix problem with Intel ICH7 chipset

https://bugs.launchpad.net/bugs/84603 Hard disk I/O randomly freezes when hald is running and optical drive is empty

https://bugs.launchpad.net/bugs/37382 ata timeout freezes system partly

https://bugs.launchpad.net/bugs/64587 log says "ata1 is slow to respond, please be patient"

https://bugs.launchpad.net/bugs/107417 ata1: port failed to respond

https://bugs.launchpad.net/bugs/104581 Random freezes with "exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2" error

I'm discarding a fail in the hard disk (so I would try to return the hardware to the vendor), since there are many people reporting the error. I hope these links could be helpful. Thank you very much to those who are trying to solve this. If this is already solved please comfirm here. I can reinstall ubuntu to verify everything you want. My new computer just cannot run correctly whith this problem.

[Fabio Povoledo (povvy)]

For live_linux_source: in my case dma is turned on after appling the trick... so strange...

[okahei (okahei)]

Same for me on Ata disk

[ 4946.800000] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 4946.800000] ata1.00: cmd c8/00:08:dc:e0:03/00:00:00:00:00/ea tag 0 cdb 0x0 data 4096 in
[ 4946.800000] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 4953.804000] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 4976.820000] ata1: port failed to respond (30 secs, Status 0xd0)
[ 4976.820000] ata1: soft resetting port
[ 4976.984000] ata1.00: ata_hpa_resize 1: sectors = 312581808, hpa_sectors = 312581808
[ 4976.992000] ata1.00: ata_hpa_resize 1: sectors = 312581808, hpa_sectors = 312581808
[ 4976.992000] ata1.00: configured for UDMA/100
[ 4976.992000] ata1: EH complete
[ 4976.996000] SCSI device sda: 312581808 512-byte hdwr sectors (160042 MB)
[ 4976.996000] sda: Write Protect is off
[ 4976.996000] sda: Mode Sense: 00 3a 00 00
[ 4976.996000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

[okahei (okahei)]

Linux alfawave 2.6.20-16-generic #2 SMP Wed May 23 01:46:23 UTC 2007 i686 GNU/Linux

[wolfmanjm (morris-wolfman)]

I tried some new things to further the resolution on this BUG.

I disconnected the DVD drive, and it did not fix the problem.

I disabled smartmond (which was monitoring my two HDs and worked fine in edgy) and the problem went away.

I no longer got sdb freezing and going away when I booted.

So in this case it turns out to be an interaction between smartmond and the new ata-piix driver and not hald in my case.

I reconnected my cd, and everything still worked, until I put an Audio CD in the drive.

then I got the following errors on my main SATA Hard drive, which resolved itself after about 30 seconds and I was able to rip that cd.

Ripping worked fine under Edgy BTW.

I really regret upgrading to fiesty, but its too late to go back :(

This happened to my primary SATA HD after inserting an Audio CD into the DVD drive...

[ 693.413304] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 693.413312] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0 cdb 0x43 data 4 in
[ 693.413314] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask 0x4 (timeout)
[ 700.452975] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 723.429090] ata1: port failed to respond (30 secs, Status 0xd0)
[ 723.429095] ata1: soft resetting port
[ 724.914305] ata1.00: ata_hpa_resize 1: sectors = 586072368, hpa_sectors = 586072368
[ 724.980859] ata1.00: ata_hpa_resize 1: sectors = 586072368, hpa_sectors = 586072368
[ 724.980864] ata1.00: configured for UDMA/133
[ 725.146800] ata1.01: configured for UDMA/33
[ 725.146812] ata1: EH complete
[ 725.147503] SCSI device sda: 586072368 512-byte hdwr sectors (300069 MB)
[ 725.147667] sda: Write Protect is off
[ 725.147670] sda: Mode Sense: 00 3a 00 00
[ 725.147900] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 725.148149] SCSI device sda: 586072368 512-byte hdwr sectors (300069 MB)
[ 725.148261] sda: Write Protect is off
[ 725.148262] sda: Mode Sense: 00 3a 00 00
[ 725.148702] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

The initial dmesg for the SATA drives

[ 3.041958] ata1.00: ata_hpa_resize 1: sectors = 586072368, hpa_sectors = 586072368
[ 3.042001] ata1.00: ATA-7: ST3300620AS, 3.AAC, max UDMA/133
[ 3.042036] ata1.00: 586072368 sectors, multi 16: LBA48 NCQ (depth 0/32)
[ 3.042071] ata1.01: ATAPI, max UDMA/33
[ 3.042104] ata1.01: applying bridge limits
[ 3.108534] ata1.00: ata_hpa_resize 1: sectors = 586072368, hpa_sectors = 586072368
[ 3.108576] ata1.00: configured for UDMA/133
[ 3.275100] ata1.01: configured for UDMA/33
[ 3.275138] scsi1 : ata_piix
[ 3.499816] ata2.00: ata_hpa_resize 1: sectors = 390721968, hpa_sectors = 390721968
[ 3.499857] ata2.00: ATA-6: ST3200822A, 3.01, max UDMA/100
[ 3.499892] ata2.00: 390721968 sectors, multi 16: LBA48
[ 3.499925] ata2.00: applying bridge limits
[ 3.509867] ata2.00: ata_hpa_resize 1: sectors = 390721968, hpa_sectors = 390721968
[ 3.509906] ata2.00: configured for UDMA/100
[ 3.510032] scsi 0:0:0:0: Direct-Access ATA ST3300620AS 3.AA PQ: 0 ANSI: 5
[ 3.510979] scsi 0:0:1:0: CD-ROM ...

[Jason Vitosky (jvitosky)]

Guys, I too have an ASUS Z35f with TS-L632D drive. But when I try using the TS-L632D_SC03.bin firmware, it keeps saying that my TS-L632D_AS05 drive is incompatible!!!

What should I do? Should I use the TS-L632D_AS99.bin firmware instead? I don't see how that could help since AS05 and AS99 have the same modification date.

[Jason Vitosky (jvitosky)]

Ehh...nevermind, I found the answer above in Keenan's directions above (HINT: I forgot the -n switch):

"1. Create a primary FAT partition and filesystem on my hard disk (or usb drive)
2. Get the firmware file from http://www.toshibaer.com/firmware/download.php?TS-L632D/TS-L632D_SC03.zip (unzip it, of course) and the flashing program from http://www.samsungodd.com/KorLib/File/sfdndos.exe and put them in the FAT filesystem.
3. Get the boot CD image from http://pioneerdvd.rpc1.org/boot_iso.zip, unzip it, and burn it on a CD-RW.
4. Boot from the CD-RW.
5. Find the FAT partition I made, navigate to it, and run "sfdndos TS-L632D_SC03.BIN PS -n" (the PS is for Primary Slave)"

[Joey Stanford (joey)]

I was able to successfully cross-flash my System76 Darter Ultra (and thereby void the warranty no doubt) using Jason Vitosky's instructions above. So far this appears to have resolved my cdrom based problems.

[Fred K Huang (fredkhuang-gmail)]

Has there been any more word from developers in regards to when/if this issue will be resolved WITHOUT the need to crossflash? As stated above, this does appear to be an issue with the linux kernel, hald, etc. And in particular, this issue does not appear in MS Windows. Crossflashing is not only an impractical and impossible task to ask the majority of Ubuntu's target base to do, but also runs the risk of voiding one's warranty or leaving one's hardware inoperable.

Can a developer at least clarify whether this issue is occuring in the linux-kernel, or hald, smartmond, etc. so we can refocus our efforts?

[wolfmanjm (morris-wolfman)]

Not only that, but in many cases cross flashing is not an option, as it happens with different models of cdrom drives, and sometimes even in the absence of a cdrom entirely. So good question, will this be fixed any time soon?

[chantra (chantra)]

I guess crossflashing is not an issue regarding to warranty as long as you can flash back to the original state (or to a supported one).

[Fred K Huang (fredkhuang-gmail)]

Regardless of whether or not it voids your warranty, the fact still stands that crossflashing is risky and certainly not a solution for the majority of Ubuntu's targeted base.

[chantra (chantra)]

There is a bug upstream, I have just added the link to launchpad

http://bugzilla.kernel.org/show_bug.cgi?id=8044

maybe a bit of activity should be shown upstream. This bug affects every single linux distros

[Tritonio (inshame)]

I has a similar problem. This is what I find in the system log.
Jun 17 22:25:50 Tritonio kernel: [22870.792000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
Jun 17 22:25:50 Tritonio kernel: [22870.792000] ata1.01: cmd a0/00:00:00:00:20/00:00:00:00:00/b0 tag 0 cdb 0x1e data 0
Jun 17 22:25:50 Tritonio kernel: [22870.792000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask 0x4 (timeout)
Jun 17 22:25:57 Tritonio kernel: [22877.792000] ata1: port is slow to respond, please be patient (Status 0xd0)
Jun 17 22:26:20 Tritonio kernel: [22900.808000] ata1: port failed to respond (30 secs, Status 0xd0)
Jun 17 22:26:20 Tritonio kernel: [22900.808000] ata1: soft resetting port
Jun 17 22:26:21 Tritonio kernel: [22901.500000] ata1.00: ata_hpa_resize 1: sectors = 234441648, hpa_sectors = 234441648
Jun 17 22:26:21 Tritonio kernel: [22901.508000] ata1.00: ata_hpa_resize 1: sectors = 234441648, hpa_sectors = 234441648
Jun 17 22:26:21 Tritonio kernel: [22901.508000] ata1.00: configured for UDMA/100
Jun 17 22:26:21 Tritonio kernel: [22901.688000] ata1.01: configured for UDMA/33
Jun 17 22:26:21 Tritonio kernel: [22901.688000] ata1: EH complete
Jun 17 22:26:21 Tritonio kernel: [22901.696000] SCSI device sda: 234441648 512-byte hdwr sectors (120034 MB)
Jun 17 22:26:21 Tritonio kernel: [22901.712000] sda: Write Protect is off
Jun 17 22:26:21 Tritonio kernel: [22901.712000] sda: Mode Sense: 00 3a 00 00
Jun 17 22:26:21 Tritonio kernel: [22901.736000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Jun 17 22:26:21 Tritonio kernel: [22901.760000] SCSI device sda: 234441648 512-byte hdwr sectors (120034 MB)
Jun 17 22:26:21 Tritonio kernel: [22901.776000] sda: Write Protect is off
Jun 17 22:26:21 Tritonio kernel: [22901.776000] sda: Mode Sense: 00 3a 00 00
Jun 17 22:26:21 Tritonio kernel: [22901.792000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

I get this error many times a day! It's really annoying because the whole system locks up for 30 seconds. At lest Wanda the fish still suims and Geyes follow the mouse pointer so I have something to play with. The funny thing is that I had this problem with windows XP too since I bought the laptop! But back then I wasn't even able to see what the problem was... I will try to update the firmware soon and post the results here or in one of the 6 duplicates. I have an Acer Aspire 9424WSMi with a TSSTcorp CD/DVDW TS-L632D.

[chantra (chantra)]

changed to bug number 8316 upstream as per Tejun Heo comment

[chantra (chantra)]

th bug number is to be upstream bug number: http://bugzilla.kernel.org/show_bug.cgi?id=8316

[Laurent (laurent-goujon)]

Same bug as #75295

[Matthias (m-kaeppler)]

I am experiencing the same issue on a Samsung R55 Notebook.

Plus, I noticed a major slowdown in Hard Drive I/O. It's ~4 times slower now than with Edgy. This is a critical issue for us, as we need to perform time consuming operations on a database. Can you please flag this bug as major or critical so it gets more attention?

[Matthias (m-kaeppler)]

Here's a performance measurement of my hard drive. I don't know if the values are normal as I don't have anything to compare it to:

/dev/sda:
 Timing cached reads: 1670 MB in 2.00 seconds = 835.11 MB/sec
 Timing buffered disk reads: 114 MB in 3.00 seconds = 37.96 MB/sec

[chantra (chantra)]

FYI:

the upstream bug: http://bugzilla.kernel.org/show_bug.cgi?id=8316 is pretty active.

A debugging patch is available: see http://bugzilla.kernel.org/show_bug.cgi?id=8316#c73

Would be great if some people could spare some time patching their kernel and sending the output there.

Looks like things are shaking a bit up there.

[Scott Henson (scotth)]

An interesting bit of information is that under kernel 2.6.22-6-generic the situation seemed to have fixed itself. I was getting the errors to the log, but I was not noticing the lockups. Now, with kernel 2.6.22-7 the lockups are becoming very noticable. Looking through the changelog I see nothing specific that would point to what changed, but its a long list and I'm sure I missed something. This is on an up-to-date gutsy install on a gazelle that was upgraded from feisty. The libata blacklist fix is not applied on this system.

[Scott Henson (scotth)]

So, I have a stupid question. Why in the world is hal polling a cdrom anyway? Can't it just listen for open/close events and check the drive then? Is the problem that nothing generates these type of events? I know people seem to have settled on this being a kernel bug, but could we perhaps convince hald to play nicer?

[Travis Reitter (treitter-dev)]

To confirm what Scott Henson said, I've noticed the lockups in 2.6.22-8 (only kernel I see in my Gutsy Tribe 3 menu.lst). This is also on a Gazelle laptop from System 76.

[Roberto Pinho (robertopinho)]

I've also had the problem, but seems to be working now with this Kernel :

 2.6.20-16-386 ( linux-image-2.6.20-16-386 )

[Adam (adam.russell)]

Kubuntu Feisty Fawn
kernel 2.6.20-16-generic

I believe the problem is occuring with my SATA hard drive, which is a Samsung SP0812C. It stops responding quite often. It seems that most people that are having this problem are fixing it with 'piix', but I have a VIA chipset rather than an Intel one. I will attach some logs below. Relevant dmesg lines:

[ 2416.505383] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 2416.505394] ata1.00: cmd c8/00:90:e6:54:5b/00:00:00:00:00/e2 tag 0 cdb 0x0 data 73728 in
[ 2416.505396] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 2416.505414] ata1: soft resetting port
[ 2416.671775] ATA: abnormal status 0x7F on port 0x0001e007
[ 2416.682280] ATA: abnormal status 0x7F on port 0x0001e007
[ 2416.693332] ata1.00: ata_hpa_resize 1: sectors = 156368016, hpa_sectors = 156368016
[ 2416.705299] ata1.00: ata_hpa_resize 1: sectors = 156368016, hpa_sectors = 156368016
[ 2416.705303] ata1.00: configured for UDMA/100
[ 2416.705314] ata1: EH complete
[ 2416.727567] SCSI device sda: 156368016 512-byte hdwr sectors (80060 MB)
[ 2416.727651] sda: Write Protect is off
[ 2416.727653] sda: Mode Sense: 00 3a 00 00
[ 2416.741889] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 2446.704938] ata1.00: limiting speed to UDMA/33:PIO4
[ 2446.704945] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 2446.704953] ata1.00: cmd c8/00:88:e6:7d:5d/00:00:00:00:00/e1 tag 0 cdb 0x0 data 69632 in
[ 2446.704955] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 2453.697731] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 2476.684824] ata1: port failed to respond (30 secs, Status 0xd0)
[ 2476.684831] ata1: soft resetting port
[ 2476.855210] ATA: abnormal status 0x7F on port 0x0001e007
[ 2476.865714] ATA: abnormal status 0x7F on port 0x0001e007
[ 2476.876786] ata1.00: ata_hpa_resize 1: sectors = 156368016, hpa_sectors = 156368016
[ 2476.888756] ata1.00: ata_hpa_resize 1: sectors = 156368016, hpa_sectors = 156368016
[ 2476.888760] ata1.00: configured for UDMA/33
[ 2476.888771] ata1: EH complete
[ 2476.909818] SCSI device sda: 156368016 512-byte hdwr sectors (80060 MB)
[ 2476.912474] sda: Write Protect is off
[ 2476.912478] sda: Mode Sense: 00 3a 00 00
[ 2476.916458] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[ 2507.060115] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 2507.060126] ata1.00: cmd c8/00:20:b6:2d:04/00:00:00:00:00/e2 tag 0 cdb 0x0 data 16384 in
[ 2507.060128] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 2514.052911] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 2537.032018] ata1: port failed to respond (30 secs, Status 0xd0)
[ 2537.032025] ata1: soft resetting port
[ 2537.198411] ATA: abnormal status 0x7F on port 0x0001e007
[ 2537.208918] ATA: abnormal status 0x7F on port 0x0001e007
[ 2537.215986] ata1.00: ata_hpa_resize 1: sectors = 156368016, hpa_sectors = 156368016
[ 2537.223961] ata1.00: ata_hpa_resize 1: sectors = 156368016, hpa_sectors = 156368016
[ 2537.223964] ata1.00: configured for U...

