TurrisOS 3.9.1 is out!

Community office
21

I’m not sure if it’s directly connected to the 3.9.1 version but today I noticed Netmetr configuration in Foris was reset. I had configured autostart every second hour. Version 3.9.1 was installed at 13:40:40 yesterday (according to schnapps list) on my Turris and I rebooted it about half an hour later. The last Netmetr report was 13:05:57 so I think the times correlate.

Is it a known issue? Should I check that other configs weren’t reset too?

1 Like
22

Same error in the log, Turris 1.1.

Output from updater-ng-localrepo.postinst:
Traceback (most recent call last):
  File "/usr/bin/localrepo", line 541, in <module>
    main()
  File "/usr/bin/localrepo", line 537, in main
    cmd_check(args)
  File "/usr/bin/localrepo", line 429, in cmd_check
    gen_updater_config()  # Just to be sure, regenerate
  File "/usr/bin/localrepo", line 48, in gen_updater_config
    with open(os.path.join(REPO_PATH, "localrepo.lua"), "w") as file:
IOError: [Errno 2] No such file or directory: '/usr/share/updater/localrepo/localrepo.lua'
23

Hello,

updated from 3.8.4. So far quite OK. I dont like this newly appears on Omnia (??? I never checked Collect Data):

2018-01-03 16:00:01 info /usr/sbin/cron[4167]: (root) CMD (/usr/share/server-uplink/contract_valid.sh)
2018-01-03 16:00:01 info /usr/sbin/cron[4168]: (root) CMD (/usr/share/server-uplink/registration_code.sh)

Also on Foris, DNSSEC check dont work anymore, it freezes on “Loading…”

24

Experience the same issue on my Turris 1.1. OS v3.9.1. Most of the time DNSSEC check freeze, but not always.

25

Most probably it just takes long time. Other tests are fast because they just try something and if you have working connection then it works really fast. But with DNSSEC it is trying to intentionally resolve bad domains. Because of that resolver tries really hard to resolve it but in the end it fails but we have to wait on those timeouts. It’s same as if you would try to ping nonexistent host, ping also waits very long time. There is no way it knows that there is no such host unless it waits it out because otherwise it could be just slow network.

26

Well, that makes sense. But I mentioned that because:

  1. freeze happens also for other checks in my case
  2. I remember that checks used to be faster in previous releases of TurrisOS.
27

IN that case you can run check_connection from terminal to see what is happening.

28

I put it wrongly. I meant, whole connetion check frozen. That wasnt happened before, the checks was pretty quick in previous foris. DNSSEC test always went quick and also on NIC page it goes quickly and well, so most propably DNSSEC itself isnt an issue. Unfortunatelly I didnt get anything on logs.

check_connection runs with all OK result within a few seconds.

https://serving.photos.photobox.com/36153722a8551a6155b6e7e6367b11084c0bd6cd4730c08f61649930084858406225ca14.jpg

29

I checked it in two browsers, and client also dont seems to be root cause.

30

The error is on the router’s side, and I personally manifest myself when the Foris is redirected to another port. The colleague who is in charge of this part already knows about the problems. Now it is up to an update to resolve this problem.

2 Likes