Hi,
I keep getting this error in update notifications:
Error notifications
Updater failed:
runtime: [string “requests”]:160: attempt to index global ‘string’ (a nil value)
Any pointers for a solution?
thanks
Marcel
Hi,
I keep getting this error in update notifications:
Updater failed:
runtime: [string “requests”]:160: attempt to index global ‘string’ (a nil value)
Any pointers for a solution?
thanks
Marcel
Hey,
the same happens here on one of our Turris which is still on 3.7.1. What can we do here? I am afraid to upgrade to 3.8 because that caused those nasty DNS issues our other Turris.
Thanks
Andre
This is bug fixed in 3.7.2. Could you please people update at least to fixups? Now you have to update updater by hand:
wget https://repo.turris.cz/omnia/packages/turrispackages/updater-ng_59.3-1_mvebu.ipk
opkg-trans -a updater-ng_59.3-1_mvebu.ipk -r opkg-trans
updater.sh
(This exact form is just for Turris Omnia and will work only until 3.8.2 probably)
Thanks a lot from my side as well. It did solve the problem indeed. However, we had updates turned on but the Turris was turned off for a while so likely missed some updates. Maybe that cause the troubles.
Hi,
My router has been off line in the cupboard for a while. However, I’m now ready to start using it but I’m getting the same
Updater failed:
runtime: [string "requests"]:160: attempt to index global 'string' (a nil value)
I tried the wget solution in the custom command, but that failed with
Command failed (Code: 2048)
What is the current fix for this please? The router states: foris version: 94.2 (3.7.1)
Thanks.
That solution was just applicable only for limited time as address provided changes with new versions of updater. Here is solution using archive version:
wget https://repo.turris.cz/archive/omnia/3.7.5/packages/turrispackages/opkg-trans_58.4.7-1_mvebu.ipk
opkg-trans -a opkg-trans_58.4.7-1_mvebu.ipk
updater.sh
Hi,
Thanks. Ran that in Custom Commands and it reported success. I left it a short while then rebooted. I’m not getting anymore messages regarding updater failing, but I note that had 3.7.5 in the command and I’m still seeing 3.7.1.
Any cause for concern or should that change when it does updates tonight?
Thanks again.
The point is to update updater to newer version then the one from Turris OS 3.7.1 that is why there is 3.7.5 in command. After running updater.sh you should end up with Turris OS 3.8.4.
The wget reports working fine, but updater.sh gives me
e[31;1mDIEe[0m:
runtime: [string “requests”]:160: attempt to index global ‘string’ (a nil value)
Are you sure that you have executed all three comands? It seems to me that you didn’t managed to update updater at all.
Hi,
Yes, my mistake. Failed to notice that first wget failed and subsequent calls were adding numbers to the end, thus opkg was also failing. Fixed now. Sitting at 3.8.4. However now I have to turn DNSSEC off or DNS fails, but that is a discussion for another time. Thanks again.
On my Turris I now get:
root@turris:~# opkg-trans -a opkg-trans_58.4.7-1_mvebu.ipk
/usr/bin/opkg-trans: line 1: syntax error: unexpected word (expecting “)”)
Device Turris - RTRS02
Serial number 38654707564
Turris OS version 3.7.1
Kernel version 3.18.58-1ac1f353df888f6a08f1362f0631fbf3-0
What can I do to update to 3.9?
Thank You
-Rick
Just do factory reset if you haven’t done a usb reflash in past and if you did then do usb reflash again. Because if referenced solution doesn’t work then I really won’t be looking in to it why now.
Ok. Did factory refresh and Turris is up to 3.7.3 (kernel 3.18.63-1ac1f353df888f6a08f1362f0631fbf3-0) at this point but I get the following on the Turris Home Page:
“Error from 2017/12/19 22:55:31
Migration to updater-ng failed: Unsuccessful creation of list of additional installed packages: Unknown error”
What should I do? Should I ssh to root@192.168.1.1 and execute:
I just dont want to screw this up.
Thank You,
-Rick
We have documentation for this (because you have Turris 1.x you should be Czech) https://doc.turris.cz/doc/cs/howto/updater_migration
Possibly it might be known error in updater’s reporting. But it can also mean some real failure during post-migration update so follow documentation.
You shouldn’t have to run commands you have pasted here. Those were for versions older then 3.7.3. We released fix in 3.7.3 and commands you pasted were just workaround if somebody stayed on 3.7 version before third fixup so that he would be able to update to 3.8. You don’t have to be doing this. This might even break migration if you haven’t finished it before you run those commands.
Thank you for the prompt response.
I did go to the Czech page (and used google xlate) and tried to update but get the same error:
“Error from 2017/12/22 02:03:36
Migration to updater-ng failed: Unsuccessful creation of list of additional installed packages: Unknown error”
This was after I followed the instructions from the page and got:
$ ssh root@192.168.1.1
root@192.168.1.1’s password:
BusyBox v1.25.1 (2017-08-01 17:18:39 CEST) built-in shell (ash)
|__ || | | || __ \ | __ \ | _| / ____|
| | | | | || |) || |) | | | | (__
| | | | | || _ / | _ / | | ___ \
| | | || || | \ \ | | \ \ | | ) |
|| _/ || _|| _|||__/
root@turris:~# get-api-crl
root@turris:~# updater-migrate.sh
Wait. You are one of the people who received Turris 1.x from Ondřej Filip? Or how is possible that you are not speaking Czech?
Anyway as I wrote. Because of problems with updater and migration race condition it reports unknown error
instead of real problem. In your case there seems to be something wrong with you internet connection.
I tried this link and it works for me (of course it’s self signed so you have to add exception in browser). So I don’t think that problem is on our servers but rather in your internet connection.
Can you please try to run following commands?
ping $(ip route | awk '/^default/{print $3}')
ping 217.31.205.50
nslookup api.turris.cz
ping api.turris.cz
ping -6 api.turris.cz
And also please run connection test in DNS tab in Foris (we interface).
I did see error re SSL. How do I tell Turris to ignore SSL? Should I edit “updater-migrate.sh” ?
The ping does work.
root@turris:~# ping $(ip route | awk ‘/^default/{print $3}’)
PING 192.168.1.1 (192.168.1.1): 56 data bytes
64 bytes from 192.168.1.1: seq=0 ttl=64 time=0.375 ms
64 bytes from 192.168.1.1: seq=1 ttl=64 time=0.268 ms
64 bytes from 192.168.1.1: seq=2 ttl=64 time=0.285 ms
root@turris:~# nslookup api.turris.cz
Server: 127.0.0.1
Address 1: 127.0.0.1 localhost
Name: api.turris.cz
Address 1: 2001:1488:ac15:ff80::101 api.turris.cz
Address 2: 217.31.192.101 api.turris.cz
root@turris:~# ping api.turris.cz
PING api.turris.cz (217.31.192.101): 56 data bytes
64 bytes from 217.31.192.101: seq=0 ttl=39 time=190.325 ms
64 bytes from 217.31.192.101: seq=1 ttl=39 time=180.130 ms
^C
— api.turris.cz ping statistics —
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 180.130/185.227/190.325 ms
root@turris:~# ping -6 api.turris.cz
PING api.turris.cz (2001:1488:ac15:ff80::101): 56 data bytes
ping: sendto: Permission denied
root@turris:~# updater-migrate.sh
WARN:Config file /etc/config/updater modified by the user. Backing up the new one into /etc/config/updater-opkg
Output from syslog-ng3.postinst:
Command failed: Not found
line not found
line not found
line not found
ERROR:
unreachable: https://repo.turris.cz/turris/lists/base.lua: SSL certificate problem: unable to get local issuer certificate
INFO:Executing postupdate hooks…
Can you please check if you have file /etc/ssl/certs/DST_Root_CA_X3.crt
on your router?
It seems to me that you don’t if so, you can try to do this:
wget --no-check-certificate https://repo.turris.cz/turris/packages/base/ca-certificates_20161130_mpc85xx.ipk
opkg-trans -a ca-certificates_20161130_mpc85xx.ipk
That will manually update/install CA certificates that you are seems to be missing.