Disk access every 5 seconds prevents spindown.

[jinnan_tonnix]

Hi all,

I'm having trouble getting my drive to spin down. It seems it's being
accessed every 5 seconds by the journalling system.

I've used Andreas's script and the 'noatime' script to remount var and
dev in memory rather than disk, and to set the 'noatime' flag on the disk.

Could someone check their system to see if they get the same result?
Thanks. Please post back!

'cat /proc/mounts' shows:
# cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/root /initrd jffs2 rw 0 0
/dev/sdb1 / ext3 rw,noatime 0 0
/proc /proc proc rw 0 0
usbdevfs /proc/bus/usb usbdevfs rw 0 0
/dev/sdb1 /dev.state ext3 rw,noatime 0 0
ramfs /dev ramfs rw 0 0
/dev/sdb1 /var.state ext3 rw,noatime 0 0
ramfs /var ramfs rw 0 0
/dev/sdb1 /share/hdd/data ext3 rw,noatime 0 0
/dev/sdb2 /share/hdd/conf ext3 rw,noatime 0 0


'Mount' shows
# mount
/dev/sdb1 on /initrd type jffs2 (rw)
/dev/sdb1 on / type ext3 (rw,noatime)
/proc on /proc type proc (rw)
usbdevfs on /proc/bus/usb type usbdevfs (rw)
/dev/sdb1 on /dev.state type ext3 (rw,noatime)
ramfs on /dev type ramfs (rw)
/dev/sdb1 on /var.state type ext3 (rw,noatime)
ramfs on /var type ramfs (rw)
/dev/sdb1 on /share/hdd/data type ext3 (rw,noatime)
/dev/sdb2 on /share/hdd/conf type ext3 (rw,noatime)






 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/nslu2-general/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/nslu2-general/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:nslu2-general-digest@...
    mailto:nslu2-general-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
    nslu2-general-unsubscribe@...

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 

[Mat-24]

I can confirm that doing what you're doing (using Andreas's script and
the noatime script) on an unslung 6.8 slug, and getting same results as
you for the cat /proc/mounts and the mount commands ('cept I'm sda not
sdb), something is accessing the disk every 5 or so seconds.

I've got a western digital mybook, with the circular disk access lights
on, so it's easy to see when it's being accessed. I am running quite a
bit on my slug at the moment though, and my next round of testing is to
disable everything and start up the services one by one.

jinnan_tonnix wrote:

> Hi all,
>
> I'm having trouble getting my drive to spin down. It seems it's being
> accessed every 5 seconds by the journalling system.
>
> I've used Andreas's script and the 'noatime' script to remount var and
> dev in memory rather than disk, and to set the 'noatime' flag on the disk.
>
> Could someone check their system to see if they get the same result?
> Thanks. Please post back!
>
> 'cat /proc/mounts' shows:
> # cat /proc/mounts
> rootfs / rootfs rw 0 0
> /dev/root /initrd jffs2 rw 0 0
> /dev/sdb1 / ext3 rw,noatime 0 0
> /proc /proc proc rw 0 0
> usbdevfs /proc/bus/usb usbdevfs rw 0 0
> /dev/sdb1 /dev.state ext3 rw,noatime 0 0
> ramfs /dev ramfs rw 0 0
> /dev/sdb1 /var.state ext3 rw,noatime 0 0
> ramfs /var ramfs rw 0 0
> /dev/sdb1 /share/hdd/data ext3 rw,noatime 0 0
> /dev/sdb2 /share/hdd/conf ext3 rw,noatime 0 0
>
>
> 'Mount' shows
> # mount
> /dev/sdb1 on /initrd type jffs2 (rw)
> /dev/sdb1 on / type ext3 (rw,noatime)
> /proc on /proc type proc (rw)
> usbdevfs on /proc/bus/usb type usbdevfs (rw)
> /dev/sdb1 on /dev.state type ext3 (rw,noatime)
> ramfs on /dev type ramfs (rw)
> /dev/sdb1 on /var.state type ext3 (rw,noatime)
> ramfs on /var type ramfs (rw)
> /dev/sdb1 on /share/hdd/data type ext3 (rw,noatime)
> /dev/sdb2 on /share/hdd/conf type ext3 (rw,noatime)
>
>

[jinnan_tonnix]

Thanks for that info Mat. I'm running Dropbear and Twonkymusic server on
my Slug, so perhaps these are writing 'something'. I'm in the same boat
as you, so it's back to starting with basic services and checking what's
writing to disk.

By the way, don't the lines
/dev/sdb1 on /dev.state type ext3 (rw,noatime)
/dev/sdb1 on /var.state type ext3 (rw,noatime)
surely means that dev.state and var.state is mounted to the disk, rather
than RAM? If so, then the kjournald process will write to disk rather
than RAM. Or have I misunderstood what's going on? (<-- quite likely)

  [:-?]

Paul.


--- In nslu2-general@..., Mat <mat@...> wrote:
>
> I can confirm that doing what you're doing (using Andreas's script and
> the noatime script) on an unslung 6.8 slug, and getting same results
as
> you for the cat /proc/mounts and the mount commands ('cept I'm sda not
> sdb), something is accessing the disk every 5 or so seconds.
>
> I've got a western digital mybook, with the circular disk access
lights
> on, so it's easy to see when it's being accessed. I am running quite a
> bit on my slug at the moment though, and my next round of testing is
to
> disable everything and start up the services one by one.
>
> jinnan_tonnix wrote:
> > Hi all,
> >
> > I'm having trouble getting my drive to spin down. It seems it's
being
> > accessed every 5 seconds by the journalling system.
> >
> > I've used Andreas's script and the 'noatime' script to remount var
and
> > dev in memory rather than disk, and to set the 'noatime' flag on the
disk.

> >
> > Could someone check their system to see if they get the same result?
> > Thanks. Please post back!
> >
> > 'cat /proc/mounts' shows:
> > # cat /proc/mounts
> > rootfs / rootfs rw 0 0
> > /dev/root /initrd jffs2 rw 0 0
> > /dev/sdb1 / ext3 rw,noatime 0 0
> > /proc /proc proc rw 0 0
> > usbdevfs /proc/bus/usb usbdevfs rw 0 0
> > /dev/sdb1 /dev.state ext3 rw,noatime 0 0
> > ramfs /dev ramfs rw 0 0
> > /dev/sdb1 /var.state ext3 rw,noatime 0 0
> > ramfs /var ramfs rw 0 0
> > /dev/sdb1 /share/hdd/data ext3 rw,noatime 0 0
> > /dev/sdb2 /share/hdd/conf ext3 rw,noatime 0 0
> >
> >
> > 'Mount' shows
> > # mount
> > /dev/sdb1 on /initrd type jffs2 (rw)
> > /dev/sdb1 on / type ext3 (rw,noatime)
> > /proc on /proc type proc (rw)
> > usbdevfs on /proc/bus/usb type usbdevfs (rw)
> > /dev/sdb1 on /dev.state type ext3 (rw,noatime)
> > ramfs on /dev type ramfs (rw)
> > /dev/sdb1 on /var.state type ext3 (rw,noatime)
> > ramfs on /var type ramfs (rw)
> > /dev/sdb1 on /share/hdd/data type ext3 (rw,noatime)
> > /dev/sdb2 on /share/hdd/conf type ext3 (rw,noatime)
> >
> >
>

[Non-text portions of this message have been removed]

[Mat-24]

A further update in the quest for drive spindown on an unslung 6.8 slug.

I found an option in Twonkyvision to suspend the media servers disk
access a certain number of minutes after a client last accesses. No
joy on this one.

I then did a "cd /opt/etc/init.d" followed by "chmod -x *" so noen of
my diversion scripts run at boot (ie apache, asterisk, twonky, mtdaapd
etc". (though I did do a "chmod +x S40sshd" so I can get back in).

I've turned off upnp, then killed off as many processes as possible,
leaving me with the following:

# ps -ef
  PID TTY     Uid        Size State Command
    1         root       1212   S   /bin/init
    2         root          0   S   [keventd]
    3         root          0   S   [ksoftirqd_CPU0]
    4         root          0   S   [kswapd]
    5         root          0   S   [bdflush]
    6         root          0   S   [kupdated]
    7         root          0   S   [mtdblockd]
    8         root          0   S   [khubd]
    9         root          0   S   [jffs2_gcd_mtd4]
   10         root          0   S   [usb-storage-0]
   11         root          0   S   [scsi_eh_0]
   16         root          0   S   [sd-mc-thread]
   27         root          0   S   [kjournald]
   50         root          0   D   [ixp425_csr]
   51         root          0   S   [ixp425 ixp0]
   54 ttyS0   root       1916   S   /bin/sh
  165         root          0   S   [kjournald]
  441         root       3208   S   /opt/sbin/sshd
  464         root       5812   S   sshd: root@ttyp0
  466 ttyp0   root       1932   S   -sh
  662         root       1924   S   /sbin/klogd -n
  667         root       1936   S   /sbin/syslogd -n
  695 ttyp0   root       1984   R   ps -ef

So not much there. I'm starting to think that spindown on unslung is
an impossible dream - I even tried using a hex editor on the
/sbin/rc.bootbin to change ext3 to ext2, and it still mounted as ext3
on boot. trying it manually (umount /dev/sda1 followed by "mount -t
ext2 /dev/sda1 /share/flash/data" gives me an error "device or
resource busy")

I'm thinking I may well be looking at openslug this weekend...

--- In nslu2-general@..., "jinnan_tonnix"
<jinnan_tonnix@...> wrote:

>
>
> Thanks for that info Mat. I'm running Dropbear and Twonkymusic server on
> my Slug, so perhaps these are writing 'something'. I'm in the same boat
> as you, so it's back to starting with basic services and checking what's
> writing to disk.
>
> By the way, don't the lines
> /dev/sdb1 on /dev.state type ext3 (rw,noatime)
> /dev/sdb1 on /var.state type ext3 (rw,noatime)
> surely means that dev.state and var.state is mounted to the disk, rather
> than RAM? If so, then the kjournald process will write to disk rather
> than RAM. Or have I misunderstood what's going on? (<-- quite likely)
>
>   [:-?]
>
> Paul.
>
>
> --- In nslu2-general@..., Mat <mat@> wrote:
> >
> > I can confirm that doing what you're doing (using Andreas's script and
> > the noatime script) on an unslung 6.8 slug, and getting same results
> as
> > you for the cat /proc/mounts and the mount commands ('cept I'm sda not
> > sdb), something is accessing the disk every 5 or so seconds.
> >
> > I've got a western digital mybook, with the circular disk access
> lights
> > on, so it's easy to see when it's being accessed. I am running quite a
> > bit on my slug at the moment though, and my next round of testing is
> to
> > disable everything and start up the services one by one.
> >
> > jinnan_tonnix wrote:
> > > Hi all,
> > >
> > > I'm having trouble getting my drive to spin down. It seems it's
> being
> > > accessed every 5 seconds by the journalling system.
> > >
> > > I've used Andreas's script and the 'noatime' script to remount var
> and
> > > dev in memory rather than disk, and to set the 'noatime' flag on the
> disk.
> > >
> > > Could someone check their system to see if they get the same result?
> > > Thanks. Please post back!
> > >
> > > 'cat /proc/mounts' shows:
> > > # cat /proc/mounts
> > > rootfs / rootfs rw 0 0
> > > /dev/root /initrd jffs2 rw 0 0
> > > /dev/sdb1 / ext3 rw,noatime 0 0
> > > /proc /proc proc rw 0 0
> > > usbdevfs /proc/bus/usb usbdevfs rw 0 0
> > > /dev/sdb1 /dev.state ext3 rw,noatime 0 0
> > > ramfs /dev ramfs rw 0 0
> > > /dev/sdb1 /var.state ext3 rw,noatime 0 0
> > > ramfs /var ramfs rw 0 0
> > > /dev/sdb1 /share/hdd/data ext3 rw,noatime 0 0
> > > /dev/sdb2 /share/hdd/conf ext3 rw,noatime 0 0
> > >
> > >
> > > 'Mount' shows
> > > # mount
> > > /dev/sdb1 on /initrd type jffs2 (rw)
> > > /dev/sdb1 on / type ext3 (rw,noatime)
> > > /proc on /proc type proc (rw)
> > > usbdevfs on /proc/bus/usb type usbdevfs (rw)
> > > /dev/sdb1 on /dev.state type ext3 (rw,noatime)
> > > ramfs on /dev type ramfs (rw)
> > > /dev/sdb1 on /var.state type ext3 (rw,noatime)
> > > ramfs on /var type ramfs (rw)
> > > /dev/sdb1 on /share/hdd/data type ext3 (rw,noatime)
> > > /dev/sdb2 on /share/hdd/conf type ext3 (rw,noatime)
> > >
> > >
> >
>
>
>
> [Non-text portions of this message have been removed]
>

[ptrsmid]

Hello Mat,

I use unslung 6.8 with a western digital harddisk.
I have open ssh, mt-daapd and twonkymedia installed and my slug does
spin down. However, the drive dus startup quite often. It is a bit
difficult to track but i get the impression de drive starts up about
every hour. Most of the time the drive feels kind of luke warm which
indicates recent activity. Only sometimes the drive feels realy cold.

It would be nice to have a utility to check when the drive is
accessed by what program but I understood from the wike that is not
available on unslung 6.8.

regards, peter

--- In nslu2-general@..., "Mat" <mat@...> wrote:
>
> A further update in the quest for drive spindown on an unslung 6.8
slug.
>
> I found an option in Twonkyvision to suspend the media servers disk
> access a certain number of minutes after a client last accesses. No
> joy on this one.
>
> I then did a "cd /opt/etc/init.d" followed by "chmod -x *" so noen
of
> my diversion scripts run at boot (ie apache, asterisk, twonky,
mtdaapd
> etc". (though I did do a "chmod +x S40sshd" so I can get back in).
>
> I've turned off upnp, then killed off as many processes as
possible,

> leaving me with the following:
>
> # ps -ef
>   PID TTY     Uid        Size State Command
>     1         root       1212   S   /bin/init
>     2         root          0   S   [keventd]
>     3         root          0   S   [ksoftirqd_CPU0]
>     4         root          0   S   [kswapd]
>     5         root          0   S   [bdflush]
>     6         root          0   S   [kupdated]
>     7         root          0   S   [mtdblockd]
>     8         root          0   S   [khubd]
>     9         root          0   S   [jffs2_gcd_mtd4]
>    10         root          0   S   [usb-storage-0]
>    11         root          0   S   [scsi_eh_0]
>    16         root          0   S   [sd-mc-thread]
>    27         root          0   S   [kjournald]
>    50         root          0   D   [ixp425_csr]
>    51         root          0   S   [ixp425 ixp0]
>    54 ttyS0   root       1916   S   /bin/sh
>   165         root          0   S   [kjournald]
>   441         root       3208   S   /opt/sbin/sshd
>   464         root       5812   S   sshd: root@ttyp0
>   466 ttyp0   root       1932   S   -sh
>   662         root       1924   S   /sbin/klogd -n
>   667         root       1936   S   /sbin/syslogd -n
>   695 ttyp0   root       1984   R   ps -ef
>
> So not much there. I'm starting to think that spindown on unslung

is
> an impossible dream - I even tried using a hex editor on the
> /sbin/rc.bootbin to change ext3 to ext2, and it still mounted as
ext3
> on boot. trying it manually (umount /dev/sda1 followed by "mount -t
> ext2 /dev/sda1 /share/flash/data" gives me an error "device or
> resource busy")
>
> I'm thinking I may well be looking at openslug this weekend...
>

[bullfrog528]

Is it possible that you've a swap file/partition on this drive? Hence
the activity.

regards

Jeremy (Bullfrog)

--- In nslu2-general@..., "ptrsmid" <peter@...> wrote:
>
> Hello Mat,
>
> I use unslung 6.8 with a western digital harddisk.
> I have open ssh, mt-daapd and twonkymedia installed and my slug
does
> spin down. However, the drive dus startup quite often. It is a bit
> difficult to track but i get the impression de drive starts up
about
> every hour. Most of the time the drive feels kind of luke warm
which
> indicates recent activity. Only sometimes the drive feels realy
cold.

>
> It would be nice to have a utility to check when the drive is
> accessed by what program but I understood from the wike that is not
> available on unslung 6.8.
>
> regards, peter
>
> --- In nslu2-general@..., "Mat" <mat@> wrote:
> >
> > A further update in the quest for drive spindown on an unslung

6.8
> slug.
> >
> > I found an option in Twonkyvision to suspend the media servers
disk
> > access a certain number of minutes after a client last accesses.
No
> > joy on this one.
> >
> > I then did a "cd /opt/etc/init.d" followed by "chmod -x *" so
noen

> of
> > my diversion scripts run at boot (ie apache, asterisk, twonky,
> mtdaapd
> > etc". (though I did do a "chmod +x S40sshd" so I can get back in).
> >
> > I've turned off upnp, then killed off as many processes as
> possible,
> > leaving me with the following:
> >
> > # ps -ef
> >   PID TTY     Uid        Size State Command
> >     1         root       1212   S   /bin/init
> >     2         root          0   S   [keventd]
> >     3         root          0   S   [ksoftirqd_CPU0]
> >     4         root          0   S   [kswapd]
> >     5         root          0   S   [bdflush]
> >     6         root          0   S   [kupdated]
> >     7         root          0   S   [mtdblockd]
> >     8         root          0   S   [khubd]
> >     9         root          0   S   [jffs2_gcd_mtd4]
> >    10         root          0   S   [usb-storage-0]
> >    11         root          0   S   [scsi_eh_0]
> >    16         root          0   S   [sd-mc-thread]
> >    27         root          0   S   [kjournald]
> >    50         root          0   D   [ixp425_csr]
> >    51         root          0   S   [ixp425 ixp0]
> >    54 ttyS0   root       1916   S   /bin/sh
> >   165         root          0   S   [kjournald]
> >   441         root       3208   S   /opt/sbin/sshd
> >   464         root       5812   S   sshd: root@ttyp0
> >   466 ttyp0   root       1932   S   -sh
> >   662         root       1924   S   /sbin/klogd -n
> >   667         root       1936   S   /sbin/syslogd -n
> >   695 ttyp0   root       1984   R   ps -ef
> >
> > So not much there. I'm starting to think that spindown on unslung
> is
> > an impossible dream - I even tried using a hex editor on the
> > /sbin/rc.bootbin to change ext3 to ext2, and it still mounted as
> ext3
> > on boot. trying it manually (umount /dev/sda1 followed by "mount -

t
> > ext2 /dev/sda1 /share/flash/data" gives me an error "device or
> > resource busy")
> >
> > I'm thinking I may well be looking at openslug this weekend...
> >
>

[bullfrog528]

As Far as I can tell
cat /proc/swaps will tell you which devices are being used for swap
space
cat /proc/meminfo will list size of swap space in use amongst other
things.
swapoff -a will turn off all swap devices/space
swapon -a will turn on all swap devices/space

I believe your probably suffering from the same reason why a memory
stick is accessed every few seconds if you haven't used the flag to
tell the system not to use the memory stick as swap when you've
unslung to a memory stick.

Jeremy (Bullfrog)

--- In nslu2-general@..., "bullfrog528" <yahoo@...> wrote:
>
> Is it possible that you've a swap file/partition on this drive?
Hence

> the activity.
>
> regards
>
> Jeremy (Bullfrog)
>
> --- In nslu2-general@..., "ptrsmid" <peter@> wrote:
> >
> > Hello Mat,
> >
> > I use unslung 6.8 with a western digital harddisk.
> > I have open ssh, mt-daapd and twonkymedia installed and my slug
> does
> > spin down. However, the drive dus startup quite often. It is a

bit
> > difficult to track but i get the impression de drive starts up
> about
> > every hour. Most of the time the drive feels kind of luke warm
> which
> > indicates recent activity. Only sometimes the drive feels realy
> cold.
> >
> > It would be nice to have a utility to check when the drive is
> > accessed by what program but I understood from the wike that is
not

> > available on unslung 6.8.
> >
> > regards, peter
> >
> > --- In nslu2-general@..., "Mat" <mat@> wrote:
> > >
> > > A further update in the quest for drive spindown on an unslung
> 6.8
> > slug.
> > >
> > > I found an option in Twonkyvision to suspend the media servers
> disk
> > > access a certain number of minutes after a client last

accesses.
> No
> > > joy on this one.
> > >
> > > I then did a "cd /opt/etc/init.d" followed by "chmod -x *" so
> noen
> > of
> > > my diversion scripts run at boot (ie apache, asterisk, twonky,
> > mtdaapd
> > > etc". (though I did do a "chmod +x S40sshd" so I can get back
in).

> > >
> > > I've turned off upnp, then killed off as many processes as
> > possible,
> > > leaving me with the following:
> > >
> > > # ps -ef
> > >   PID TTY     Uid        Size State Command
> > >     1         root       1212   S   /bin/init
> > >     2         root          0   S   [keventd]
> > >     3         root          0   S   [ksoftirqd_CPU0]
> > >     4         root          0   S   [kswapd]
> > >     5         root          0   S   [bdflush]
> > >     6         root          0   S   [kupdated]
> > >     7         root          0   S   [mtdblockd]
> > >     8         root          0   S   [khubd]
> > >     9         root          0   S   [jffs2_gcd_mtd4]
> > >    10         root          0   S   [usb-storage-0]
> > >    11         root          0   S   [scsi_eh_0]
> > >    16         root          0   S   [sd-mc-thread]
> > >    27         root          0   S   [kjournald]
> > >    50         root          0   D   [ixp425_csr]
> > >    51         root          0   S   [ixp425 ixp0]
> > >    54 ttyS0   root       1916   S   /bin/sh
> > >   165         root          0   S   [kjournald]
> > >   441         root       3208   S   /opt/sbin/sshd
> > >   464         root       5812   S   sshd: root@ttyp0
> > >   466 ttyp0   root       1932   S   -sh
> > >   662         root       1924   S   /sbin/klogd -n
> > >   667         root       1936   S   /sbin/syslogd -n
> > >   695 ttyp0   root       1984   R   ps -ef
> > >
> > > So not much there. I'm starting to think that spindown on

unslung
> > is
> > > an impossible dream - I even tried using a hex editor on the
> > > /sbin/rc.bootbin to change ext3 to ext2, and it still mounted
as
> > ext3
> > > on boot. trying it manually (umount /dev/sda1 followed
by "mount -
> t
> > > ext2 /dev/sda1 /share/flash/data" gives me an error "device or
> > > resource busy")
> > >
> > > I'm thinking I may well be looking at openslug this weekend...
> > >
> >
>

[jinnan_tonnix]

--- In nslu2-general@..., "bullfrog528" <yahoo@...> wrote:
>
> As Far as I can tell
> cat /proc/swaps will tell you which devices are being used for swap
> space
> cat /proc/meminfo will list size of swap space in use amongst other
> things.
> swapoff -a will turn off all swap devices/space
> swapon -a will turn on all swap devices/space
>

I'll have a look at this over the weekend.


I'm beginning to think that a spin-down is an impossible dream too!

I've set Twonkymusic player to sleep after 20 minutes and stopped its
automatic scans, so I don't think this has anything to do with things.
Furthermore, I, like Mat, have taken the x bit from the startup
directory so I can test it without Twonky and OpenSSH don't running.
Even when I do this, I still get the faint blink of the drive LED
every 5 seconds.



I think the problem is something to do with the journalling. The
kjournald process is set to write every 5 seconds, but what the hell
it's writing is beyond me. I too have tried to mount the drives as
ext2, but with no success.



Peter - you say your disks spin down. Would you mind posting the
output of 'cat /proc/mounts' and 'mount' for me please? That way I can
see if my disk mounts are similar to yours. I still have my suspicions
over the way dev.state and var.state appear in the mount table.


Thanks to all..

[Mat-24]

I did wonder about the swap thing. I totally forgot about it in
earlier testing, as it doesn't show up in cat /proc/mounts and mount.

