Hard disk I/O randomly freezes when hald is running and optical drive is empty
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Linux |
Invalid
|
High
|
|||
linux (Ubuntu) |
Invalid
|
Undecided
|
Unassigned | ||
linux-source-2.6.20 (BOSS) |
Invalid
|
Undecided
|
Unassigned | ||
linux-source-2.6.20 (Ubuntu) |
Won't Fix
|
Medium
|
Unassigned | ||
linux-source-2.6.22 (Ubuntu) |
Won't Fix
|
Medium
|
Unassigned |
Bug Description
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:
[ 188.960000] res 40/00:03:
[ 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
Changed in linux: | |
status: | Unknown → Rejected |
Changed in linux: | |
status: | Unknown → Confirmed |
Changed in linux: | |
status: | Confirmed → In Progress |
Changed in linux: | |
status: | In Progress → Incomplete |
Changed in linux-source-2.6.20: | |
status: | Confirmed → Triaged |
Changed in linux: | |
status: | Incomplete → Confirmed |
Changed in linux-source-2.6.22: | |
assignee: | nobody → ubuntu-kernel-team |
importance: | Undecided → Medium |
status: | New → Triaged |
Changed in linux-source-2.6.20: | |
status: | Triaged → Won't Fix |
Changed in linux-source-2.6.22: | |
status: | Triaged → Won't Fix |
Changed in linux-source-2.6.20: | |
assignee: | nobody → phillip-lougher |
status: | New → Invalid |
Changed in linux: | |
assignee: | nobody → rifi58 |
Changed in linux: | |
status: | Incomplete → Confirmed |
Changed in linux: | |
status: | Confirmed → Invalid |
Changed in linux: | |
importance: | Unknown → High |
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.