-
Notifications
You must be signed in to change notification settings - Fork 1.7k
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
rtkrcv 3.4.3 b26 Segmentation fault #272
Comments
This error was added with version 2.4.3 b24, if you run rtkrcv with option "-s". If you start the rtk server manually, it works. |
I confirm this issue. Manual starting works. Automatic start causes SEGFAULT. |
This is ARM Debian 8 (raspian jessie) |
Hallo, Me too had the same problem working with the version of rtklib 2.4.3 b26 on a raspberry. The option -s in this version gives rise to a segmentation fault (on the contrary, this option for rtklib 2.4.2 worked fine...). Any hint how to fix it? Thank you |
Hello,
use version 2.4.3 b23, checkout with:
git checkout -b rtklib_2.4.3_b23 5622a5d
David
Il giorno 26/apr/2017, alle ore 15:54, Tiziano Cosso <[email protected]> ha scritto:
Hallo,
Me too had the same problem working with the version of rtklib 2.4.3 b26 on a raspberry. The option -s in this version gives rise to a segmentation fault (on the contrary, this option for rtklib 2.4.2 worked fine...).
Any hint how to fix it?
Thank you
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or mute the thread.
Ing. David Zuliani
OGS - Istituto Nazionale di Oceanografia e di Geofisica Sperimentale
dep: Centro di Ricerche Sismologiche - CRS
Via Treviso, 55 - 33100 Cussignacco UDINE Italy
ph. +39 0432 1450132
E-mail: [email protected]
|
Going back to prior version is a temp fix. Looks like b24 changes broke it. |
Any ETA for fix? |
I checked and saw some poorly implemented and threading causing this issue. Here is a quick+dirty but working fix if you want to use autostart: You should add lines with + and recompile rtkrcv. When you do this patch, RTKRCV will ALWAYS AUTOSTART. I think this is the right way anyways, Again, don't use -s option. So HERE IS THE FIX: In rtkrcv.c ONLY:
|
Just tested my patch and it works. You should not use -s. It will just auto start the server without it. Using -s causes it to segfault. |
Sry to warm up an old issue even though it's not solved yet. I ran into the same issue, but the "patch" from @sezeryalcin didn't work for me, as that one requires a login into the rtkrcv app, or into one of the virtual terminals via telnet or something like that. My usecase requires a standalone start. So I don't have a real fix, but I went a little bit deeper into the code and the problem is essentially that the app can't print its messages somewhere as there is neither a virtual terminal or a real one. So under the assumption that the app is started headless and no one connects via telnet, there is no use in printing the messages at that point. Fixing it the right way would need a more thorough rewrite of that part of the message handling process.
|
@kme-it Hi, You patch are great, but it doesn't help to start rtkrcv completely headless: I've able to start it in terminal.
|
Hi @Lankaster, I am not sure if I understand you correctly. I wrote this patch so one could start this via systemd for example. This works with this patch. You could start this in terminal manually too, but then you don't need this fix. The one from sezeryalcin should work just fine. |
Hi @kme-it I try to use it on OpenWrt device headless (without terminal) via init.d: https://github.com/openwrt/packages/tree/master/utils/rtklib |
@Lankaster Is it working from a terminal in an interactive mode? |
@kme-it With this patch we're able to start rtkrcv from terminal manually, but "-s" option doesn't work: we can see only "Segment fault" error. Hier is strace log:
|
Hi @Lankaster, first I would try without the patch in interactive mode if the segfaults are happening there too. And after that in interactive mode with the patch, to bisect the scenario. I am right now unsure if you might need the other patch to with the -s option or if you can skip the -s option. Maybe you should try with both patches. I don't have access to the the working instance with this setup here right now, so I can't look it up and at least right now I don't remember if my setup startet with or without the option. The basic idea behind this patch is to disable all logging to a virtual terminal (which doesn't exist). This should be straight forward and i don't see how it might leed to a segfault. I assume the patch applied without errors. If not it's difficult to debug like this. A stacktrace would be nice, or if available a debugbuild with remote debugging might help too. |
Hi, Regards, |
…-cmd-string-init rtknavi_qt serverStart: initialize cmd strings emptry
Dear all
i'm trying rtkrcv both on my mac and on my linux machine. I've compiled and installed the source code without errors but when I run the command:
rtkrcv -s -p 2950 -o myconf.conf
I get: Segmentation fault
I've tried the same myconf.conf file with a previous 3.4.3 b16 release and it works.
Can you help me?
The text was updated successfully, but these errors were encountered: