08:30 <fhumblot> Hey there, I'm running openwrt 22-03 on an ath79 target (ar9331 SoC) and am having a weird sporadic issue. When I issue a reboot command, the system sometimes gets into a state where it gives an I/O error on any command.

08:31 <fhumblot> It seems that it unmounts the MTD partitions and then fails to proceed with rebooting. The only thing that then works is to power cycle the device. Has anyone run into this issue before?

11:53 <fhumblot> The last log I seem to get from dmesg before getting thrown out of the device is: https://pastebin.com/ARv9G6UD afterwards I need a power cycle before it successfully boots again

12:35 <karlp> fhumblot: you could perhaps try this: https://github.com/etactica/openwrt/commit/7152087e9c1eef7490e4d8afa3b417e5f6e74f1b

12:36 <karlp> though, for me, "reboot" was always reliable, it was reboots caused by the watchdog that were unreliable,

12:36 <karlp> so sounds... not really the same?

12:39 <karlp> https://github.com/etactica/openwrt/commit/2f6dba833bed8130267b6cf69d0b7c0bdefc0734 is more up to date if you do want to try that

12:40 <karlp> I have it for 5.10 as well https://github.com/etactica/openwrt/commit/a2ebd39201d2bd788e43bf5b55018c74e13f1d2c but then switched to sunxi

12:57 <lu_zero> anybody else beside me is having problems in entering the recovery partition on the bpi-r3 with the 23.05?

13:20 <fhumblot> karlp: thanks for the hint! Is there a particular reason that this isn't upstreamed (to openwrt or Linux itself)? I also noticed that the commit message speaks of carambola2 devices, which is also the device I'm using lol so this may actually be it

13:46 <Slimey> heh

13:46 <Slimey> you could like kill some important system process ;)

14:35 <fhumblot> Slimey: heh, but if the watchdog triggers, you have bigger problems I suppose

18:02 <stintel> I'm working on a protocol handler for OpenThread ... I have a bunch of checks to fail setup. I call proto_setup_failed when any of the checks fail, but this results in an (seemingly) endless loop of netifd trying to bring up the interface. how can I avoid that loop?

18:02 <stintel> code is here: https://github.com/openwrt/packages/blob/5ec2b3b67436051bf95a5ef14cb8842c441c56a4/net/openthread-br/files/openthread-proto.sh

18:16 <stintel> proto_block_restart seems useful

20:48 <KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_lantiq.html has been updated. (95.9% images and 99.9% packages reproducible in our current test framework.)

21:16 <Slimey> lol

21:16 <Slimey> just messing around over here...

21:16 <Slimey> [ 0.000000] Booting Linux on physical CPU 0x0

21:16 <Slimey> [ 0.000000] CPU: ARMv7 Processor [410fc075] revision 5 (ARMv7), cr=30c5387d

21:16 <Slimey> [ 0.000000] Linux version 6.1.42 (builder@buildhost) (arm-openwrt-linux-muslgnueabi-gcc (OpenWrt GCC 12.3.0 r23685-7e7eb5312d) 12.3.0, GNU ld (GNU Binutils) 2.40.0) #0 SMP Sat Aug 5 16:42:45 2023

21:16 <Slimey> [ 0.000000] OF: fdt: Machine model: Broadcom BCM947622

21:19 <Slimey> did not expect it to actually start heh