no really build related...
but for compiling most stuff natively on the router you have to use a chroot/lxc/docker/full-distro
debian chroots were my 'goto' choice... but there are probably better lxc/docker options now...
no really build related...
but for compiling most stuff natively on the router you have to use a chroot/lxc/docker/full-distro
debian chroots were my 'goto' choice... but there are probably better lxc/docker options now...
I have several Error: Parsing the config file failed!
in the log. How can I diagnose what this is ? This message also appears when I ssh to the pi4b.
just grep -C6 around it... i.e.;
logread | grep -C6 'the config file failed'
(sanitize if needed before posting)
edit: looks like related to collectd;
/usr/sbin/collectd:Error: Parsing the config file failed!
so also...
/etc/init.d/luci_statistics restart; sleep 6; logread | tail -n20
Sorry, I've seen the solution earlier in this thread, now that error has gone.
But I have other errors, such as Mon Jan 17 09:42:00 2022 kern.warn kernel: [ 9.979134] brcmfmac mmc1:0001:1: Direct firmware load for brcm/brcmfmac43455-sdio.raspberrypi,4-model-b.bin failed with error -2
this one is semi-normal... ( can be ignored ) ... at least I think it is... i'd need to boot the stock(official) image to confirm... but I get this message too... and wifi still works... afaik it falls back to brcmfmac43455-sdio.bin
or something...
Thanks. I have wifi disabled in the pi, maybe that's what's causing it.
Another one: when I ssh to the pi some login messages are in red
MonJan1709:44:102022 plugin: rrd_update_r failed: /tmp/rrd/rpi-aveiro-router/cpu-3/percent-softirq.rrd: opening
MonJan1709:44:132022 62.253264] tailroom_fix: ethtool -K eth0 rx off [skip_notailroomindmesg]
MonJan1709:44:132022 62.260160] EEE_DISABLE [off]
Do I need to care about this ?
sometimes, these appear when something is new and it cant read them on first load...
(semi-normal)
if it keeps re-appearing on (service) restart... then sometimes you need to get rid of your history when plugins change and whatnot...
/etc/init.d/luci_statistics stop
rm -rf /tmp/rrd/*; rm -rf /boot/psave/rrd/*
/etc/init.d/luci_statistics start
the other two messages are normal...
Thanks. That solved the first message.
Now the last one (I hope). If I reboot using command line I get this error message cat: can't open '/sys/class/net/br-lan/address': No such file or directory
but the pi reboots.
o good... that's a new one
will need to track down where it is coming from...
thankyou!
grep localversion /etc/custom/buildinfo.txt
localversion="5.0.11-32"
fixed-ish... for a quick workaround before the next build you can run this;
sed -i 's|br-lan/address|br-lan/address 2>/dev/null|g' /usr/sbin/reboot
Fixed. You are the greatest. Thank you very much.
Hello, i've just upgraded from version 5.0.11-32 to 5.0.19-7 (because the banner shows new update found, usually it shows no-available-update) and noticed my frequency stuck at 600mhz but vcgencmd get_throttled shows 0x0.
Also, my logs keeps spammed with dnsmasq and after a while the router itself will reboot.
the last version if i remember correctly that havent problem for me is 1.0.10-3.
here is my logs and other relevant info. Thanks.
firstly thankyou very much for the detailed debugging info!
secondly... can I see please
cat /etc/config/dhcp | grep -v ipset
###################### and next time you have a working build
uci show dhcp | grep -v ipset | grep -v host
(you can remove client mac addresses / reservations if present)
(this is out of interest to see whats up with ujail... but chances are I will remove it)
hmmmm... interesting you seem to be using wifi dongle as wan/wwan...
Tue Jan 18 14:31:52 2022 user.notice kopijahe 2.2.2: Percobaan login. Interface: wwan. Hasil: Gagal Login (User Anda Sedang Aktif)
(is the above message travelmate? login script? dengan saya senang sekali bertumu dengan anda!)
fyi.. I just pushed 5.0.19-9_r18609
(stable||current) which no longer has ujail
... it would be good if you can see how that goes...
sorry for the problems but glad you reported them in detail, what i'm seeing is so unusual this may take some time to troubleshoot... I can also upload older builds for you if needed...
or you can also just install ( sysupgrade -R ) one of these older builds which should probably be ok?;
here is my ipset output
config dnsmasq
option domainneeded '1'
option localise_queries '1'
option rebind_protection '1'
option rebind_localhost '1'
option local '/lan/'
option domain 'lan'
option expandhosts '1'
option authoritative '1'
option readethers '1'
option leasefile '/tmp/dhcp.leases'
option localservice '1'
option ednspacket_max '1232'
option dnsforwardmax '2300'
option min_cache_ttl '270'
option cachesize '5000'
list address '/router/192.168.1.4'
option confdir '/tmp/dnsmasq.d'
list server '/welcome2.wifi.id/118.98.44.10'
list server '/welcome2.wifi.id/118.98.44.100'
list server '127.0.0.1#5054'
list server '127.0.0.1#5053'
config dhcp 'lan'
option interface 'lan'
option start '100'
option limit '150'
option leasetime '12h'
option dhcpv4 'server'
option dhcpv6 'server'
option ra 'server'
option ra_slaac '1'
list ra_flags 'managed-config'
list ra_flags 'other-config'
config dhcp 'wan'
option interface 'wan'
option ignore '1'
config host
option mac 'DC:A6:32:xx:xx:xx'
option ip '192.168.1.203'
option leasetime '365d'
config host
option mac '48:F1:7F:xx:xx:xx'
option ip '192.168.1.201'
option leasetime '365d'
config domain
option ip '192.168.1.203'
option name 'rpi2.lan'
config host
option mac '00:E0:4C:xx:xx:xx'
option ip '192.168.1.202'
option leasetime '365d'
config host
option mac '00:0C:29:xx:xx:xx'
option ip '192.168.1.204'
option leasetime '365d'
option dns '1'
config domain
option name 'welcome2.wifi.id'
option ip '10.233.16.13'
config host
option mac 'DC:A6:32:xx:xx:xx'
option ip '192.168.1.205'
option leasetime '365d'
config domain
option ip '192.168.1.205'
option name 'rpi1.lan'
config host
option name 'sg105e'
option dns '1'
option mac '28:EE:52:xx:xx:xx'
option ip '192.168.1.3'
option leasetime '365d'
config host
option mac '00:E0:4C:xx:xx:xx'
option ip '192.168.1.206'
option leasetime '365d'
config odhcpd 'odhcpd'
option maindhcp '0'
option leasefile '/tmp/hosts/odhcpd'
option leasetrigger '/usr/sbin/odhcpd-update'
option loglevel '4'
Tue Jan 18 14:31:52 2022 user.notice kopijahe 2.2.2: Percobaan login. Interface: wwan. Hasil: Gagal Login (User Anda Sedang Aktif)
Nice to meet you too :D. This one is bash script (kopijahe) for login to wifi.id (public isp with landing page), i set manually to run at start up in local.rc with this command, since that was recommended by script author.
kopijahe -m 2 -z &
-m is multi-profile login, -z is send output to syslog, and & i think is to keep the script run in background. Should i remove that line in my local.rc?
not yet.. if it was working before... then best thing is to try the new/old build without ujail first...
Okay, i've already upgraded to 5.0.19-9 and CPU still stuck at 600mhz. Do i need to wait a while for the error to appear to post other logs?
yes... i've got some other stuff to try for cpu problem...
if this is 2020 Sep 3 then could be related to this
############ fwV:Sep3 vlV:000138a1
will need to see if anyone else is running that version and it's working for them or you could just try updating the bootloader...
but first I want to see...
the oom is very strange but may be related to ujail... but... there are 100 processes of
/bin/nft-qos-helper.sh
but that should not run if nft-qos is disabled (or at least it is suppost to just exit, not sure why I don't have the problem also... maybe it has some wan detection code in it or something that does not work on your system).... you can delete it if you want... if you get oom again (related to it)
(you can delete the first log upload)
will need to see if anyone else is running that version and it's working for them or you could just try updating the bootloader...
what command should i run to update the bootloader from openwrt (or just replace the file insde boot partition?) and which version i should use?
but first I want to see...
- dnsmasq errors go away
- oops/oom go away
- if your regular stuff works or not
i will inform you in case the problems goes away.
(you can delete the first log upload)
will do.
you can try this but I have not tested in a while and needs internet...
rpi4_eeprom.sh update # or rpi4_eeprom.sh updaterecommended
have been looking at that script... very cool...