Re: [Voyage-linux] Voyage 0.10: Ethernet bug when moving CF-disks to other boards discovered

2016-11-26 Thread Jon Meek
Hi Andreas - I have checked my notes and this has been "an issue" going
back to at least SuSE Linux (a fine German product) in 2006. So, it is not
Voyage-linux specific, but maybe Debian related.
Here is at least one good reason for this: suppose you have a server
running in a data center and you need to add some more Ethernet ports. You
have the DC people put in the card and turn the machine back on. It would
be very bad if the port assignment algorithm decided to assign one of the
new ports to eth0. That new port would likely not even be plugged into a
switch port at boot time and you would have to use the serial console (if
you have one) to sort out the mess.

However, I agree that your situation was frustrating, and I have
experienced it myself.

Jon


On Sat, Nov 26, 2016 at 2:15 PM, Andreas Delleske 
wrote:

> Hi Jon,
>
> > I don't think that this is a bug. Those rules keep your Ethernet port
> names
> > consistent over time. In my cloning notes I have a step to delete the
> > appropriate lines before moving the disk / CF card to a new system.
>
> Objection, your honour :)
>
> At least when we chose DHCP I guess everyone wants the device working
> at a new place in a new hardware. It's not even possible to swap a
> broken board this way. Everyone should know this at least, and I think
> this behaviour had been different in previous versions of voyage
> (which is fantastic BTW).
>
> So I think the installer could offer to add your modification..
>
> And yes, I would have expected at least an eth1 popping up ..
>
> --
> Andreas
>



-- 
Jon T. Meek, Ph.D.
 https://linkedin.com/in/meekjt
 https://meekj.github.io
___
Voyage-linux mailing list
Voyage-linux@list.voyage.hk
http://list.voyage.hk/mailman/listinfo/voyage-linux


Re: [Voyage-linux] Voyage 0.10: Ethernet bug when moving CF-disks to other boards discovered

2016-11-26 Thread Andreas Delleske
Hi Jon,

> I don't think that this is a bug. Those rules keep your Ethernet port names
> consistent over time. In my cloning notes I have a step to delete the
> appropriate lines before moving the disk / CF card to a new system.

Objection, your honour :)

At least when we chose DHCP I guess everyone wants the device working
at a new place in a new hardware. It's not even possible to swap a
broken board this way. Everyone should know this at least, and I think
this behaviour had been different in previous versions of voyage
(which is fantastic BTW).

So I think the installer could offer to add your modification..

And yes, I would have expected at least an eth1 popping up ..

-- 
Andreas

___
Voyage-linux mailing list
Voyage-linux@list.voyage.hk
http://list.voyage.hk/mailman/listinfo/voyage-linux


Re: [Voyage-linux] Voyage 0.10: Ethernet bug when moving CF-disks to other boards discovered

2016-11-26 Thread Jon Meek
I don't think that this is a bug. Those rules keep your Ethernet port names
consistent over time. In my cloning notes I have a step to delete the
appropriate lines before moving the disk / CF card to a new system.

I would have expected a new line for eth1 to be added to the file, which is
not what you want, of course.

Jon


On Sat, Nov 26, 2016 at 8:00 AM, Andreas Delleske 
wrote:

> Hi list,
>
> I guess I have discovered a bug:
>
> With an Alix board and Voyage Linux 0.10.0, I prepared a Compact Flash
> with the usual script.
>
> It started fine, I got ssh access etc., DHCP was working, Ethernet..
>
> Then when I stopped the machine (orderly shutdown) and plugged the
> CF-card into another ALIX board (with a different Ethernet MAC, the
> Ethnernet interface eth0 did not even come up at boot - it was not
> present even in "ifconfig".
>
> The problem is, that when starting up and the following file does not
> yet exist, the system "hardwires" the MAC address into the file, like
> so:
>
> xx:~# cat /etc/udev/rules.d/70-persistent-net.rules
> # This file was automatically generated by the /lib/udev/write_net_rules
> # program, run by the persistent-net-generator.rules rules file.
> #
> # You can modify it, as long as you keep each rule on a single
> # line, and change only the value of the NAME= key.
>
> # PCI device 0x1106:0x3053 (via-rhine)
> SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*",
> ATTR{address}=="00:0d:b9:17:fc:d0", ATTR{dev_id}=="0x0",
> ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"
>
> I think it is this problem.
>
> http://unix.stackexchange.com/questions/110496/cloned-linux-
> system-and-etc-udev-rules-d-70-persistent-net-rules
>
> Could the voyage code be adapted or is this a Debian thing? Come
> something be done?
>
>
> --
> Cheers
> Andreas
>
> ___
> Voyage-linux mailing list
> Voyage-linux@list.voyage.hk
> http://list.voyage.hk/mailman/listinfo/voyage-linux
>



-- 
Jon T. Meek, Ph.D.
 https://linkedin.com/in/meekjt
 https://meekj.github.io
___
Voyage-linux mailing list
Voyage-linux@list.voyage.hk
http://list.voyage.hk/mailman/listinfo/voyage-linux


Re: [Voyage-linux] Voyage 0.10.0 on ALIX wants to fsck :-) - does that make sense?

2016-11-26 Thread Michael Steinmann
do this:
remountro
e2fsck /dev/sda1



2016-11-26 14:04 GMT+01:00 Andreas Delleske :

> Hi all,
>
> after having successfully installed Voage Linux 0.10.0 on a Compact
> Flash disk for an ALIX board, the startup messages are reminding me to
> use fsck as the data structure would be OK yet they recommend using
> fsck:
>
> Nov 24 15:10:58 xxx vmunix: [   12.260186] EXT2-fs (sda1): warning:
> mounting unchecked fs, running e2fsck is recommended
>
> How do I do it? Or should I ignore that?
>
> Thanks
>
> --
> Andreas
>
> ___
> Voyage-linux mailing list
> Voyage-linux@list.voyage.hk
> http://list.voyage.hk/mailman/listinfo/voyage-linux
>
___
Voyage-linux mailing list
Voyage-linux@list.voyage.hk
http://list.voyage.hk/mailman/listinfo/voyage-linux


[Voyage-linux] Voyage 0.10.0 on ALIX wants to fsck :-) - does that make sense?

2016-11-26 Thread Andreas Delleske
Hi all,

after having successfully installed Voage Linux 0.10.0 on a Compact
Flash disk for an ALIX board, the startup messages are reminding me to
use fsck as the data structure would be OK yet they recommend using
fsck:

Nov 24 15:10:58 xxx vmunix: [   12.260186] EXT2-fs (sda1): warning:
mounting unchecked fs, running e2fsck is recommended

How do I do it? Or should I ignore that?

Thanks

-- 
Andreas

___
Voyage-linux mailing list
Voyage-linux@list.voyage.hk
http://list.voyage.hk/mailman/listinfo/voyage-linux


[Voyage-linux] Voyage 0.10: Ethernet bug when moving CF-disks to other boards discovered

2016-11-26 Thread Andreas Delleske
Hi list,

I guess I have discovered a bug:

With an Alix board and Voyage Linux 0.10.0, I prepared a Compact Flash
with the usual script.

It started fine, I got ssh access etc., DHCP was working, Ethernet..

Then when I stopped the machine (orderly shutdown) and plugged the
CF-card into another ALIX board (with a different Ethernet MAC, the
Ethnernet interface eth0 did not even come up at boot - it was not
present even in "ifconfig".

The problem is, that when starting up and the following file does not
yet exist, the system "hardwires" the MAC address into the file, like
so:

xx:~# cat /etc/udev/rules.d/70-persistent-net.rules
# This file was automatically generated by the /lib/udev/write_net_rules
# program, run by the persistent-net-generator.rules rules file.
#
# You can modify it, as long as you keep each rule on a single
# line, and change only the value of the NAME= key.

# PCI device 0x1106:0x3053 (via-rhine)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*",
ATTR{address}=="00:0d:b9:17:fc:d0", ATTR{dev_id}=="0x0",
ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"

I think it is this problem.

http://unix.stackexchange.com/questions/110496/cloned-linux-system-and-etc-udev-rules-d-70-persistent-net-rules

Could the voyage code be adapted or is this a Debian thing? Come
something be done?


-- 
Cheers
Andreas

___
Voyage-linux mailing list
Voyage-linux@list.voyage.hk
http://list.voyage.hk/mailman/listinfo/voyage-linux