-
Notifications
You must be signed in to change notification settings - Fork 492
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’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
TCP hangup #1435
Comments
BTW: TCP client receives messages happily. |
More information on your attempted configuration, please? |
Protocol TCP, Address: 0.0.0.0 DataPort: 10110 but happens on every DataPort I used until now, no UserComment, Priority does not matter. |
Still not sure what is is that you are trying to do. Create a TCP server? If so, then the address should be something other than 0.0.0.0. |
Oh sorry, then I do have a wrong understanding of the documentation, which states: |
kapitan... |
Depends on what you understand of 'which side is client and which is server'. |
Probably something going on here, although the server (0.0.0.0) configuration is commonly used on other linux platforms without problem. Would be worth running under gdb, and getting a stack trace on the fault, if possible. |
Ive found using O as a server in that config oV50 it becomes a very fragile thing on macOS as well, but I am still analysing the root cause(s) I have found rebooting after setting up oV50 as a server seems to stop the initial crash caused by a client initiating the TCP connection but about 4 hours and things get really slow and a reboot Is required of the app For it to act as a server is it using GPSD? I know I have GPSD installed on my mac Instances but don’t have the service running as per the brew install advice, so I assumed this was something to do with it ? If it isn’t configured as a server the tcp connections are a lot more robust. Dashboard can be a bit sluggish in seeing the connection changes and to start parseing on not parseing sentences. Will grab a crash dump and upload it next time I am playing with tcp config’s /Ron |
We do not use the GPSD library. I found it to be fragile, also. We parse GPSD messages as received. |
Here you go with the gdb output ... I connected two times and moved the cursor a bit and with the second time disconnecting. GDB output: ` ` |
Valgrind output ... see line 553: |
And from macOS .... /Ron Process: OpenCPN [22801] Date/Time: 2019-07-05 15:52:28.782 +1000 Time Awake Since Boot: 330000 seconds System Integrity Protection: enabled Notes: Translocated Process Crashed Thread: 0 Dispatch queue: com.apple.main-thread Exception Type: EXC_BAD_ACCESS (SIGSEGV) Termination Signal: Segmentation fault: 11 VM Regions Near 0x71: Thread 0 Crashed:: Dispatch queue: com.apple.main-thread Thread 1: Thread 2: Thread 3: Thread 4: Thread 5: Thread 6:: com.apple.CFSocket.private Thread 7:: com.apple.NSEventThread Thread 8: Thread 0 crashed with X86 Thread State (64-bit): Logical CPU: 4 Binary Images: External Modification Summary: VM Region Summary:
REGION TYPE SIZE COUNT (non-coalesced) Model: MacPro6,1, BootROM 130.0.0.0.0, 8 processors, 8-Core Intel Xeon E5, 3 GHz, 64 GB, SMC 2.20f18 |
scadaperspective Why you have OpenCPN here: /private/var/folders/*/OpenCPN.app |
Because it is a Mac Pro |
Even a Mac Pro runs macOS. So you are doing something wrong. |
Apologies to everyone else CarCode, I have resisted replying to you on other occasions and mediums however I decided to try to give you one last opportunity to pull your head in as we say in Australia. Please do pull your head in go sailing and stop bothering nice people. This reply is the last time I will respond to you so your 15 minutes with me is now over. As far as I am concerned Car Code your comment sadly is as frequently and usually the case nothing to do with what is being posted. It add's no benefit to me or the conversation at hand and your comments directed towards me are completely unsolicited. Please don't communicate with me unless I solicit such a response, unless you use / provide your real first name and not some pseudonym or at the very least you are actually being helpful. I have not had you introduced to me as being part of the actual support team or development team, so why are you commenting at all here unless it is to report or convey an end user issue? Do you work for Apple computer's Inc USA ? No, I suspect not as you would know better otherwise. The Machine the dump was captured from IS an Apple Mac Pro business workstation build by Apple computers Inc, it is a standard end user build and not a special features build, It was purchased from them by me as a regular apple business customer, all software applications including operating system was/is installed by them and it was shipped to Apple Australia (Sydney) for final testing before being shipped to me and remained under a controlled chain of custody now residing with me. It comes with the configuration I purchased from them. I have made no alterations to the machine apart from installing software and hardware approved for use by Apple computers Inc. I am using this equipment correctly, in fact actually my usage is expressly approved by Apple computers Inc as being fair and reasonable use of the product and in full accordance with their terms and conditions of sale in Australia. So you're wrong. The problems with wxWidgets and for Apple specifically with Xcode frameworks I have been experiencing and are now resolved are external to and un related to the issues associated with wxWidgets 3.1.2 or 3.1.3 for that matter, are more general OpenGL compatibility issues related to the support library, issues that I and others are genuinely experiencing and trying to assist in resolving, people including Mike are reporting is the actual topic being discussed, when you, as usual, rudely interjected. Please don't do it any more, and more importantly for you please don't do it towards me specifically as I will take affirmative actions to have this pseudonym you are using be blocked here and on the cruisers forum if you continue. Ron Southworth |
@kapitan-k @bdbcat Has this been resolved? Or should it stay open? Original poster needs to close it. |
No resolution. Not important for me, no further issues / comments of others. |
Using Linux Mint 19.1 and OpenCPN version 5.0.0 OpenCPN hangs up when I connect via TCP client (in golang or telnet, does not matter) ... port does not matter neither.
CPU usage for OpenCPN moves then to 100% and does not respond until I disconnect the client. After disconnection and further click on the chart OpenCPN crashes.
Log shows nothing special.
Any ideas?
The text was updated successfully, but these errors were encountered: