Replies: 2 comments 1 reply
|
Is newt fully restarting or is it just the websocket bouncing? If its
fully restarting could you post more of the logs around the restart as
something would be triggering it to crash or restart.
|
0 replies
|
Here are some more logs, as requested, on the older versions of NEWT i was on, it seemed to be a complete restart, would show the whole startup status, more verbosity, but with the lastest version, here is the logs surrounding the even on the NEWT VM: All of these seem to correspond to random network scans, and TCP attempts via some TCP rules i have configured in Pangolin, as I mentioned above. As for the Pangolin host, I am waiting for another instance to happen so i can capture the full log, haven't seen any in the last 24 hours. |
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Good afternoon, all,
I have been banging my head on the wall with Pangolin, and the NEWT tunnel system. I have been really liking the setup, and currently I am running the server on a VPS connection, with about ~20ms of latency to my served network. I am on the latest version of Pangolin, running in docker, and I have the latest version of NEWT running in a VM on my DMZ network. The issue is seemingly randomly the NEWT connection closes, seemingly gracefully and then reconnects. This would not be an issue, but when using some of the services I am hosting, this causes a drop of connection. Barring TCP connection resets from outside network scans, I have no errors in the logs, just the following:
Pangolin server side:
Newt Server side:
The time discrepancies are simply because one log is in EST and the other is in UTC.
Running iperf3 tests between the two device's i have not seen any latency increases or packet loss during these times, and the outage time is almost inconceivable besides 1-3 seconds of increased load times for sites, and RTC connections dropped.
System resources are also not an issue, both of machines are consistently under 30% resources.
Does anyone have any ideas what could be causing this? The timing is random, sometimes once a day, sometimes once an hour, could the tcp connection resets cause this?
Let me know if there is any other information that would help investigate this, as I would be happy to provide it. I could not find much about this, besides some older posts seeing a similar issue due to packet loss, and i will also add that all of the endpoints that are having issues are on the same machine as the Pangolin tunnel endpoint.
Thank you!
All reactions