noflushd tries to spindown USB drive
Status: Inactive
Brought to you by:
nold
Menu
▾
▴
#16 noflushd tries to spindown USB drive
If noflushd is started while a USB drive is present, it
will try to spin down the USB drive (/dev/sda) instead
of /dev/hda.
On these occations when it is started
(/etc/init.d/noflushd start) it says:
Starting No Flush Daemon: pdflush wakeup interval 0
suspiciously low, overriding to 5 seconds
And when it comes time to spin down /dev/hda it doesn't
and xconsole says:
Jun 29 17:07:10 capra noflushd[5926]: Spinning down
/dev/sda.
Strange. My noflushd file has: PARAMS="-v -n 33,1
/dev/hda" No mention on /dev/sda. This happens for both
USB flash drives, and USB hard drive enclosures.
Thanks.
Logged In: YES
user_id=7832
Thanks for the report. Can you please provide additional
information to aid my debugging:
* The versions of noflushd and the kernel you're running.
* The output of "noflushd -d -v -n 1 /dev/hda" with a USB
drive connected.
* The contents of /proc/partitions and /proc/diskstat (or
/proc/stat on 2.4 kernels).
* If this is on a 2.6 kernel, are you using udev?
Logged In: YES
user_id=1304594
I'm using a 2.6.10 Kernel compiled from source using make-kpkg:
capra:/# uname -a
Linux capra 2.6.10-kinetic #1 Fri Mar 25 07:40:21 EST 2005
i686 GNU/Linux
Running an up to date Testing (Etch) distribution of Debian.
capra:/# noflushd -V
noflushd - idle disk daemon. Version 2.7.5.
Error: Don't know how to spindown /dev/sda
Error: No valid timeout found for /dev/hda
~~~~~~~~
capra:/# /etc/init.d/noflushd stop
Stopping No Flush Daemon: noflushd.
Starting directly from command line like this seems to work
as expected:
capra:/# noflushd -d -v -n 1 /dev/hda
Kernel talks oldstyle dev
Probing for valid blkdev at /dev/hda
Detected oldstyle dev at /dev/hda
Kernel talks oldstyle dev
Added entry for /dev/hda (3, 0)
Added entry for /dev/sda (8, 0)
Added disk /dev/hda, current timeout 60, state 2
Controlling pdflush daemon
Kernel talks linux 2.5-style stat
pdflush stopped
Check interval 5
Allocated stat entry for (3, 0)
Allocated stat entry for (22, 0)
Allocated stat entry for (7, 0)
Allocated stat entry for (7, 1)
Allocated stat entry for (7, 2)
Allocated stat entry for (7, 3)
Allocated stat entry for (7, 4)
Allocated stat entry for (7, 5)
Allocated stat entry for (7, 6)
Allocated stat entry for (7, 6)
Allocated stat entry for (7, 7)
Allocated stat entry for (8, 0)
rw disks: 0 stopped, 2 spinning
pdflush restarted with interval 500.
State 1, sleeping 1 seconds
pdflush stopped
Check interval 6
Disk /dev/hda, Time left 54.
[SNIP]
Disk /dev/hda, Time left 0.
Spinning down /dev/hda.
Syncing /dev/hda
Syncing /dev/hda1
Syncing /dev/hda2
Syncing /dev/hda3
Syncing /dev/hda4
Syncing /dev/hda5
Syncing /dev/hda6
Syncing /dev/loop0
Syncing /dev/hda
Syncing /dev/hda1
Syncing /dev/hda2
Syncing /dev/hda3
Syncing /dev/hda4
Syncing /dev/hda5
Syncing /dev/hda6
Syncing /dev/loop0
Sending spindown command to /dev/hda.
