Happens for me when I open the statistics in LuCI. Try:
/etc/init.d/rpcd restartYou made that conclusion based on these two command failed? We tested it and rainbow intensity works even with these two rows.
Can confirm that rainbow still have some issue. I configured rainbow in config file like this
config led 'all'
option status 'auto'
option color 'green'
option brightness '005'
config led 'power'
option status 'auto'
option color 'red'
config led 'lan'
option status 'auto'
option color 'green'
config led 'wan'
option status 'auto'
option color 'blue'
config led 'wlan_1'
option status 'auto'
option color 'green'
config led 'wlan_2'
option status 'auto'
option color 'green'
It works in version 5.4.4 but rainbow config is completely ignored in version 6.0.1.
Same goes for setting ledās intensity.
Did you intruduced breaking changes in rainbow command or in its config file?
2 Likes
As I mentioned earlier thereās nothing in /var/log/lighttpd, but today I discovered yet another thing. If I stop lighttpd and try to start it from /usr/sbin, it complains :80 is in use:
root@turris:~# service lighttpd status
running
root@turris:~# /etc/init.d/lighttpd stop
root@turris:~# service lighttpd status
Command failed: Not found
Failed to parse json data: unexpected end of data
inactive
root@turris:~# /etc/init.d/lighttpd status
Command failed: Not found
Failed to parse json data: unexpected end of data
inactive
root@turris:~# /usr/sbin/lighttpd -f /etc/lighttpd/lighttpd.conf
2022-10-22 11:26:06: (../src/network.c.540) can't bind to socket: 192.168.1.1:80: Address in use
If I change lighttpd port to e.g. 81 and start the service, I can open 192.168.1.1:81, I can choose reForis/Luci/pakon. I tried reForis, it however complains about the connection being insecure when I want to log in. If I click āClick here to redirect to secure protocol.ā, I get
Secure Connection Failed
An error occurred during a connection to 192.168.1.1:81. SSL received a record that exceeded the maximum permissible length.
Error code: SSL_ERROR_RX_RECORD_TOO_LONG
Itās as if something was hogging port 80, but so far I could not figure out what it isā¦ Iām not running nginx or anything else.
Edit: after changing lighttpd port to 81 Iām now able to connect even just to 192.168.1.1 over HTTP without specifying any port. No idea what is going on here 
1 Like
Right. NOT amused here. This updated completely ruined my MOX classic. Not sure why it went to TOS6 since that one is on HBS. Cant get it working again. Have now my old synology router to type thisā¦
The LED blinks 2 timesā¦
i do get the internal wizzard page, and after that a server 500 error
edit, breathing again. works now on a cable modem, but not yet with a VLAN PPPOEā¦
Okā¦PPPoe working , now open vpn is making troubleā¦
Options error: Unrecognized option or missing or extra parameter(s) in /etc/openvpn/turristhuis.ovpn:289: remote-cert-tls (2.5.3)
funny enough the generated .ovpn cfg works on a windows machine here,
1 Like
I tried this already. Didnāt help.
Web interface was unavailable after upgrade: āConnection refusedā, no ports was listening.
I used athmoss method and discovered:
root@turris:/etc/lighttpd# /usr/sbin/lighttpd -f /etc/lighttpd/lighttpd.conf
Undefined config variable: var.home_dir
...: (../src/configfile.c.2162) source: /etc/lighttpd/conf.d/30-rrdtool.conf line: 18 pos: 39 parser failed somehow near here: +
...: (../src/configfile.c.2162) source: /etc/lighttpd/lighttpd.conf line: 35 pos: 0 parser failed somehow near here: (EOL)
I replaced the home_dir with :
"/var/www/lighttpd.rrd"
MOX classic, HBK branch, .5 GB, 2x WiFi, simple config, all seems OK. (*)(%)
TO 2016, HBS branch, 2 GB, 2x WiFi, HaaS, RIPE Atlas, Sentinel, lxc, simple config, all seems OK. (*) (#)
(*) still some troubles regarding raiinbow, needs troubleshooting.
(%) except SDIO WiFi.
(#) firewall log in Sentinel failing.
1 Like
I realized that we are now released to HBS TOS 6 only just now today, but being on TOS 6 HBL and then HBK it is not much difference as I already had TOS 6 for more than half a year on both omnia and turris 1x.
I am so happy that we finally moved forward to brighter future a.k. to be more close to master openwrt development even not really but to track progress on master with some defined delay to allow stable release will be definitely good thing as well leaving the past of TOS 5 behind and the progress is unstoppable.
Reading release notes kind of remind me Roy from IT crowd answering phone and yelling āDid you tried to turn it off and on again ?ā it is bit lol 
but probably team had quite hot and stresfull moments on release day.
Anyway I switched my routers to HBS but now feel a bit nostalgic as I may probably be missing that adrenalin not knowing what colors will rainbow led shine in the morning after nightly updates and restart during night. Also not knowing if I would be able to connect to internet and have wifi to work on home office that day. 
Now we could laugh about those days on development build and I really feel nostalgic to the point that my life will became boring on HBS so I consider to give a ride to TOS 7 HBL again to help.
5 Likes
I do not have RRDTool on my end.
I however made some progress with the troubleshooting. It seems TOS6 does not like my lighttpd being bound to a specific IP:
server.bind = "192.168.1.1"
Commenting this out and rebooting Omnia seems to solve the web interfaces unreachability. Not sure why binding the web server to a specific IP is a problem though.
I also have this, from /var/log messages it seems sentinel cannot renew a certificate:
root@turris:~# grep sentinel /var/log/messages
Oct 22 14:26:15 turris sentinel-dynfw-client[2326]: 2022-10-22 16:26:15,451 - WARNING - Unable to renew certificate (another try after 8 sec): [Errno -3] Try again
Oct 22 14:26:23 turris sentinel-fwlogs[5501]: ERROR: Packet handling failed: Resource temporarily unavailable
Oct 22 14:26:36 turris sentinel-fwlogs[7822]: ERROR: Packet handling failed: Resource temporarily unavailable
Oct 22 14:26:58 turris sentinel-fwlogs[7826]: ERROR: Packet handling failed: Resource temporarily unavailable
Oct 22 14:27:06 turris sentinel-fwlogs[7939]: ERROR: Packet handling failed: Resource temporarily unavailable
Oct 22 14:27:30 turris sentinel-fwlogs[7942]: ERROR: Packet handling failed: Resource temporarily unavailable
Oct 22 14:27:46 turris sentinel-fwlogs[7966]: ERROR: Packet handling failed: Resource temporarily unavailable
Oct 22 14:27:46 turris procd: Instance sentinel-fwlogs::instance1 s in a crash loop 6 crashes, 10 seconds since last crash
Does someone have an idea how to fix this? Thank you all for your help!
ss -tlp | grep :www should answer that.
Hello,
Sentinel firewall in Sentinel failing too :
Oct 23 07:00:59 turris sentinel-dynfw-client[4014]: ipset v7.6: Error in line 1: Element cannot be deleted from the set: itās not added
Oct 23 07:00:59 turris sentinel-dynfw-client[4014]: 2022-10-23 09:00:59,386 - WARNING - Error running ipset command: return code 1.
Oct 23 09:01:19 turris sentinel: INFO [certgen.action_spec_init:89] Valid certificate found
Oct 23 07:01:31 turris sentinel-fwlogs[25125]: ERROR: Packet handling failed: Resource temporarily unavailable
Oct 23 07:01:54 turris sentinel-fwlogs[25776]: ERROR: Packet handling failed: Resource temporarily unavailable
oh boyā¦my MOX syslog is totally cluttered with this one : Failed to send to ff02::1%guest_turris@br-guest-turris (Address not available)
and this : dnsmasq-dhcp[29052]: read /etc/ethers - 0 addresses
and this : turris kernel: [38388.737264] page_pool_release_retry() stalled pool shutdown 4 inflight 3438 sec
anyone?
1 Like
I have the same: Failed to send to ff02::1%guest_turris@br-guest-turris (Address not available)
Additionally to my lighttp issue mentioned in post #3 I realized that āFirewall logsā failed in ReForis - Sentinel - Overview.
In TurrisOS 6.0 it helped to deactivate Firewall logs and activate them again. Now it does not just not help but deactivating is impossible.
My device is Turris Omnia 1GB (blue, rev. v0).
Both things worked fine till TurrisOS 5.4.4
Might have to do with this in reforis> guest network ? : This network currently doesnāt contain any devices. The changes you make here will become fully functional after you assign a network interface to this network.
Apperantly this is used for wifi guest account, and i do not have that running?
nOOb hereā¦if someone can check this i t would be nice!
I just got around to upgrading my Indiegogo 2GB Turris.
Still poking around, but it looks really good. LuCI seems a little finicky, but I think itās just short timeouts for the interface, as in how long it keeps me logged in before it decides I need to re-log in.
It even kept all my funky DNS settings.
Thank you Turris team.
My simpler Omnia upgraded from 5.4.3 to 6.0.1 without issues today. Just one - DNS forwarding stopped working and I had to switch it off (Iāve already reported it).
The process was, however, still not ideal. I got disconnected from wifi quite early on in the process. Then I watched progress connected via VPN from WAN over mobile data - but only in reforis, as both ssh and luci were dead. Then I received the ārestart is neededā notification, but before I could do anything, the router was rebooted automatically (I guess thatās what āresolvesā the open firewall issue, so ok). After the first restart, wifi came back on, but was very unstable, connected and disconnected every minute. After a second reboot, everything works. Iām just surprised that no schnapshot was created after the update (only the pre- snapshot was). Updater finished with exit code 64, so is that a reason why the schnappshot was not created?
I accidentally discovered that the reForis - Administration - Snapshots module is broken.
Schnapps in the terminal works, the reForis menu does not respond - after some time the API reports āTimeout error occurred.ā. I didnāt find anything suspicious in the log.
Restarting the router wonāt change anything.
Updated from 5.X directly to 6.0.1 and it worked perfectly.
- Wifi was up and running after reboot
- Old VLAN configuration was still working (I will migrate to new configuration style in the future)
- LuCi, Reforis and SSH were working after reboot (I only had to reactivate my custom ssl configuration for lighttpd)
- LEDs were working (never changed settings)
Thank you very much for your work!
3 Likes