ZyXEL NSA320 booted from USB...once!

[iwonbigbro]

I'll give it a go. Is the resume on power on state persistent, or does it need to be reminded after each boot? I remember an old pentium III WOL motherboards would reset the WOL state after a WOL event.

[WarheadsSE]

As far as I know, it stays until capacitor drains.

[iwonbigbro]

Okay, I set the resume state to be set on boot and have rebooted a couple of times now. Still no joy.

$this->bbcode_second_pass_code('', '
[ OK ] Started Login Service.
[ OK ] Started Set the power resume state on boot.
[FAILED] Failed to start A basic dhcp ethernet connection.
See 'systemctl status netctl@eth0.service' for details.

Arch Linux 3.1.10-22-ARCH (ttyS0)

alarm login: root
Password:
Last login: Fri May 12 06:10:26 on ttyS0
[root@alarm ~]# systemctl status netctl@eth0.service
netctl@eth0.service - A basic dhcp ethernet connection
Loaded: loaded (/etc/systemd/system/netctl@eth0.service; enabled)
Active: failed (Result: exit-code) since Tue 1965-07-20 00:36:29 MDT
Docs: man:netctl.profile(5)
Process: 112 ExecStart=/usr/lib/network/network start %I (code=exited, status=1/FAILURE)
')

Still the same deal, these messages in the journal:

$this->bbcode_second_pass_code('', '
Jul 20 00:36:29 alarm network[112]: Starting network profile 'eth0'...
Jul 20 00:36:29 alarm systemd[1]: netctl@eth0.service: main process exited, code=exited, status=1/FAILURE
Jul 20 00:36:29 alarm systemd[1]: Failed to start A basic dhcp ethernet connection.
Jul 20 00:36:29 alarm systemd[1]: Unit netctl@eth0.service entered failed state.
')

If I try to start dhcpcd manually, I get the following:

$this->bbcode_second_pass_code('', '
Jul 20 00:36:29 alarm systemd[1]: Starting dhcpcd on eth0...
Jul 20 00:36:29 alarm dhcpcd[418]: version 6.0.5 starting
Jul 20 00:36:29 alarm dhcpcd[418]: eth0: soliciting an IPv6 router
Jul 20 00:36:29 alarm dhcpcd[418]: eth0: soliciting a DHCP lease
Jul 20 00:36:29 alarm dhcpcd[418]: timed out
Jul 20 00:36:29 alarm dhcpcd[418]: exited
Jul 20 00:36:29 alarm systemd[1]: dhcpcd@eth0.service: control process exited, code=exited status=1
Jul 20 00:36:29 alarm systemd[1]: Failed to start dhcpcd on eth0.
Jul 20 00:36:29 alarm systemd[1]: Unit dhcpcd@eth0.service entered failed state.
')

Not seeing any DHCP requests coming through. About to run tcpdump to see if there is any traffic at all.

[iwonbigbro]

Nothing coming through on the interface.

$this->bbcode_second_pass_code('', '
root@yellowtail:~# tcpdump -i eth0 -vv
tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 65535 bytes
')

[pepedog]

Try this
poweroff
Remove power lead for 5 seconds
If that fails, plug in USB keyboard and alt-ctrl-del once booted
This is so flaky, method one used to work every time, until recently

[iwonbigbro]

Sure enough the power resume works, powered on as soon as I plugged in the DC supply again. At first, thought I had the same fault:

$this->bbcode_second_pass_code('', '
[ OK ] Started Set the power resume state on boot.
[FAILED] Failed to start A basic dhcp ethernet connection.
See 'systemctl status netctl@eth0.service' for details.
')

But then, as if by magic!

$this->bbcode_second_pass_code('', '
[ 25.569813] mv643xx_eth_port mv643xx_eth_port.0: eth0: link up, 100 Mb/s, full duplex, flow control disabled
[ 25.579872] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready

Arch Linux 3.1.10-22-ARCH (ttyS0)

alarm login: [ OK ] Started dhcpcd on eth0.
[ OK ] Reached target Multi-User System.
[ OK ] Reached target Graphical Interface.
[ 36.078048] eth0: no IPv6 routers present
')

Seems the Microsoft style power cycling approach seemed to do the trick!

$this->bbcode_second_pass_code('', '
# ifconfig
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 10.42.0.48 netmask 255.255.255.0 broadcast 10.42.0.255
inet6 fe80::ee43:f6ff:fec0:b1e8 prefixlen 64 scopeid 0x20<link>
')

Thanks!

[WarheadsSE]

I'll skip the gory details, but the EE's @ ZyXEL decided to get crafty, and the result is this.

[iwonbigbro]

I'm happy enough now. At least I can unplug the serial cable and tidy up my desk a bit.

I find it odd that they are actively putting money into stopping people from hacking the hardware!? Sure, stop remote hacks for security reasons, but hacks where you have to physically connect a cable or device? I see nothing wrong with that, they still got £60 out of me and now they don't even have to support me.

[WarheadsSE]

That was not their intent. They had other goals.