lime-config do not configure properly ramips network

[p4u]

As reported in the mailing list there is a problem with the lime-config network interface detection when using it in ramips architecture.

root@LiMe-16da88:~# lime-config Clearing wireless config... Clearing network config... Disabling odhcpd Cleaning dnsmasq Disabling 6relayd... /usr/lib/lua/lime/network.lua:155: attempt to index local 'dev' (a nil value) stack traceback: /usr/bin/lime-config:45: in function </usr/bin/lime-config:45> C: in function 'foreach' /usr/lib/lua/lime/network.lua:192: in function 'scandevices' /usr/lib/lua/lime/network.lua:217: in function </usr/lib/lua/lime/network.lua:208> C: in function 'xpcall' /usr/bin/lime-config:45: in function 'main' /usr/bin/lime-config:65: in main chunk

Configuring system... Let uhttpd listen on IPv4/IPv6

root@LEDE:~# ip link 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UNKNOWN qlen 1000 link/ether 78:a3:51:16:da:88 brd ff:ff:ff:ff:ff:ff 3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 78:a3:51:16:da:89 brd ff:ff:ff:ff:ff:ff 4: wlan1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 78:a3:51:16:da:88 brd ff:ff:ff:ff:ff:ff 5: br-lan: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP qlen 1000 link/ether 78:a3:51:16:da:88 brd ff:ff:ff:ff:ff:ff 6: eth0.1@eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br-lan state UP qlen 1000 link/ether 78:a3:51:16:da:88 brd ff:ff:ff:ff:ff:ff 7: eth0.2@eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP qlen 1000 link/ether 78:a3:51:16:da:89 brd ff:ff:ff:ff:ff:ff

root@LEDE:~# iw list| grep Wiphy Wiphy phy1 Wiphy phy0

[p4u]

The problem might be here:

https://github.com/libremesh/lime-packages/blob/develop/packages/lime-system/files/usr/lib/lua/lime/network.lua#L196

EDIT: it is not

[ilario]

A temporary workaround consists in commenting out this line in network.lua: https://github.com/libremesh/lime-packages/blob/7cda1ad852055e85a3fec73764826ec00ed735bd/packages/lime-system/files/usr/lib/lua/lime/network.lua#L178 which had been introduced two years ago by @G10h4ck in caad3ac. I copy from the lime-users mailing list my explanation of the bad behaviour:

On 04/14/2017 05:53 AM, Juergen Kimmel wrote:

Here is /etc/config/network

config device 'lan_dev' option name 'eth0.1' option macaddr '78:a3:51:16:da:88'

config device 'wan_dev' option name 'eth0.2' option macaddr '78:a3:51:16:da:89'

Ok, now can you try to comment out (prepending "--!") line 178 in /usr/lib/lua/lime/network.lua ?

function owrt_device_parser(section) dev_parser(section["name"]) --! dev_parser(section["ifname"]) end

https://github.com/libremesh/lime-packages/blob/develop/packages/lime-system/files/usr/lib/lua/lime/network.lua#L178

According to LEDE and OpenWrt wikis [1,2], sections of type device in /etc/config/network should have an ifname option. LibreMesh scripts assume that there is such option and the crash you experienced could be due to the lack of an "option ifname 'eth0'" in device sections in /etc/config/network on ramips. I hope this works... :) Ilario

[1] https://wiki.openwrt.org/doc/uci/network#devices [2] https://lede-project.org/docs/user-guide/switch_configuration#devices

[dangowrt]

We should switch to use our own version of config-generate and use /etc/board.d stuff instead asap...

On Tue, Apr 18, 2017 at 03:27:38AM -0700, Ilario Gelmetti wrote:

A temporary workaround consists in commenting out this line in network.lua: https://github.com/libremesh/lime-packages/blob/7cda1ad852055e85a3fec73764826ec00ed735bd/packages/lime-system/files/usr/lib/lua/lime/network.lua#L178 which had been introduced two years ago by @G10h4ck in caad3ac. I copy from the lime-users mailing list my explanation of the bad behaviour:

On 04/14/2017 05:53 AM, Juergen Kimmel wrote:

Here is /etc/config/network

config device 'lan_dev' option name 'eth0.1' option macaddr '78:a3:51:16:da:88'

config device 'wan_dev' option name 'eth0.2' option macaddr '78:a3:51:16:da:89'

Ok, now can you try to comment out (prepending "--!") line 178 in /usr/lib/lua/lime/network.lua ?

function owrt_device_parser(section) dev_parser(section["name"]) --! dev_parser(section["ifname"]) end

https://github.com/libremesh/lime-packages/blob/develop/packages/lime-system/files/usr/lib/lua/lime/network.lua#L178

According to LEDE and OpenWrt wikis [1,2], sections of type device in /etc/config/network should have an ifname option. LibreMesh scripts assume that there is such option and the crash you experienced could be due to the lack of an "option ifname 'eth0'" in device sections in /etc/config/network on ramips. I hope this works... :) Ilario

[1] https://wiki.openwrt.org/doc/uci/network#devices [2] https://lede-project.org/docs/user-guide/switch_configuration#devices

-- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/libremesh/lime-packages/issues/117#issuecomment-294770210

[p4u]

@ilario I think your workaround would break other platforms.

IMO the workaround should add a first line in dev_parser() function like: if dev == nil then return end

That should solve the crash too.

However I don't really understand what is the point of these sections:

config device 'lan_dev' option name 'eth0.1' option macaddr '78:a3:51:16:da:88'

I never saw them before, it's a ramips thing? What happens if eth0.1 is defined here but also in "br-lan" ?

[p4u]

We should switch to use our own version of config-generate and use /etc/board.d stuff instead asap...

@dangowrt that is probably right, but it will be a work maybe for the next release

[dangowrt]

On Tue, Apr 18, 2017 at 03:48:55AM -0700, Pau wrote:

@ilario I thin kyour workaround would break other platforms.

IMO the workaround should add a first line in dev_parser() function like: if dev == nil then return end

That should solve the crash too.

However I don't really understand what is the point of these sections:

config device 'lan_dev' option name 'eth0.1' option macaddr '78:a3:51:16:da:88'

I never saw them before, it's a ramips thing? What happens if eth0.1 is defined here but also in "br-lan" ?

It would have the same mac address as the WAN interface. Please have a look here to understand how config is generated and realize that there is no point to re-process the generated config to generate yet another config.

https://git.lede-project.org/?p=source.git;a=blob;f=package/base-files/files/bin/config_generate;h=6230b0892f04c9853706a06a9547c7e5cc249625;hb=HEAD#l83