I've stripped all the processes down again, and run a swapoff -a

I've now got


# ps -ef
  PID TTY     Uid        Size State Command
    1         root       1212   S   /bin/init
    2         root          0   S   [keventd]
    3         root          0   S   [ksoftirqd_CPU0]
    4         root          0   S   [kswapd]
    5         root          0   S   [bdflush]
    6         root          0   S   [kupdated]
    7         root          0   S   [mtdblockd]
    8         root          0   S   [khubd]
    9         root          0   S   [jffs2_gcd_mtd4]
   10         root          0   S   [usb-storage-0]
   11         root          0   S   [scsi_eh_0]
   16         root          0   S   [sd-mc-thread]
   27         root          0   S   [kjournald]
   50         root          0   D   [ixp425_csr]
   51         root          0   S   [ixp425 ixp0]
   54 ttyS0   root       1916   S   /bin/sh
  163         root          0   S   [kjournald]
  422         root       3208   S   /opt/sbin/sshd
  494         root       1936   S   /sbin/syslogd -n
  496         root       1924   S   /sbin/klogd -n
  500         root       5812   R   sshd: root@ttyp0
  502 ttyp0   root       1920   S   -sh
  505 ttyp0   root       1984   R   ps -ef

and cat /proc/swaps shows nothing mounted.

I'm starting to wonder if the WD mybook drive is one which just won't
spin down on the slug - it spins down when the slug is turned off, and
spins up when the slug is turned on, but it just sits there accessing
away, and spinning full time.

Back to the drawing board...


--- In nslu2-general@..., "bullfrog528" <yahoo@...> wrote:

>
> As Far as I can tell
> cat /proc/swaps will tell you which devices are being used for swap
> space
> cat /proc/meminfo will list size of swap space in use amongst other
> things.
> swapoff -a will turn off all swap devices/space
> swapon -a will turn on all swap devices/space
>
> I believe your probably suffering from the same reason why a memory
> stick is accessed every few seconds if you haven't used the flag to
> tell the system not to use the memory stick as swap when you've
> unslung to a memory stick.
>
> Jeremy (Bullfrog)
>
> --- In nslu2-general@..., "bullfrog528" <yahoo@> wrote:
> >
> > Is it possible that you've a swap file/partition on this drive?
> Hence
> > the activity.
> >
> > regards
> >
> > Jeremy (Bullfrog)
> >
> > --- In nslu2-general@..., "ptrsmid" <peter@> wrote:
> > >
> > > Hello Mat,
> > >
> > > I use unslung 6.8 with a western digital harddisk.
> > > I have open ssh, mt-daapd and twonkymedia installed and my slug
> > does
> > > spin down. However, the drive dus startup quite often. It is a
> bit
> > > difficult to track but i get the impression de drive starts up
> > about
> > > every hour. Most of the time the drive feels kind of luke warm
> > which
> > > indicates recent activity. Only sometimes the drive feels realy
> > cold.
> > >
> > > It would be nice to have a utility to check when the drive is
> > > accessed by what program but I understood from the wike that is
> not
> > > available on unslung 6.8.
> > >
> > > regards, peter
> > >
> > > --- In nslu2-general@..., "Mat" <mat@> wrote:
> > > >
> > > > A further update in the quest for drive spindown on an unslung
> > 6.8
> > > slug.
> > > >
> > > > I found an option in Twonkyvision to suspend the media servers
> > disk
> > > > access a certain number of minutes after a client last
> accesses.
> > No
> > > > joy on this one.
> > > >
> > > > I then did a "cd /opt/etc/init.d" followed by "chmod -x *" so
> > noen
> > > of
> > > > my diversion scripts run at boot (ie apache, asterisk, twonky,
> > > mtdaapd
> > > > etc". (though I did do a "chmod +x S40sshd" so I can get back
> in).
> > > >
> > > > I've turned off upnp, then killed off as many processes as
> > > possible,
> > > > leaving me with the following:
> > > >
> > > > # ps -ef
> > > >   PID TTY     Uid        Size State Command
> > > >     1         root       1212   S   /bin/init
> > > >     2         root          0   S   [keventd]
> > > >     3         root          0   S   [ksoftirqd_CPU0]
> > > >     4         root          0   S   [kswapd]
> > > >     5         root          0   S   [bdflush]
> > > >     6         root          0   S   [kupdated]
> > > >     7         root          0   S   [mtdblockd]
> > > >     8         root          0   S   [khubd]
> > > >     9         root          0   S   [jffs2_gcd_mtd4]
> > > >    10         root          0   S   [usb-storage-0]
> > > >    11         root          0   S   [scsi_eh_0]
> > > >    16         root          0   S   [sd-mc-thread]
> > > >    27         root          0   S   [kjournald]
> > > >    50         root          0   D   [ixp425_csr]
> > > >    51         root          0   S   [ixp425 ixp0]
> > > >    54 ttyS0   root       1916   S   /bin/sh
> > > >   165         root          0   S   [kjournald]
> > > >   441         root       3208   S   /opt/sbin/sshd
> > > >   464         root       5812   S   sshd: root@ttyp0
> > > >   466 ttyp0   root       1932   S   -sh
> > > >   662         root       1924   S   /sbin/klogd -n
> > > >   667         root       1936   S   /sbin/syslogd -n
> > > >   695 ttyp0   root       1984   R   ps -ef
> > > >
> > > > So not much there. I'm starting to think that spindown on
> unslung
> > > is
> > > > an impossible dream - I even tried using a hex editor on the
> > > > /sbin/rc.bootbin to change ext3 to ext2, and it still mounted
> as
> > > ext3
> > > > on boot. trying it manually (umount /dev/sda1 followed
> by "mount -
> > t
> > > > ext2 /dev/sda1 /share/flash/data" gives me an error "device or
> > > > resource busy")
> > > >
> > > > I'm thinking I may well be looking at openslug this weekend...
> > > >
> > >
> >
>

[jinnan_tonnix]

--- In nslu2-general@..., "Mat" <mat@...> wrote:
>
> I did wonder about the swap thing. I totally forgot about it in
> earlier testing, as it doesn't show up in cat /proc/mounts and mount.
>

I'll try turning the swap off tonight, but I think I'll end up with
the same result as you (i.e. waste of time)

> I'm starting to wonder if the WD mybook drive is one which just won't
> spin down on the slug - it spins down when the slug is turned off, and
> spins up when the slug is turned on, but it just sits there accessing
> away, and spinning full time.
>

I've got a Buffalo 300GB drive. It's got some funky power management
features; I've set it to switch off when the Slug's off and to spin
down after 10mins of inactivity. I know it works as I've tried it on a
Windows machine. The problem is that it's never inactive!!! Every 5
seconds - blip! - little red disk light. I'm starting to take it
personally!