[nicosiaf77 (francesco-nicosia)]

Ubuntu Feisty Fawn
kernel 2.6.20-16-generic

I have an Acer Aspire 9424WSMi with a TSSTcorp CD/DVDW TS-L632D

Aug 31 02:28:37 gabema kernel: [ 1094.812000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
Aug 31 02:28:37 gabema kernel: [ 1094.812000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0 cdb 0x25 data 8 in
Aug 31 02:28:37 gabema kernel: [ 1094.812000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask 0x4 (timeout)
Aug 31 02:28:44 gabema kernel: [ 1101.812000] ata1: port is slow to respond, please be patient (Status 0xd0)
Aug 31 02:29:07 gabema kernel: [ 1124.828000] ata1: port failed to respond (30 secs, Status 0xd0)
Aug 31 02:29:07 gabema kernel: [ 1124.828000] ata1: soft resetting port
Aug 31 02:29:08 gabema kernel: [ 1125.564000] ata1.00: ata_hpa_resize 1: sectors = 234441648, hpa_sectors = 234441648
Aug 31 02:29:08 gabema kernel: [ 1125.572000] ata1.00: ata_hpa_resize 1: sectors = 234441648, hpa_sectors = 234441648
Aug 31 02:29:08 gabema kernel: [ 1125.572000] ata1.00: configured for UDMA/100
Aug 31 02:29:08 gabema kernel: [ 1125.752000] ata1.01: configured for UDMA/33
Aug 31 02:29:08 gabema kernel: [ 1125.752000] ata1: EH complete
Aug 31 02:29:08 gabema kernel: [ 1125.768000] SCSI device sda: 234441648 512-byte hdwr sectors (120034 MB)
Aug 31 02:29:08 gabema kernel: [ 1125.776000] sda: Write Protect is off
Aug 31 02:29:08 gabema kernel: [ 1125.776000] sda: Mode Sense: 00 3a 00 00
Aug 31 02:29:08 gabema kernel: [ 1125.784000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Aug 31 02:29:08 gabema kernel: [ 1125.784000] SCSI device sda: 234441648 512-byte hdwr sectors (120034 MB)
Aug 31 02:29:08 gabema kernel: [ 1125.784000] sda: Write Protect is off
Aug 31 02:29:08 gabema kernel: [ 1125.784000] sda: Mode Sense: 00 3a 00 00
Aug 31 02:29:08 gabema kernel: [ 1125.784000] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA

I don't find any updated firmware for CD/DVDW TS-L632D, please help me.

[Gareth Fitzworthington (mapping-gp-deactivatedaccount)]

Readers of this bug commentary should also examine the following related bugs:
Bug #75295
Bug #117441

These show the problems that the TSSTcorpCD/DVD TS-L632D optical drive has been causing for both Linux & MSWindows users.
They also show that the solution requires a firmware upgrade.

[ori (ori-livneh)]

I can confirm that the SC04 firmware from Samsung fixed this issue for my TSST TS-L632D. I used Hiren's Boot CD to boot into DOS with USB support and launched the firmware upgrade from a USB jumpdrive.

[Flávio Etrusco (etrusco)]

I truly hope there's less people affected by these libata issues than it seems, or 7.10 will be a fiasco :-(
I'm "kind of lucky" that I only started to have problems in 2.6.22, in Feisty devel, and problem is not exactly this, but a probably less annoying one (that I can't find anymore on launchpad :-$ ) that freezes during boot for 2 minutes.

I've been following these issues on launchpad and lkml and it seems to be a big deal.
Granted, 2.6.23 works beautifully for me; I don't know whether it was the several SiS 5513 fixes that went in, the (previously unimplemented in libata) cable (speed) detection, whether my LG 4480B combo drive was blacklisted or whether updating the firmware helped the new kernel. All I know is that 30s still seems way to long even if we were talking SCSI devices and trying 3 times before dropping to ATA33 (still) seems pointless to me. But who am I to tell :-/

It is also unfortunate that Ubuntu removed (most of?) the old IDE modules/drivers, thus impeding the option to simply blacklist the (corresponding) libata driver...

[nicosiaf77 (francesco-nicosia)]

I can confirm that the SC04 firmware from Samsung fixed this issue for my TSST TS-L632D. My Acer 9424 go to 100% of Performance and nothing freeze of system.

