udev 60-persistent-storage.rules vol_id cancels hdd spin-down

Bug #388506 reported by phoenix
16
This bug affects 2 people
Affects Status Importance Assigned to Milestone
hdparm (Ubuntu)
Fix Released
Undecided
Unassigned
udev (Ubuntu)
Invalid
Medium
Unassigned

Bug Description

Binary package hint: udev

My secondary hdd (not mounted) did not go into standby and spun up again right after a "hdparm -y /dev/sdb" command.

the cause is /lib/udev/vol_id beeing executed on a change action caused by this rule in /lib/udev/rules.d/60-persistent-storage.rules:

# probe filesystem metadata of disks
KERNEL!="sr*", IMPORT{program}="vol_id --export $tempnode"

which causes an access to the hdd _after_ a spin-down event -> hdd spins up again
(My hdd model: SAMSUNG HD103UJ (Spinpoint F1, 1TB))

quick fix for me:
KERNEL!="sr*", ACTION=="add", IMPORT{program}="vol_id --export $tempnode"

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

This is not a duplicate, but is a known bug caused by hdparm opening the block device for writing; it is not clear that it needs to do so.

Changed in udev (Ubuntu):
importance: Undecided → Medium
status: New → Triaged
Revision history for this message
jpkotta (jpkotta) wrote :

Indeed, hdparm does not need to open the file RW. I have only tested RO hdparm with -t, -T, -y, -Y, -i, -I, and -C, and not very thoroughly.

Revision history for this message
jpkotta (jpkotta) wrote :

hdparm-9.15 fixes this issue (it opens the file RO).

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

Great, looks like 9.15 is in karmic already

Changed in udev (Ubuntu):
status: Triaged → Invalid
Changed in hdparm (Ubuntu):
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.