Re: Socket handle leak and active handle warning with Vibe-D
On Monday, 15 January 2024 at 22:19:56 UTC, Steven Schveighoffer wrote: You may have to do the same thing I did with redis: https://github.com/vibe-d/vibe.d/pull/2372 Good luck! I would also say, I don't know why Windows doesn't do the same trace info debug thing, except that probably whomever added it didn't care about windows. Many thanks for the assistance.
Re: Socket handle leak and active handle warning with Vibe-D
On Monday, 15 January 2024 at 18:40:00 UTC, bomat wrote: Sorry, I probably should have mentioned I was on Windows. For testing it under Linux I commented out the call to `connectMongoDB`, since I don't have it installed there - and the warning went away. Interesting, I did not suspect that as the source of the problem at all. :P I'm now looking into how to clean up MongoDB connections properly, but don't see anything besides `cleanupConnections()` (which I'm already calling without any effect). Maybe I need to initialize it differently... I'll experiment a bit. You may have to do the same thing I did with redis: https://github.com/vibe-d/vibe.d/pull/2372 Good luck! I would also say, I don't know why Windows doesn't do the same trace info debug thing, except that probably whomever added it didn't care about windows. -Steve
Re: Socket handle leak and active handle warning with Vibe-D
On Monday, 15 January 2024 at 17:45:16 UTC, Steven Schveighoffer wrote: Which driver are you using? In the posix driver, it should mention (and use) the debug flag `EventCoreLeakTrace`. https://github.com/vibe-d/eventcore/blob/7fa0a15fa541c3fcf65640ee332fd3a09c34730c/source/eventcore/drivers/posix/driver.d#L130 I didn't realize this wasn't across the board... -Steve Sorry, I probably should have mentioned I was on Windows. For testing it under Linux I commented out the call to `connectMongoDB`, since I don't have it installed there - and the warning went away. Interesting, I did not suspect that as the source of the problem at all. :P I'm now looking into how to clean up MongoDB connections properly, but don't see anything besides `cleanupConnections()` (which I'm already calling without any effect). Maybe I need to initialize it differently... I'll experiment a bit.
Re: Socket handle leak and active handle warning with Vibe-D
On Monday, 15 January 2024 at 17:24:40 UTC, bomat wrote: On Sunday, 14 January 2024 at 20:36:44 UTC, Steven Schveighoffer wrote: There should be a version you can enable that tells you where that socket handle was allocated. That might give you a further clue as to why it's not closed when the system shuts down. I think the program tells you which version to enable when this happens, but if not, let me know and I'll find it. Thanks for the response, Steve. Hmmm, not sure if I'm missing something, but this is all the output I get from the program: ``` [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP requests on ::1:8080 [main( ) INF] Stopped to listen for HTTP requests on 127.0.0.1:8080 Warning: 1 socket handles leaked at driver shutdown. Warning: 1 socket handles leaked at driver shutdown. ``` Unless there's some switch to make it more verbose? Which driver are you using? In the posix driver, it should mention (and use) the debug flag `EventCoreLeakTrace`. https://github.com/vibe-d/eventcore/blob/7fa0a15fa541c3fcf65640ee332fd3a09c34730c/source/eventcore/drivers/posix/driver.d#L130 I didn't realize this wasn't across the board... -Steve
Re: Socket handle leak and active handle warning with Vibe-D
On Sunday, 14 January 2024 at 20:36:44 UTC, Steven Schveighoffer wrote: There should be a version you can enable that tells you where that socket handle was allocated. That might give you a further clue as to why it's not closed when the system shuts down. I think the program tells you which version to enable when this happens, but if not, let me know and I'll find it. -Steve Thanks for the response, Steve. Hmmm, not sure if I'm missing something, but this is all the output I get from the program: ``` [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP requests on ::1:8080 [main( ) INF] Stopped to listen for HTTP requests on 127.0.0.1:8080 Warning: 1 socket handles leaked at driver shutdown. Warning: 1 socket handles leaked at driver shutdown. ``` Unless there's some switch to make it more verbose?
Re: Socket handle leak and active handle warning with Vibe-D
On Saturday, 13 January 2024 at 20:49:54 UTC, bomat wrote: I am still getting this in 2024 and vibe.d 0.9.7: ``` Warning: 1 socket handles leaked at driver shutdown. ``` I was wondering if maybe someone has new info on this... There should be a version you can enable that tells you where that socket handle was allocated. That might give you a further clue as to why it's not closed when the system shuts down. I think the program tells you which version to enable when this happens, but if not, let me know and I'll find it. -Steve
Re: Socket handle leak and active handle warning with Vibe-D
I am still getting this in 2024 and vibe.d 0.9.7: ``` Warning: 1 socket handles leaked at driver shutdown. ``` I was wondering if maybe someone has new info on this...
Re: Socket handle leak and active handle warning with Vibe-D
On Friday, 1 January 2021 at 22:07:28 UTC, Selim Ozel wrote: [2] https://github.com/SelimOzel/vibe_noLeaks I don't see anything abnormal in this code though. Will trying it later today myself. Could you show the actual socket code causing that leak?
Re: Socket handle leak and active handle warning with Vibe-D
On Tuesday, 5 January 2021 at 21:12:01 UTC, Selim Ozel wrote: On Monday, 4 January 2021 at 17:17:10 UTC, Steven Schveighoffer wrote: On 1/2/21 12:52 PM, Selim Ozel wrote: After logging into to 127.0.0.1 for a single time in my browser, if I do a ctrl+c it still leaks two socket handles. With connection Windows 10: Running .\vibe_noleaks.exe [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. [() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP^ requests on C::1:8080 [main( ) INFC:\Software\vibe_noLeaks>] Stopped to listen for HTTP requests on 127.0.0.1:8080 Warning: 2 socket handles leaked at driver shutdown. Warning: 2 socket handles leaked at driver shutdown. This is normal. The server uses keepalive connections, so that in case any more requests arrive on the same connection, the initial connection setup does not need to be established. Well, at least that is what I think is happening. If you want a few seconds (I think 5 or so), then you won't get these. It would be good if vibe-d could provide a way to shut down any keepalive connections when the server is shutting down. -Steve That's interesting. I actually started to dive deeper into those and tried to pinpoint the lines of code that result in additional open sockets upon new http connections; although my understanding of vibe-d is a bit too low at this point to figure out what's exactly happening. I think from a user perspective having something a bit friendlier on warning side would be helpful. Bausshf built a wrapper around vibe.d sockets called cheetah. Haven't used it myself but it looks more abstracted than than actual vibe.d code. Can't speak of the quality since I haven't used it myself. Also it's got both a server and client abstraction but only the server example is shown. Still not low-level though. https://github.com/bausshf/cheetah/wiki/Simple-Socket-Server-Example I've used the very vibe.d example you referenced in the past and didn't have any trouble with it. So it could possibly be a bug or regression. If you've have wsl2 installed on Windows, try running on the Linux side too. Do you have any suggestions in mind towards that? I have a bit of time this week and I could take a stab at it. B
Re: Socket handle leak and active handle warning with Vibe-D
On Monday, 4 January 2021 at 17:17:10 UTC, Steven Schveighoffer wrote: On 1/2/21 12:52 PM, Selim Ozel wrote: After logging into to 127.0.0.1 for a single time in my browser, if I do a ctrl+c it still leaks two socket handles. With connection Windows 10: Running .\vibe_noleaks.exe [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. [() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP^ requests on C::1:8080 [main( ) INFC:\Software\vibe_noLeaks>] Stopped to listen for HTTP requests on 127.0.0.1:8080 Warning: 2 socket handles leaked at driver shutdown. Warning: 2 socket handles leaked at driver shutdown. This is normal. The server uses keepalive connections, so that in case any more requests arrive on the same connection, the initial connection setup does not need to be established. Well, at least that is what I think is happening. If you want a few seconds (I think 5 or so), then you won't get these. It would be good if vibe-d could provide a way to shut down any keepalive connections when the server is shutting down. -Steve That's interesting. I actually started to dive deeper into those and tried to pinpoint the lines of code that result in additional open sockets upon new http connections; although my understanding of vibe-d is a bit too low at this point to figure out what's exactly happening. I think from a user perspective having something a bit friendlier on warning side would be helpful. Do you have any suggestions in mind towards that? I have a bit of time this week and I could take a stab at it. Best, S
Re: Socket handle leak and active handle warning with Vibe-D
On Sunday, 3 January 2021 at 23:53:54 UTC, aberba wrote: On Friday, 1 January 2021 at 22:07:28 UTC, Selim Ozel wrote: I created the simplest possible example as explained by the Vibe-D community in [1]. The exact source code of what I run is in [2]. On Windows I get a socket handle leak warning on shutdown with crtl+c from terminal after running the executable. [...] On Ubuntu 20.04 I get leaking drivers warning with the same process. [...] I really don't know what this is all about but it is at the core of my Vibe-D development. So any pointers you might have would be very helpful to me. Thanks in advance. S [1] https://vibed.org/blog/posts/a-scalable-chat-room-service-in-d [2] https://github.com/SelimOzel/vibe_noLeaks Add this to your dub.json file to fix it "versions": [ "VibeHighEventPriority" ] This issue should be fixed in next vibe.d release Thanks. Not sure if relevant to this one but I came across that one as well before [1]. The symptom was "The simple hello world app I build with vibe-d does not seem to work on the second compile+execution." on an Ubuntu 20.04 EC2. [1] https://forum.dlang.org/post/mailman.6758.1605999004.31109.digitalmar...@puremagic.com
Re: Socket handle leak and active handle warning with Vibe-D
On 1/4/21 12:17 PM, Steven Schveighoffer wrote: If you want a few seconds *wait* a few seconds -Steve
Re: Socket handle leak and active handle warning with Vibe-D
On 1/3/21 6:53 PM, aberba wrote: On Friday, 1 January 2021 at 22:07:28 UTC, Selim Ozel wrote: I created the simplest possible example as explained by the Vibe-D community in [1]. The exact source code of what I run is in [2]. On Windows I get a socket handle leak warning on shutdown with crtl+c from terminal after running the executable. [...] On Ubuntu 20.04 I get leaking drivers warning with the same process. [...] I really don't know what this is all about but it is at the core of my Vibe-D development. So any pointers you might have would be very helpful to me. Thanks in advance. S [1] https://vibed.org/blog/posts/a-scalable-chat-room-service-in-d [2] https://github.com/SelimOzel/vibe_noLeaks Add this to your dub.json file to fix it "versions": [ "VibeHighEventPriority" ] This issue should be fixed in next vibe.d release No, this is a different issue. That issue is when the server doesn't shut down so you can't re-bind to the port without killing it with a SIGKILL signal. And thankfully it should be fixed in the latest release (not vibe.d, but eventcore I think). -Steve
Re: Socket handle leak and active handle warning with Vibe-D
On 1/2/21 12:52 PM, Selim Ozel wrote: After logging into to 127.0.0.1 for a single time in my browser, if I do a ctrl+c it still leaks two socket handles. With connection Windows 10: Running .\vibe_noleaks.exe [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. [() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP^ requests on C::1:8080 [main( ) INFC:\Software\vibe_noLeaks>] Stopped to listen for HTTP requests on 127.0.0.1:8080 Warning: 2 socket handles leaked at driver shutdown. Warning: 2 socket handles leaked at driver shutdown. This is normal. The server uses keepalive connections, so that in case any more requests arrive on the same connection, the initial connection setup does not need to be established. Well, at least that is what I think is happening. If you want a few seconds (I think 5 or so), then you won't get these. It would be good if vibe-d could provide a way to shut down any keepalive connections when the server is shutting down. -Steve
Re: Socket handle leak and active handle warning with Vibe-D
On Friday, 1 January 2021 at 22:07:28 UTC, Selim Ozel wrote: I created the simplest possible example as explained by the Vibe-D community in [1]. The exact source code of what I run is in [2]. On Windows I get a socket handle leak warning on shutdown with crtl+c from terminal after running the executable. [...] On Ubuntu 20.04 I get leaking drivers warning with the same process. [...] I really don't know what this is all about but it is at the core of my Vibe-D development. So any pointers you might have would be very helpful to me. Thanks in advance. S [1] https://vibed.org/blog/posts/a-scalable-chat-room-service-in-d [2] https://github.com/SelimOzel/vibe_noLeaks Add this to your dub.json file to fix it "versions": [ "VibeHighEventPriority" ] This issue should be fixed in next vibe.d release
Re: Socket handle leak and active handle warning with Vibe-D
On Friday, 1 January 2021 at 22:07:28 UTC, Selim Ozel wrote: I created the simplest possible example as explained by the Vibe-D community in [1]. The exact source code of what I run is in [2]. On Windows I get a socket handle leak warning on shutdown with crtl+c from terminal after running the executable. [...] On Ubuntu 20.04 I get leaking drivers warning with the same process. [...] I really don't know what this is all about but it is at the core of my Vibe-D development. So any pointers you might have would be very helpful to me. Thanks in advance. S [1] https://vibed.org/blog/posts/a-scalable-chat-room-service-in-d [2] https://github.com/SelimOzel/vibe_noLeaks For further reference, I also went through this issue [1]. It seems like I am seeing the same behavior as kinexis-uk. S [1] https://github.com/vibe-d/vibe.d/issues/2245
Re: Socket handle leak and active handle warning with Vibe-D