[CassieMoondust (cassie-lx)]

I have the same bug like described in the bugdesciption on a nVidia 590 based mainboard (ASUS M2N32 WS Professional).
There are two SATA Samsung 250GB harddisks connected and one Samsung SH-182M dvd-writer at the IDE-port of the board.
With this issue i cannont create a working software raid.
Can i use the same workaround as described above (blacklist ata_piix) or is this for intel-boards only?

[K (kkumar)]

I do have this problem with my new SATA disk. OS is running on ATA disk, so ubuntu runs fine but my mythtv and all other media writes into my SATA and it is keep freezing. My mother board and hard drive as follows
GIGABYTE GA-K8N Pro-SLI 939 NVIDIA nForce4 SLI ATX AMD Motherboard with 1394b
Western Digital SE16 500GB SATA2 7200RPM 16MB Cache 8.9MS NCQ Hard Drive OEM

[CassieMoondust (cassie-lx)]

I've solved the bug for me, one of the sata cables was bad - i've changed all cables to original asus cables (wich delivered with my ASUS-Board)
Didn't believe that a cable can cause so much trouble...now it works for about two weeks 24 hours a day without an error.

[Wolfram Arnold (wolframarnold)]

etrustco:
"I truly hope there's less people affected by these libata issues than it seems, or 7.10 will be a fiasco :-(
...
It is also unfortunate that Ubuntu removed (most of?) the old IDE modules/drivers, thus impeding the option to simply blacklist the (corresponding) libata driver.."

I can confirm this. I'm not even using cutting edge hardware. Ubuntu 5.10 has fewer disk issues than 7.04 and 7.10. 7.10 is the worst. I get these freeze-up events, and after a lot of Googling and reading tried to force the system back to the old ide-driver, just to realize that the piix module isn't part of Gusty any more, and ide-generic doesn't give me DMA :-(

[K (kkumar)]

It is really pain for me,

I have two disks, first one is ATA , 10GB, loaded with OS, running fine.

Second one is SATA, 500GB, this is for all my files, media etc. This disk
keep freezing and it is really annoying when I try to download some file
from internet or my mythtv automatic recording, watching tv etc.

I will be glad if some one tells me how to restore drive back online without
rebooting my machine. dmesg logs attached.

On Nov 15, 2007 9:44 PM, Wolfram Arnold < <email address hidden>> wrote:

> etrustco:
> "I truly hope there's less people affected by these libata issues than it
> seems, or 7.10 will be a fiasco :-(
> ...
> It is also unfortunate that Ubuntu removed (most of?) the old IDE
> modules/drivers, thus impeding the option to simply blacklist the
> (corresponding) libata driver.."
>
> I can confirm this. I'm not even using cutting edge hardware. Ubuntu
> 5.10 has fewer disk issues than 7.04 and 7.10. 7.10 is the worst. I
> get these freeze-up events, and after a lot of Googling and reading
> tried to force the system back to the old ide-driver, just to realize
> that the piix module isn't part of Gusty any more, and ide-generic
> doesn't give me DMA :-(
>
> --
> Hard disk I/O randomly freezes when hald is running and optical drive is
> empty
> https://bugs.launchpad.net/bugs/84603
> You received this bug notification because you are a direct subscriber
> of the bug.
>

[Db0 (db0)]

I can confirm I have something quite similar.

http://ubuntuforums.org/showthread.php?p=3784478

I am going to try playing with the cables a bit and see if I can disaple that piix.

[suoko (suoko)]

Problem is still there.
As I posted here http://ubuntuforums.org/showthread.php?t=598580, feisty kernel under gutsy + blacklisting some modules is a way to solve this problem, although this causes problems with usb devices (i.e. I can't mount camera anymore although feisty could do it with no problems) and cpu scaling.
I guess we'd need a new gutsy kernel or a customized one.

[suoko (suoko)]

THE SOLUTION IS HERE!
https://bugs.launchpad.net/linux/+bug/75295/comments/97
thanks all

[pdm (patrice-sancey)]

Hi,

this solution is for people who have TSS corp DVD driver. I have a Matshita one.

wolfram Arnold wrote that piix module is not in Gutsy : what do you mean : can't we add piix in /etc/initramfs-tools/modules ?

What can we do to make Gutsy possible to (simply) use ?

[K (kkumar)]

Does it help me by disconnecting my DVD drive from mother board? I use DVD
drive very very rare.

On Nov 20, 2007 8:41 AM, pdm <email address hidden> wrote:

> Hi,
>
> this solution is for people who have TSS corp DVD driver. I have a
> Matshita one.
>
> wolfram Arnold wrote that piix module is not in Gutsy : what do you mean
> : can't we add piix in /etc/initramfs-tools/modules ?
>
> What can we do to make Gutsy possible to (simply) use ?
>
> --
> Hard disk I/O randomly freezes when hald is running and optical drive is
> empty
> https://bugs.launchpad.net/bugs/84603
> You received this bug notification because you are a direct subscriber
> of the bug.
>

[Laurent (laurent-goujon)]

I'm still using the ide_cd module (sata_nv is disabled on Gutsy) and also have the problem so disabling ata-piix or sata modules won't have any effect.

[Igor Lautar (igorl)]

I confirm this bug on HP nc8430 with upgraded drive.

Summary:
I just upgraded existing drive (Fujitsu MHV2100BH) to Seagate 7200.2 (ST9200420ASG). Previous drive worked fine with edgy and feisty (did not tried it with gutsy).
After upgrade, I went and installed feisty. The same lockups (as described here) apear on random.

[Igor Lautar (igorl)]

Typo in my previous comment:
"After upgrade, I went and installed feisty"
I've actually installed gutsy.

[mezhaka (mezhaka)]

I confirm that the problem has gone after updating the firmware of the cd drive. it is an asus laptop (6000 series) with the afore mentioned TSST drive. I used the blah-blahSC04 version of firmware. sfdndos utility mentioned above. tried to make a bootable usb, but could not boot using it, so i burned boot CD from http://pioneerdvd.rpc1.org/index.html#BOOTISO

thanks to all of those who contributed.

[Kevin P (kevin-cybercolloids)]

A few more notes from testing. A lot of people talking about CD-ROM drives. I have two optical drives, one Samsung and one NEC.

I disconnected the Samsung drive - same problem.

Connect both optical drives plus Samsung SATA harddrive - problem

Connect both optical drives plus Maxtor SATA harddrive - problem

Disconnect SATA hardrive - no problem

From what I can tell the issue is with the SATA harddrive and not with any of the optical drives. Currently I am using the system everyday with both optical drives and an IDE hard drive with no problems. As soon as I reconnect the SATA hard drive - problems.

The SATA harddrive has been used before in a Ubuntu box with no problems - these issues developed sometime around an upgrade to Feisty. Currently I am using Gutsy with exactly the same problems.

Another point - the piix workaround doesn't seem to be the key to the issue - I have a via board and I get the problem.

[arjanhs (arjan-advance)]

I'm having the same problem after updating the Gutsy kernel to 2.6.22-14, after a reboot i got the following errors:

ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
ata3.00: cmd c8/00:20:88:d3:d7/00:00:00:00:00/e1 tag 0 cdb 0x0 data 16384 in
res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
ata3: port is slow to respond, please be patient (Status 0xd0)
ata3: device not ready (errno=-16), forcing hardreset
ata3: soft resetting port
ata3.00: configured for UDMA/133
ata3: EH complete
488397168 512-byte hardware sectors (250059 MB)
Write Protect is off
sd 2:0:0:0: [sda] Mode Sense: 00 3a 00 00
sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA

It's a new drive which i'm using for three months now.

[K (kkumar)]

I have same problem with my SATA, it works good in windows. I have
another question. If disk fails in the middle of copying then, I had to
reboot to set it back online. Is there any way to re-attach this disk
without rebooting?

Arjanhs wrote:
> I'm having the same problem after updating the Gutsy kernel to
> 2.6.22-14, after a reboot i got the following errors:
>
> ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
> ata3.00: cmd c8/00:20:88:d3:d7/00:00:00:00:00/e1 tag 0 cdb 0x0 data 16384 in
> res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
> ata3: port is slow to respond, please be patient (Status 0xd0)
> ata3: device not ready (errno=-16), forcing hardreset
> ata3: soft resetting port
> ata3.00: configured for UDMA/133
> ata3: EH complete
> 488397168 512-byte hardware sectors (250059 MB)
> Write Protect is off
> sd 2:0:0:0: [sda] Mode Sense: 00 3a 00 00
> sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
>
> It's a new drive which i'm using for three months now.
>
>

--
With Regards,
Kiran Kamsetti

[Luka Renko (lure)]

This is also reproducible with latest Hardy version on HP nw8440 with Seagate disk ST980825AS

[Luka Renko (lure)]

Maybe we should implement hal-info quirk that would disable polling on such HW, as it was done for Dell laptops - see bug 48499

[Leann Ogasawara (leannogasawara)]

Luka,

Care to quickly verify the issue exists against 2.6.24-11 and reattach your dmesg as well as lspci -vvnn output? Thanks.

[richard philippe (rifi58)]

Ubuntu run normaly only when a cd/dvd is inside...

Why the hell isn't this one year old bug marked as critical ?!?

[Kevin P (kevin-cybercolloids)]

I agree this bug seems to be serious - it renders your computer useless. A 30s delay in hard drive access is pretty useless in my books. Also the CD in a CD-Drive trick doesn't seem to work for me. The bug seems to be present for both Maxtor and Samsung hard drives I have tried. I have a small network of Ubuntu desktops and servers in our company and currently have an order for new equipment on hold because of this bug. New motherboards tend to be mainly SATA and I am not confident enough to buy a load of new SATA boards only to find they don't run with Ubuntu.

[Luka Renko (lure)]

I have noticed that during this hick-ups, "htop" shows at least one (in most cases both) fully loaded (100%), even though that the top process on the list occupies cca 10%. htop also presents that CPU load as red, so I suspect this is CPU time used in kernel. I suspect kernel is spinning on something in this case.

I use amd64 kernel on HP nw8440 laptop with Seagate 7200 RPM disk (ST980825AS)

[Gareth Fitzworthington (mapping-gp-deactivatedaccount)]

Can anybody still experiencing this problem test the following possible solution?
http://linux-ata.org/faq.html#combined
The slow down as a result of the PATA/SATA combination may be causing at least some of the above mentioned problems.
This will probably only affect those with Intel chipsets.

[Brian Murray (brian-murray)]

I'm unassigning this bug as bugs should only be assigned when someone is working on a fix for the bug and this doesn't seem to be the case.

