New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We鈥檒l occasionally send you account related emails.
Already on GitHub? Sign in to your account
socket connection refused after sleeping with rdm server left running #15
Comments
@gnfisher Thanks for the report! Glad you're finding it useful, too. That's really strange behavior, I haven't run into that yet! I'd happily accept a PR that fixes the broken behavior. I wonder if there are any logs, rdm or otherwise that might help pinpoint what's killing the server, or why it's dying? Maybe something in |
Thanks, I've checked and the log is empty at the moment. Perhaps my manually removing the socket and restarting rdm server wiped it? Next time it happens I'll check the logs again. |
I've had the issue occur again. The logs are empty. If I try to stop the server, I do get a log:
It seems like the |
I can reproduce this if I close the terminal I ran If I run TLDR rdm works fine as it is. Perhaps an upgrade would be to daemonize the process with launchd/systemd? Or alternatively to listen for SIGHUP and shut itself down. If that is interesting, I'd be interested in poking around at how that might be solved. Thanks again for creating this! |
I think @brasic mentioned wanting something similar. If we could build a command that would allow rdm to easily inject itself into launchd (maybe homebrew can do this?) I'd be supportive of adding it to the repo/binary. I had looked into doing something like |
I fear that using launchd might have essentially the same user-namespace problem, but it's worth a shot. I'm interested in taking a crack at this since RDM is something I use every day and not having to nohup it would be nice. (Don't let this stop anyone else from working on this though, it would not be the first thing on my open source backlog 馃槄) |
Coming back to this for a second, we could also handle I could imagine adding a |
It can't hurt to add the option to ignore HUP or do it by default but launchctl is working great for me. The PR got a little large so I extracted most of it into a new library that should be pretty easy to reuse: https://github.com/brasic/launchd I should bring it out of draft pretty soon: #16 |
OK, found the time to polish that PR up and it's ready to review. I've been using it for a few weeks and have seen no issues managing rdm with launchd instead of manually nohupping it. |
I think this was fixed via #16, so closing for now |
馃憢 Hello! This utility greatly improves the quality of my life, thanks for making it and sharing it!
I've run into a rough edge that I'd thought I'd capture in an issue.
While having a running rdm server in the background, I close my laptop. When I open the laptop again, the rdm server is unreachable.
In a codespace, I'll try to view a PR on github and get the following:
If I check on my host machine (macOS) and try to to stop the running server, I get the following error:
My current solution is to
rm -rf <path to the socket>
and restart the server. It would be great if the server could persist, or if it stopped itself when the machine went into a state where the socket gets "lost" (I know almost nothing about unix sockets or what might be happening).If you're open to it, I might try to put a PR together in my spare time as a project to start learning go, but I will be very slow to do this 馃暫
The text was updated successfully, but these errors were encountered: