WAN (eth2) issue on turris v1 with 5.x
Summary
After upgrading to Turris OS 5.x WAN link (eth2) reconnecting few times per day. This causing short internet interruptions. To exclude ISP from possible reasons i put switch in the middle. Issue was by upgrade to TOS6. There are few kernel patches related to gianfar
support in the master which seems to be relevant. There are not a lot of them, so IMO backporting is not very hard to do. As it was suggested on the forum i decided to report this issue on the gitlab.
How to reproduce
Install TOS 5.x on the Turris 1.x and monitor kernel events. In my case it usually happens during working hours, so likely related to the network activity (number of sessions?). Changing from DHCP to static does not fix the issue. Usually i have 2-5 such events per day.
[247392.032542] fsl-gianfar ffe26000.ethernet eth2: Link is Down
[247392.119413] IPv6: ADDRCONF(NETDEV_UP): eth2: link is not ready
[247401.312447] fsl-gianfar ffe26000.ethernet eth2: Link is Up - 1Gbps/Full - flow control off
[247401.312469] IPv6: ADDRCONF(NETDEV_CHANGE): eth2: link becomes ready
After TOS6 upgrade
There are no WAN interface events after upgrade to TOS6, system is up and running very stable.