[Melekai (mgeuken)]

as of today im experiencing similar issues. im really unsure if this is the right place. but basically when i have a SATA drive plugged in the system wont boot.
previously with just one SATA and main disk as a IDE, i would boot but get random 10-30 seconds freezes in the system.
when the sata drive was unplugged those errors were gone.

when ONLY one sata drive is plugged in. it will randomly boot the live cd and randomly crash. (hardy 64/beta/rc)
when more than one sata drive is plugged in. will never boot. always some kind of error. ... live environment just wont load.
with any other version on ide drivers. this does not happen.

[Kevin P (kevin-cybercolloids)]

The problem is solved on my machine - I replaced the SATA cabling with better quality Akasa latch cables instead of the cheap cables that come with the motherboards. So far I have installed and run Ubuntu/Hardy and installed Gentoo (on another partition) and compiled X/Gnome with no problems. The machine has been rebooted several times and has not frozen once.

So I can confirm that one possible cause of the error -

[ 231.478375] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

Could be poor quality cabling.

Googling around problems with cheap SATA cabling are very common.

[Jochen Garcke (jochen-bugs)]

I also had problems like this on my nc6220 laptop starting with gutsy, often when moving the laptop the system hang for a bit and activity in the (empty) cd drive could be heard.

Since upgrading to Hardy this seems to be over, only very short breaks if at all when moving the laptop to access the cd drive.

[weks (na18)]

i also have this problem with my asus notebook with ata hdd...pc locks up randomly and here is system log...is there any fix for this?

[kiev1 (sys-sys-admin)]

This kernel bug
 this problem already whole year

for me she showed up one time in the floor of hour, however as a result of this problem I lost a mysql database - mysql innodb not start - "Accertion error" - did not help even "innodb_force_recovery = 4", backup was an a week remoteness - the works of whole department lost data for a few days, the management simply in shock - I going to discharge from job (((

this problem already whole year:
-----------
I'm stumped trying to track down the below intermittent problem.....
I've confirmed this problem on 2.6.19, 2.6.20 and 2.6.21.
http://lkml.org/lkml/2007/6/14/154
http://kerneltrap.org/mailarchive/linux-kernel/2007/6/14/103765
http://kerneltrap.org/node/16175
http://lkml.org/lkml/2007/6/14/154
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/217920
https://bugs.launchpad.net/ubuntu/+bug/164183
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/229747
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/159521
https://bugs.launchpad.net/ubuntu/+bug/164183
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/187146
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/221437
https://bugs.launchpad.net/ubuntu/+bug/226600

SUSE:

ata errors, system freeze
https://bugzilla.novell.com/show_bug.cgi?id=393675

System lockup with concurrent acces to SATA disks on Promise PDC20378
http://lists.opensuse.org/opensuse-bugs/2008-02/msg03458.html

Kernel panic / system hang / sata_promise
https://bugzilla.novell.com/show_bug.cgi?id=350907

DELL Poweredge 2970 hangs sometimes (ata1)
https://bugzilla.novell.com/show_bug.cgi?id=359333

Fedora:
ata device crashing system in Fedora 8
http://www.experts-exchange.com/OS/Linux/Distributions/Fedora/Q_23125450.html

problème de mise à jour
http://forums.fedora-fr.org/viewtopic.php?pid=253930

Kernel 2.6.24.x boot problem - Anyone , Any idea
http://fcp.surfsite.org/modules/newbb/viewtopic.php?viewmode=flat&order=ASC&topic_id=54760&forum=10

Thought though with the newest hard drive with support of NCQ such is not present, ... also same:

"With this kernel I’m getting frequent temporary freezes (system comes back responsive after a minute or so…)."
http://kerneltrap.org/mailarchive/linux-kernel/2008/1/8/546296

[Leann Ogasawara (leannogasawara)]

The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would appreciate it if you could please test this newer 2.6.27 Ubuntu kernel. There are one of two ways you should be able to test:

1) If you are comfortable installing packages on your own, the linux-image-2.6.27-* package is currently available for you to install and test.

--or--

2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4. Please watch http://www.ubuntu.com/testing for Alpha5 to be announced. You should then be able to test via a LiveCD.

Please let us know immediately if this newer 2.6.27 kernel resolves the bug reported here or if the issue remains. More importantly, please open a new bug report for each new bug/regression introduced by the 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please specifically note if the issue does or does not appear in the 2.6.26 kernel. Thanks again, we really appreicate your help and feedback.

[Kevin P (kevin-cybercolloids)]

I no longer have this problem after replacing the SATA data cable with a
better quality latch cable. Recently I rebuilt the computer with the
problem and used a cheap cable again, how quickly you forget - the
problem recurred. The computer is now running OK with a set of good
quality cables. The computer is in an office and is used every day, all
day so it is getting a good work out - and its stable.

Kevin.

Leann Ogasawara wrote:
> The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the
> upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would
> appreciate it if you could please test this newer 2.6.27 Ubuntu kernel.
> There are one of two ways you should be able to test:
>
> 1) If you are comfortable installing packages on your own, the linux-
> image-2.6.27-* package is currently available for you to install and
> test.
>
> --or--
>
> 2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer
> 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4.
> Please watch http://www.ubuntu.com/testing for Alpha5 to be announced.
> You should then be able to test via a LiveCD.
>
> Please let us know immediately if this newer 2.6.27 kernel resolves the
> bug reported here or if the issue remains. More importantly, please
> open a new bug report for each new bug/regression introduced by the
> 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please
> specifically note if the issue does or does not appear in the 2.6.26
> kernel. Thanks again, we really appreicate your help and feedback.
>
> ** Tags added: cft-2.6.27
>
>

[Adam (adam.russell)]

I am unable to contribute any further to this bug, as I am no longer using the hardware in question. I will be unsubscribing.

