I have the following problem.
After a temporary loss of connection (reboot of router, short loss of internet from isp side, etc.) the storage node does not reinitiate the connection to satellites. In the dashboard it shows as online, but last contact increases, network traffic is 0.
Only after restarting the node things are back to normal.
Is this expected? Shouldnt the node regain connectivity without a restart when internet is back?
I am running two nodes, one on windows and one on linux. Both have the same problem.
1 Like
In my case, after a loss of connection, the nodes seems recover without interruption, but I think the problem could depend of the type of installation (like Docker settings), your network settings, router, and so on. Be sure to keep your local IP addresses fixed in Windows/Linux settings + in your router to be sure nothing is changing when you’re reconnected.
You can also use free service like Uptime Robot and get notified each time your connection dropped, so you could eventually react and check at time.
I know also some users prefer auto-restart their nodes after a period of time to avoid this kind of issue, but it will hurt your downtime a bit and also your hard drive for loading the required data after a new run. So maybe could you check again all your settings and make the test by restarting your router manually to see if the problem still persist.
If your node comes online quickly after going down, the satellites usually haven’t noticed it was offline yet and traffic will continue right away. However, if the satellites notice your node is offline, they will exclude them for selection until it’s spotted back online again. This is why after longer down time it takes a while until you get traffic again, because the satellite would first have to do an uptime check to know you’re back online.
A restart helps because the node checks in with the satellites itself and gets marked as online again right away. But it should get traffic again automatically, if you give it some time as well.
1 Like
Thank you both for your answers. This explains it then. Next time I will then just restart the nodes if i realise the network interruption, otherwise ill let the node take care of it.