Timeout on rssh
chikamobina opened this issue · comments
Hi. When I use "ssh -D 5005 -J server hostname", then I start to use this port as a proxy, after a while rssh kills my session by timeout, is there any way to remove or change this? I start e.g. a network scan, some long running exploits and after a while rssh just kills me and everything hangs up.
This was a long time ago, in your old updates, then it stopped killing the session, if I use the port, it lived as long as the port is used, but now it's like before, it kills the session just by timeout, both linux and windows.
Hm. I was having an issue like this but assumed it was flakey networking.
I haven't made any changes to that aspect of rssh so not sure how its decided to change.
I'll do some regression testing.
You can disable timeouts entirely with --timeout 0
However I think this problem might be different than just the timeout mechanism.
Got it. Yeah, I also thought it was a connection at first, but then some update made it stable, but now it's throwing out again.
Very annoying, I'll see if I can get a situation that replicates it. But otherwise I may temporarily just remove the timeout feature and see if that helps
Cool I can replicate it.
For my future reference, its not just jump (-J) hosting, its all connections.
Additional information. After the first disconnection it seems to become stable.
@chikamobina Are you using the clients in foreground mode? (--foreground) as it seems I can only get this behavior to happen in that mode?
No, I use it without it. But I use garble all the time. For Windows I have a .dll.
Glad to hear the windows dll works, I honestly havent tested it since I made it.
Okay, I've removed the code that was causing this issue (at least the foreground crashing bit). Am doing a release as this is quite not good in terms of bugginess.
Tell me if you get the weird disconnections still please.
@chikamobina can you try v2.1.1 and see if you get the same disconnects on your backgrounded clients?
I cant replicate backgrounded clients dying, I've identified the change that causes foreground clients to die (although not why it causes death)
I was just testing this on the last version
Okay, I've removed the code that was causing this issue (at least the foreground crashing bit). Am doing a release as this is quite not good in terms of bugginess.
Tell me if you get the weird disconnections still please.
I'll let you know when I test it, thank you.
As I have been using this myself without issue Im going to close this. Feel free to re-open it if you are still experiencing issues.
Okay, I have tested and this problem observed in my lab with the firewall only, in other places everything works fine, thank you.