[Christopher Berner (cberner)]

I just started experiencing this bug after I upgraded from Hardy to Intrepid. Let me know if there is any information I can provide.

[Launchpad Janitor (janitor)]

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.

[sideshowmel (sideshowmellemel)]

I've read through this and countless other posts about this issue (which I am also experiencing). It truly does render the computer useless.

The only big thing that jumps out at me in this thread is that everyone is mentioning problems with their CD/DVD drives, and upgrading firmware, etc. So... I don't even HAVE an optical drive installed on the system, and this problem happens every time I boot.

I do have an Intel chipset, and I was thinking I'd try blacklisting the ata piix module, but currently I am remote (over ssh), and the computer takes several minutes to respond even to shell commands. So I'd venture to say this isn't only a problem with optical drives. All I have is 3 internal SATA drives. No external USB, no optical, no floppy. I also tried stopping the HAL daemon, and initially it seemed JUST A LITTLE better, although the messages still show up in the syslog. I don't know if I should try to blacklist ata piix while I'm remote, as I've noticed in reading through posts that it might render the machine unbootable.

Anything else I can try? This is running Intrepid with all recent updates, 3 internal SATA drives, two of which are using mdadm for 3 RAID1 partitions and 1 RAID0 partition. And I've done countless diagnostics and all hardware appears normal. Also, I've removed the third (non-RAID) drive and the problem still happens. errors:

Jan 5 10:10:15 kaya kernel: [ 6132.945046] ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan 5 10:10:15 kaya kernel: [ 6132.945062] ata3.00: cmd c8/00:08:61:36:db/00:00:00:00:00/e5 tag 0 dma 4096 in
Jan 5 10:10:15 kaya kernel: [ 6132.945066] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Jan 5 10:10:15 kaya kernel: [ 6132.945074] ata3.00: status: { DRDY }
Jan 5 10:10:15 kaya kernel: [ 6132.945090] ata3: soft resetting link
Jan 5 10:10:15 kaya kernel: [ 6133.181465] ata3.00: configured for UDMA/100
Jan 5 10:10:15 kaya kernel: [ 6133.181487] ata3: EH complete
Jan 5 10:10:15 kaya kernel: [ 6133.194973] sd 2:0:0:0: [sda] 234441648 512-byte hardware sectors (120034 MB)
Jan 5 10:10:15 kaya kernel: [ 6133.195229] sd 2:0:0:0: [sda] Write Protect is off
Jan 5 10:10:15 kaya kernel: [ 6133.195236] sd 2:0:0:0: [sda] Mode Sense: 00 3a 00 00
Jan 5 10:10:15 kaya kernel: [ 6133.221708] sd 2:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Jan 5 10:10:46 kaya kernel: [ 6163.908049] ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan 5 10:10:46 kaya kernel: [ 6163.908063] ata3.00: cmd c8/00:20:89:30:db/00:00:00:00:00/e5 tag 0 dma 16384 in
Jan 5 10:10:46 kaya kernel: [ 6163.908065] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Jan 5 10:10:46 kaya kernel: [ 6163.908071] ata3.00: status: { DRDY }
Jan 5 10:10:46 kaya kernel: [ 6163.908083] ata3: soft resetting link
Jan 5 10:10:46 kaya kernel: [ 6164.144466] ata3.00: configured for UDMA/100
Jan 5 10:10:46 kaya kernel: [ 6164.144487] ata3: EH complete
Jan 5 10:10:46 kaya kernel: [ 6164.161017] sd 2:0:0:0: [sda] 234441648 512-byte hardware sectors (120034 MB)
Jan 5 10:10:46 kaya kernel: [ 6164.161303] sd 2:0:0:0: [sda] Write Pr...

[sideshowmel (sideshowmellemel)]

So it would appear that this is, in fact a hardware issue (at least in my case). Like I said, my mem, mobo, proc, and HDD's all were good. The one thing I never considered: power supply.

It would appear that one of my drives was consistently receiving too little power. It just occured to me out of the blue when I was sitting there dealing with this problem and heard what I thought were my fans spinning down, then back up again, within 2 seconds.

I had been using a splitter that splits one Molex 4-pin adapter into 2 SATA power adapter. I changed the wire configurations around so that each 4-pin Molex was only allocated one HDD per. Since then I show 2 days of system uptime with no recurrence of this problem.

I've since ordered a new power supply, as this one is most certainly about to die.

Thanks for the responses, everyone. Sorry to waste time and effort!

[Bryan Wu (cooloney)]

As sideshowmel reported, this bug is invalidate at all. So close it to invalid.

-Bryan

[kamahat (kamahat)]

Same probleme and I had to wait a long time to find a solution : flashing the firmware of my latop cdrom.

I've got an ACER Aspire 9410, the cdrom is a TSSTCorp TS-L632D
And acer only deliver 1 firmware : AC01

As stated on the other post, crossing flashing is okay, I've flashed witch :" SC04 - Original Samsung Computer Firmware "
And all my probleme goes away

a source to find firmwares : http://backfire.rpc1.org/tsstcorp/index.php?path=TS-L632D/

PS : to crossflash I've done it under windows with the binary suplied "sfdnwin " and the option "-nocheck"
some information here also : http://forum.rpc1.org/viewtopic.php?p=37412#37417