On Saturday, 2 January 2021 at 00:28:43 UTC, Steven Schveighoffer wrote: On 1/1/21 5:07 PM, Selim Ozel wrote: I created the simplest possible example as explained by the Vibe-D community in [1]. The exact source code of what I run is in [2]. On Windows I get a socket handle leak warning on shutdown with crtl+c from terminal after running the executable. [...] On Ubuntu 20.04 I get leaking drivers warning with the same process. [...] I really don't know what this is all about but it is at the core of my Vibe-D development. So any pointers you might have would be very helpful to me. Thanks in advance. 1. the sockets are leaked for a reason that is pretty obscure -- namely, the GC might need to access those sockets as the process is shut down. Prior to this, the end result of vibe.d server was frequently a segfault. 2. The reason they are leaking is most likely because you still have a listening socket somewhere. I wish it would tell you how that socket was allocated, but it doesn't. To fix, make sure all your listening sockets are closed. In my vibe.d app, I have the following: auto listener = listenHTTP(settings, router); scope(exit) listener.stopListening(); I also clean up my session store connection (something I had to add support for in vibe.d), which was a different source of leaking handles. I also clean up database connections, which might be cached. And finally, even with all this, I still get leaking driver messages if an HTTP keepalive socket is open. I really feel like vibe.d should give you the option of not printing this message, as most of the time, it's something you can ignore. -Steve Hey Steve. Thanks a ton for all the tips. FWIW I am writing down my findings below because just maybe they might be helpful for someone else later on. The scope guard seems to have fixed some of the leak complaints. Unfortunately there is still a leak after a single connection. Without connection Windows 10: Running .\vibe_noleaks.exe [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. [() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP requests on ::1:8080 [main() INF] Stopped to listen for HTTP requests on 127.0.0.1:8080 Without connection Ubuntu 20.04: Running ./vibe_noleaks [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. ^C[main() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP requests on ::1:8080 [main() INF] Stopped to listen for HTTP requests on 127.0.0.1:8080 After logging into to 127.0.0.1 for a single time in my browser, if I do a ctrl+c it still leaks two socket handles. With connection Windows 10: Running .\vibe_noleaks.exe [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. [() INF] Received signal 2. Shutting down. [main() INF] Stopped to listen for HTTP^ requests on C::1:8080 [main( ) INFC:\Software\vibe_noLeaks>] Stopped to listen for HTTP requests on 127.0.0.1:8080 Warning: 2 socket handles leaked at driver shutdown. Warning: 2 socket handles leaked at driver shutdown. I think vibe-d is also leaking more sockets when there is a web interface attached (not included in my toy repository). These haven't stopped me from developing but they are just things I wanted to write down and learn more before building even more infrastructure with it. I might dive into vibe-d codebase at some point too. Best, S
Re: Socket handle leak and active handle warning with Vibe-D
On 1/1/21 5:07 PM, Selim Ozel wrote: I created the simplest possible example as explained by the Vibe-D community in [1]. The exact source code of what I run is in [2]. On Windows I get a socket handle leak warning on shutdown with crtl+c from terminal after running the executable. [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. [() INF] Received signal 2. Shutting down. Warning: 2 socket handles leaked at driver^ Cshutdown On Ubuntu 20.04 I get leaking drivers warning with the same process. [main() INF] Listening for requests on http://[::1]:8080/ [main() INF] Listening for requests on http://127.0.0.1:8080/ [main() INF] Please open http://127.0.0.1:8080/ in your browser. ^C[main() INF] Received signal 2. Shutting down. Warning (thread: main): leaking eventcore driver because there are still active handles FD 6 (streamListen) FD 7 (streamListen) Warning (thread: main): leaking eventcore driver because there are still active handles FD 6 (streamListen) FD 7 (streamListen) I really don't know what this is all about but it is at the core of my Vibe-D development. So any pointers you might have would be very helpful to me. Thanks in advance. 1. the sockets are leaked for a reason that is pretty obscure -- namely, the GC might need to access those sockets as the process is shut down. Prior to this, the end result of vibe.d server was frequently a segfault. 2. The reason they are leaking is most likely because you still have a listening socket somewhere. I wish it would tell you how that socket was allocated, but it doesn't. To fix, make sure all your listening sockets are closed. In my vibe.d app, I have the following: auto listener = listenHTTP(settings, router); scope(exit) listener.stopListening(); I also clean up my session store connection (something I had to add support for in vibe.d), which was a different source of leaking handles. I also clean up database connections, which might be cached. And finally, even with all this, I still get leaking driver messages if an HTTP keepalive socket is open. I really feel like vibe.d should give you the option of not printing this message, as most of the time, it's something you can ignore. -Steve
