Updater selhal - Turris Omnia 5.1.4 HBS

Již podruhé tato notifikace:
Updater selhal: runtime: [string “requests”]:417: [string “utils”]:420: Getting URI (https://repo.turris.cz/hbs/omnia/lists/pkglists/datacollect.lua) failed: Couldn’t resolve host ‘repo.turris.cz’

Z připojeného počítače je stránka běžně dostupná. Připojeno k UPC.

díky za rady

datacollect.lua this is the key to the problem. Which versions are you trying to upgrade from? It can happen when you skip the branch (HBS, HBT, etc.) and then go back to a version that doesn’t include that set of packages in re (Foris). One method can be to rollback from a version of Turris OS to the same branch you want to upgrade to and then upgrade (for example, roll back to HBS and then upgrade to HBS).

This is a completely new Turris.

What is your current version of Turris OS? Omnia? MOX? Shield?

Are you sure? I think the issue is elsewhere.

failed: Couldn’t resolve host ‘repo.turris.cz’

Would be good to know when did you receive this notification.

I stated in the header :slight_smile:

Today. The page is normally available from a connected computer

In fact that lua script is present in the hbs repository. But I had a similar problem going back from HBL to HBS. There was the script for Ripe Atlas which is featured in reForis for HBL, but not for HBS.

Problems with DNS? Test from the terminal in Turris OS: nslookup repo.turris.cz.

Looking to add some packages? I don’t think there are any updates since 5.1.4.

The PC can use its own various DNS according to the settings on the PC. Then try from the router terminal.

PING proxy.turris.cz (217.31.192.69) 56(84) bytes of data.
64 bytes from proxy.turris.cz (217.31.192.69): icmp_req=1 ttl=58 time=6.41 ms
64 bytes from proxy.turris.cz (217.31.192.69): icmp_req=2 ttl=58 time=7.02 ms

Name: repo.turris.cz
repo.turris.cz canonical name = proxy.turris.cz
Name: proxy.turris.cz
Address 1: 217.31.192.69
repo.turris.cz canonical name = proxy.turris.cz
Address 2: 2001:1488:ac15:ff80::69

This makes us understand that the host is reachable. You can also get wget https://repo.turris.cz/hbs/omnia/lists/pkglists/datacollect.lua, again from router. It will download the script, which once downloaded, you can safely delete. So we are sure that the resource is available to Turris OS and therefore the problem moves elsewhere. Could you also tell us which packages have you installed from reForis?
Take a look at too:
https://forum.test.turris.cz/t/getting-uri-failed-for-luci-controls/13575

Sběr datSoftware for participation in data collection and distributed adaptive firewall.
Průzkum využitíSběr dat o užívání routeru (nainstalované balíčky, typ připojení k internetu a jiné).
Dynamický FirewallPřidává pravidla pro firewall k blokování útočníků zachycených pomocí Turris sítě pro sběr dat.
Záznamy z Firewall

Try disabling everything from the packages menu on reForis and then try again to start the update. Or, if you don’t have any particular configurations, you can flash the router with the stock firmware again. But I still don’t understand what you are trying to update / install. The more information you give, the easier it is to help you.

I don’t update anything manually. I have auto-update turned on. I’m not an expert so I don’t do anything out of date.

In what context do you get that error? Where do you see that message?

ReForis interface - notification and by email

The point you are probably missing is that unless you are getting this notification every four hours then you do not have problem with updates but rather with network in specific moments.
Updater is run every four hours. This notification is send if in that run updater fails to contact our servers.
If the notification is sent only sporadically then the culprit is going to be Internet outage. Check if notification does not match with ISP service outage.

1 Like