Disk was up and running
rw disks: 1 stopped, 1 spinning
Syncing /dev/sda
Syncing /dev/sda1
State 2, sleeping 1 seconds
Check interval 2
rw disks: 1 stopped, 1 spinning
State 2, sleeping 1 seconds
Check interval 1
rw disks: 1 stopped, 1 spinning
State 2, sleeping 1 seconds
Check interval 1
rw disks: 1 stopped, 1 spinning
State 2, sleeping 1 seconds
Check interval 1
rw disks: 1 stopped, 1 spinning
~~~~~~~~~~~
That seems to work properly! Hmm. Yet starting via
/etc/init.d/noflushd script causes the bug.
So I tried running my machines command with -d:
capra:/# /etc/init.d/noflushd stop
Stopping No Flush Daemon: noflushd.
capra:/# noflushd -d -v -n 33,1 /dev/hda
Kernel talks oldstyle dev
Probing for valid blkdev at /dev/hda
Detected oldstyle dev at /dev/hda
Kernel talks oldstyle dev
Added entry for /dev/hda (3, 0)
Added entry for /dev/sda (8, 0)
Added disk /dev/hda, current timeout 1980, state 2
pdflush wakeup interval 0 suspiciously low, overriding to 5
seconds
Controlling pdflush daemon
Kernel talks linux 2.5-style stat
pdflush stopped
Check interval 5
Allocated stat entry for (3, 0)
Allocated stat entry for (22, 0)
Allocated stat entry for (7, 0)
Allocated stat entry for (7, 1)
Allocated stat entry for (7, 2)
Allocated stat entry for (7, 3)
Allocated stat entry for (7, 4)
Allocated stat entry for (7, 5)
Allocated stat entry for (7, 6)
Allocated stat entry for (7, 7)
Allocated stat entry for (8, 0)
rw disks: 0 stopped, 2 spinning
pdflush restarted with interval 500.
State 1, sleeping 33 seconds
[ISSUED sudo /usr/bin/killall -s HUP noflushd FROM ICEWM MENU]
Switching to next timeout set.
Timeout update. Disk /dev/sda, timeout 60, left -1208102648
Timeout update. Disk /dev/hda, timeout 60, left 60
pdflush stopped
Check interval 12
Disk /dev/sda, Time left -1208102648.
Spinning down /dev/sda.
Syncing /dev/sda
Syncing /dev/sda1
Syncing /dev/sda
Syncing /dev/sda1
Sending spindown command to /dev/sda.
Segmentation fault
capra:/# cat /proc/partitions
major minor #blocks name
3 0 19535040 hda
3 1 12892131 hda1
3 2 1951897 hda2
3 3 1 hda3
3 4 3389715 hda4
3 5 795186 hda5
3 6 506016 hda6
7 0 1536000 loop0
8 0 128000 sda
8 1 127984 sda1
capra:/# cat /proc/diskstats
3 0 hda 8196 818 411919 276580 5910 10285 125620
886140 0 181080 1162990
3 1 hda1 235 235 8395 8395
3 2 hda2 152 152 9 9
3 3 hda3 0 0 0 0
3 4 hda4 7393 385114 3590 29024
3 5 hda5 994 24466 3685 83736
3 6 hda6 224 1792 557 4456
22 0 hdc 0 0 0 0 0 0 0 0 0 0 0
7 0 loop0 0 0 0 0 0 0 0 0 0 0 0
7 1 loop1 0 0 0 0 0 0 0 0 0 0 0
7 2 loop2 0 0 0 0 0 0 0 0 0 0 0
7 3 loop3 0 0 0 0 0 0 0 0 0 0 0
7 4 loop4 0 0 0 0 0 0 0 0 0 0 0
7 5 loop5 0 0 0 0 0 0 0 0 0 0 0
7 6 loop6 0 0 0 0 0 0 0 0 0 0 0
7 7 loop7 0 0 0 0 0 0 0 0 0 0 0
8 0 sda 1 0 8 10 0 0 0 0 0 10 10
8 1 sda1 0 0 0 0
It seems using two timeouts triggers the bug.
I don't think i am using udev. Been compiling my own kernels
since 2.4 and remember not selecting the new dev system when
going to 2.6. How can I check?
Thanks.