I hope Peter reads this and lists his disk mounts; if his shows the
same (and his disk spins down) then I'm stumped. Maybe his result will
give us a clue.



> Back to the drawing board...
>

I know how you feel. I'd really like to leave the Slug running (as
switching it off is a PITA - the front button does nothing) but with
an attached disk that goes to sleep.

I'll have another stab at it this weekend.
Cheers.

[Peter Chant]

On Friday 17 November 2006 16:49, jinnan_tonnix wrote:

>
> I hope Peter reads this and lists his disk mounts; if his shows the
> same (and his disk spins down) then I'm stumped. Maybe his result will
> give us a clue.

Just checking, I presume you don't mean me, my disk does not spin down.

I'll report back if I ever get any luck with my second Icy Box, when it gets
its fault sorted.

Pete

[ptrsmid]

I can not access my slug at the moment
I try to look into it but it might be not before the weekend

regards, Peter

--- In nslu2-general@..., "jinnan_tonnix"
<jinnan_tonnix@...> wrote:
>
> --- In nslu2-general@..., "Mat" <mat@> wrote:
> >
> > I did wonder about the swap thing. I totally forgot about it in
> > earlier testing, as it doesn't show up in cat /proc/mounts and
mount.
> >
>
> I'll try turning the swap off tonight, but I think I'll end up with
> the same result as you (i.e. waste of time)
>
> > I'm starting to wonder if the WD mybook drive is one which just
won't
> > spin down on the slug - it spins down when the slug is turned
off, and
> > spins up when the slug is turned on, but it just sits there
accessing
> > away, and spinning full time.
> >
>
> I've got a Buffalo 300GB drive. It's got some funky power
management
> features; I've set it to switch off when the Slug's off and to spin
> down after 10mins of inactivity. I know it works as I've tried it
on a
> Windows machine. The problem is that it's never inactive!!! Every 5
> seconds - blip! - little red disk light. I'm starting to take it
> personally!
>
> I hope Peter reads this and lists his disk mounts; if his shows the
> same (and his disk spins down) then I'm stumped. Maybe his result
will
> give us a clue.
>
>
>
> > Back to the drawing board...
> >
>
> I know how you feel. I'd really like to leave the Slug running (as
> switching it off is a PITA - the front button does nothing) but
with
> an attached disk that goes to sleep.
>
> I'll have another stab at it this weekend.
> Cheers.
>

[jinnan_tonnix]

--- In nslu2-general@..., "ptrsmid" <peter@...> wrote:
>
> I can not access my slug at the moment
> I try to look into it but it might be not before the weekend
>
> regards, Peter
>

Thanks Peter. If you could post the result of 'mount' and 'cat
/proc/mount' that would be very helpful. At least I'll know if I'm on
the right track.

[caldna]

FWIW, I have a 500GB WD MyBook Premium (USB+Firewire) and it also does
NOT spindown using unslung 6.8 + Andreas' script (rc.bootbin) +
noatime script (rc.local).  The slug also accesses the drive about
every 5 seconds and it doesn't matter if the drive has been natively
formatted by the slug, formatted in a Linux machine, or NTFS formatted
in Windows.

However, using unslung 5.5 and the other two scripts, the drive WILL
spindown after 10 minutes as expected.  I'm sticking to 5.5 until a
solution for 6.8 is found.





--- In nslu2-general@..., Mat <mat@...> wrote:

>
> I can confirm that doing what you're doing (using Andreas's script and
> the noatime script) on an unslung 6.8 slug, and getting same results as
> you for the cat /proc/mounts and the mount commands ('cept I'm sda not
> sdb), something is accessing the disk every 5 or so seconds.
>
> I've got a western digital mybook, with the circular disk access lights
> on, so it's easy to see when it's being accessed. I am running quite a
> bit on my slug at the moment though, and my next round of testing is to
> disable everything and start up the services one by one.
>
> jinnan_tonnix wrote:
> > Hi all,
> >
> > I'm having trouble getting my drive to spin down. It seems it's being
> > accessed every 5 seconds by the journalling system.
> >
> > I've used Andreas's script and the 'noatime' script to remount var and
> > dev in memory rather than disk, and to set the 'noatime' flag on

the disk.

> >
> > Could someone check their system to see if they get the same result?
> > Thanks. Please post back!
> >
> > 'cat /proc/mounts' shows:
> > # cat /proc/mounts
> > rootfs / rootfs rw 0 0
> > /dev/root /initrd jffs2 rw 0 0
> > /dev/sdb1 / ext3 rw,noatime 0 0
> > /proc /proc proc rw 0 0
> > usbdevfs /proc/bus/usb usbdevfs rw 0 0
> > /dev/sdb1 /dev.state ext3 rw,noatime 0 0
> > ramfs /dev ramfs rw 0 0
> > /dev/sdb1 /var.state ext3 rw,noatime 0 0
> > ramfs /var ramfs rw 0 0
> > /dev/sdb1 /share/hdd/data ext3 rw,noatime 0 0
> > /dev/sdb2 /share/hdd/conf ext3 rw,noatime 0 0
> >
> >
> > 'Mount' shows
> > # mount
> > /dev/sdb1 on /initrd type jffs2 (rw)
> > /dev/sdb1 on / type ext3 (rw,noatime)
> > /proc on /proc type proc (rw)
> > usbdevfs on /proc/bus/usb type usbdevfs (rw)
> > /dev/sdb1 on /dev.state type ext3 (rw,noatime)
> > ramfs on /dev type ramfs (rw)
> > /dev/sdb1 on /var.state type ext3 (rw,noatime)
> > ramfs on /var type ramfs (rw)
> > /dev/sdb1 on /share/hdd/data type ext3 (rw,noatime)
> > /dev/sdb2 on /share/hdd/conf type ext3 (rw,noatime)
> >
> >
>

[donotdespisethesnake]

I have similar problem, with a WD Essential drive. This used to spin
down on 5.5, although it was not the root then. With 6.8 there is a
regular disk access. I don't have any extra apps running.

Actually, I wonder if the problem is related to a process scanning the
USB. To my ear, the disk access sounds a lot like a disk reset,
perhaps something sending an Identify command.

What does the "onetouch_detect" process do, I wonder?


--- In nslu2-general@..., "caldna" <caldna@...> wrote:
>
> FWIW, I have a 500GB WD MyBook Premium (USB+Firewire) and it also does
> NOT spindown using unslung 6.8 + Andreas' script (rc.bootbin) +
> noatime script (rc.local).  The slug also accesses the drive about
> every 5 seconds and it doesn't matter if the drive has been natively

[ptrsmid]

Just a quick reply, tomorrow morning i'm of for the weekend again.
Next week I'll try to report wat actions I took to let the harddisk
sleep.

# cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/root /initrd jffs2 rw 0 0
/dev/sda1 / ext3 rw,noatime 0 0
/proc /proc proc rw 0 0
usbdevfs /proc/bus/usb usbdevfs rw 0 0
/dev/sda1 /dev.state ext3 rw,noatime 0 0
ramfs /dev ramfs rw 0 0
/dev/sda1 /var.state ext3 rw,noatime 0 0
ramfs /var ramfs rw 0 0
/dev/sda1 /share/flash/data ext3 rw,noatime 0 0
/dev/sda2 /share/flash/conf ext3 rw,noatime 0 0

# mount
/dev/sda1 on /initrd type jffs2 (rw)
/dev/sda1 on / type ext3 (rw,noatime)
/proc on /proc type proc (rw)
usbdevfs on /proc/bus/usb type usbdevfs (rw)
/dev/sda1 on /dev.state type ext3 (rw,noatime)
ramfs on /dev type ramfs (rw)
/dev/sda1 on /var.state type ext3 (rw,noatime)
ramfs on /var type ramfs (rw)
/dev/sda1 on /share/flash/data type ext3 (rw,noatime)
/dev/sda2 on /share/flash/conf type ext3 (rw,noatime)
#

hope it helps.

regards, peter

--- In nslu2-general@..., "jinnan_tonnix"
<jinnan_tonnix@...> wrote:

>
> --- In nslu2-general@..., "ptrsmid" <peter@> wrote:
> >
> > I can not access my slug at the moment
> > I try to look into it but it might be not before the weekend
> >
> > regards, Peter
> >
>
> Thanks Peter. If you could post the result of 'mount' and 'cat
> /proc/mount' that would be very helpful. At least I'll know if I'm on
> the right track.
>

[mururoa]

My own 2 cents :
- remount all using the atime parameter
- track wich files are modified/accessed using find based on access time
- track wich process access the files using fuser
- take the appropriate actions so that the process dont access the files
- verify
- remount all using the noatime parameter

ptrsmid wrote:

Just a quick reply, tomorrow morning i'm of for the weekend again.
Next week I'll try to report wat actions I took to let the harddisk
sleep.

[ptrsmid]

A little bit more information:
two scripts I put in /unslung:

/unslung/rcbootbin:

#!/bin/sh
# /unslung/rc.bootbin
# Tested with unslung firmware 5.5
# create ramdisk for /var and /dev

/bin/echo "Creating ramfs for /dev:"
/bin/mount -t ramfs ramfs /mnt/tmpmnt
/bin/cp -rp /dev/* /mnt/tmpmnt/
/bin/mkdir -p /dev.state
/bin/mount -o bind /dev /dev.state
/bin/mount -o bind /mnt/tmpmnt /dev
/bin/umount /mnt/tmpmnt
/bin/echo "Creating ramfs for /var:"
/bin/mount -t ramfs ramfs /mnt/tmpmnt -o maxsize=256
/bin/cp -rp /var/* /mnt/tmpmnt/
/bin/mkdir -p /var.state
/bin/mount -o bind /var /var.state
/bin/mount -o bind /mnt/tmpmnt /var
/bin/umount /mnt/tmpmnt

# restart logging daemons
/bin/killall syslogd
/bin/killall klogd

# Return and continue execution of system rc.
return 1
# [EOF]


/unslung/rc.local:

#!/bin/sh
# /unslung/rc.local
# A diversion script to remount the drive(s) without access
# times being recorded (the update of access times can
# prevent drives sleeping)

# Usually it is enough just to do /sda1 as this is the usually
# the one that holds the system.
/bin/echo "Remounting /dev/sda1 with noatime"
/bin/mount -o remount,rw,noatime /dev/sda1
/bin/echo  "Remounting /dev/sda2 with noatime"
/bin/mount -o remount,rw,noatime /dev/sda2

# return 1
# EOF - include this line

I recently reinstalled unslung 6.8 to the slug with one maxtor
onetouch 250 mb harddisk attached.
Currently I run openssh, mt-daapd and twonkyvision on the slug.
Haven't done much more
Here's what's running on the slug:

# ps
  PID TTY     Uid        Size State Command
    1         root       1212   S   /bin/init
    2         root          0   S   [keventd]
    3         root          0   S   [ksoftirqd_CPU0]
    4         root          0   S   [kswapd]
    5         root          0   S   [bdflush]
    6         root          0   S   [kupdated]
    7         root          0   S   [mtdblockd]
    8         root          0   S   [khubd]
    9         root          0   S   [jffs2_gcd_mtd4]
   10         root          0   S   [usb-mxo-ot0]
   11         root          0   S   [usb-storage-0]
   12         root          0   S   [scsi_eh_0]
   19         root          0   S   [sd-mc-thread]
   28         root          0   S   [kjournald]
   51         root          0   D   [ixp425_csr]
   52         root          0   S   [ixp425 ixp0]
   55 ttyS0   root       1916   S   /bin/sh
  164         root          0   S   [kjournald]
  165         root       1936   S   /sbin/syslogd -n
  166         root       1924   S   /sbin/klogd -n
  228         root      10992   S   upnpd &>/dev/null
  238         root      10992   S   upnpd &>/dev/null
  239         root      10992   S   upnpd &>/dev/null
  243         root      10992   S   upnpd &>/dev/null
  244         root      10992   S   upnpd &>/dev/null
  342         root       2152   S   /usr/sbin/thttpd -
C /etc/thttpd.conf
  366         root       6156   S   /usr/sbin/smbd -D
  368         root       6156   S   /usr/sbin/smbd -D
  369         root       4824   S   /usr/sbin/nmbd -D
  376         root       1200   S   /bin/inetd
  391         root       1952   S   /usr/sbin/QuickSet
  398         root       1904   S   /usr/sbin/USB_Detect
  402         root       1900   S   /usr/sbin/USB_Detect
  406         root       1884   S   /usr/sbin/onetouch_detect
  410         root       1884   S   /usr/sbin/onetouch_detect
  423         root       1296   S   /usr/sbin/crond
  448         root       3200   S   /opt/sbin/sshd -
f /opt/etc/openssh/sshd_con
  455         guest      3960   S   /opt/sbin/mt-daapd -
c /opt/etc/mt-daapd/mt-
  456         guest     13900   S   /opt/sbin/mt-daapd -
c /opt/etc/mt-daapd/mt-
  459         guest     13900   S   /opt/sbin/mt-daapd -
c /opt/etc/mt-daapd/mt-
  460         guest     13900   S   /opt/sbin/mt-daapd -
c /opt/etc/mt-daapd/mt-
  462         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  469         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  470         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  471         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  472         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  473         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  474         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  475         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  477         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  478         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  479         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  484         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  487         root      41604   S   /opt/twonkymedia -D -inifile
twonkypeter.in
  504         guest     13900   S   /opt/sbin/mt-daapd -
c /opt/etc/mt-daapd/mt-
28178         root      10992   S   upnpd &>/dev/null
28328         root       5808   S   sshd: root@ttyp0
28330 ttyp0   root       1920   S   -sh
28339 ttyp0   root       1984   R   ps
#


The disk does spin down but occasionaly spins up again. It's
difficult to find out but i guess the slug spins up on average about
every hour.
It would be nice to have a tool to check what program is accessing
the hardisk but I understood from the wiki that's not available for
unslung 6.8. If anyone has better ideas let me know.
Hope this will help

regards, peter




--- In nslu2-general@..., "ptrsmid" <peter@...> wrote:
>
> Just a quick reply, tomorrow morning i'm of for the weekend again.
> Next week I'll try to report wat actions I took to let the
harddisk

> sleep.
>
> # cat /proc/mounts
> rootfs / rootfs rw 0 0
> /dev/root /initrd jffs2 rw 0 0
> /dev/sda1 / ext3 rw,noatime 0 0
> /proc /proc proc rw 0 0
> usbdevfs /proc/bus/usb usbdevfs rw 0 0
> /dev/sda1 /dev.state ext3 rw,noatime 0 0
> ramfs /dev ramfs rw 0 0
> /dev/sda1 /var.state ext3 rw,noatime 0 0
> ramfs /var ramfs rw 0 0
> /dev/sda1 /share/flash/data ext3 rw,noatime 0 0
> /dev/sda2 /share/flash/conf ext3 rw,noatime 0 0
>
> # mount
> /dev/sda1 on /initrd type jffs2 (rw)
> /dev/sda1 on / type ext3 (rw,noatime)
> /proc on /proc type proc (rw)
> usbdevfs on /proc/bus/usb type usbdevfs (rw)
> /dev/sda1 on /dev.state type ext3 (rw,noatime)
> ramfs on /dev type ramfs (rw)
> /dev/sda1 on /var.state type ext3 (rw,noatime)
> ramfs on /var type ramfs (rw)
> /dev/sda1 on /share/flash/data type ext3 (rw,noatime)
> /dev/sda2 on /share/flash/conf type ext3 (rw,noatime)
> #
>
> hope it helps.
>
> regards, peter
>
> --- In nslu2-general@..., "jinnan_tonnix"
> <jinnan_tonnix@> wrote:
> >
> > --- In nslu2-general@..., "ptrsmid" <peter@> wrote:
> > >
> > > I can not access my slug at the moment
> > > I try to look into it but it might be not before the weekend
> > >
> > > regards, Peter
> > >
> >
> > Thanks Peter. If you could post the result of 'mount' and 'cat
> > /proc/mount' that would be very helpful. At least I'll know if

I'm on
> > the right track.
> >
>

[jinnan_tonnix]

Thanks for the info Peter - really helpful.

Bit of a long shot here. Someone in this forum mentioned that it
might be the USB scanning function. I also seem to remember that the
two USB ports behave slightly differently (hubs etc).

My disk is plugged into USB port 1. Taking a stab in the dark; Peter
is yours plugged into port 2?



--- In nslu2-general@..., "ptrsmid" <peter@...> wrote:
>
> A little bit more information:
> two scripts I put in /unslung:
>

You use the same scripts as I do.


>
> I recently reinstalled unslung 6.8 to the slug with one maxtor
> onetouch 250 mb harddisk attached.
> Currently I run openssh, mt-daapd and twonkyvision on the slug.
> Haven't done much more
> Here's what's running on the slug:

I use Dropbear instead of OpenSSH. Might look into this.



>
>
> The disk does spin down but occasionaly spins up again. It's
> difficult to find out but i guess the slug spins up on average
about
> every hour.

Perhaps Twonky is scanning your disk?



Cheers.

[Robert Hammond]

In message <ekfg88+3ipt@...>, ptrsmid <peter@...> writes
<large snip>
>
>The disk does spin down but occasionaly spins up again. It's
>difficult to find out but i guess the slug spins up on average about
>every hour.
>It would be nice to have a tool to check what program is accessing
>the hardisk but I understood from the wiki that's not available for
>unslung 6.8. If anyone has better ideas let me know.
>Hope this will help
>
On my 5.5 NSLU2 it would also spin up every hour but I eventually
tracked this down to a cron job script,  the script that Linksys use to
correct the clock every hour.  I have changed mine to correct the clock
once every day (at midnight) as follows :-

Using a Linux style text editor edit the file /etc/crontab

Look for the line
1 0 * * * root /usr/sbin/hwclock -s &>/dev/null

Change this too
0 0 * * * root /usr/sbin/hwclock -s &>/dev/null

You will need to re-start the cron program or just re-boot your NSLU2
for this fix to load up.  Any manual changes to this file seem to
survive a re-boot but suggest you should check just to make sure.


--
Robert Hammond
PGP:0x154144DA