Reinhard, Am 12.09.20 um 16:43 schrieb Reinhard Vicinus: >>> thanks, for your reply and the information. Sorry for my late reply, but >>> I had only today time to test. I did try to get the spoa server working >>> on a ubuntu bionic (18.04.4) with haproxy 2.2.3-2ppa1~bionic from the >>> vbernat ppa. I could compile the spoa server with python 3.6 support >>> from the latest github sources without obvious problems and it also >>> started without problems with the example python script (./spoa -d -f >>> ps_python.py). >>> >>> If I start haproxy with the following command: >>> >>> haproxy -f spoa-server.conf -d >>> >>> haproxy seg faults on the first request to port 10001 >> This is a bug in HAProxy then. Do you happen to have a core dump / stack >> trace? > Yes I have a core dump. But I am somewhat rusty in analyzing them. So > any pointers what to do with the core dump is appreciated. Also the > segmentation fault only occurs if the spoa server is running so the > problem is probably somewhere in the code regarding the connection to > the spoa server.
Install the haproxy-dbgsym package to install the debug symbols to make the stack trace readable. Then: gdb haproxy <corefile> In gdb use 'bt' to get the backtrace of the thread that killed the process. Possibly 't a a bt' to see what the other threads were doing. Ideally file an issue within the tracker: https://github.com/haproxy/haproxy/issues. It's easier to track it within there and there's a template to fill in. >>> If I start haproxy with the additional parameter -Ws then it does not >>> seg fault, but only the first and every 4th request get (correctly?) >>> forwarded to the spoa server, the 3 requests in between get answered >>> with an empty %[var(sess.iprep.ip_score)]. >>> >>> [...] >>> >>> I am unsure if I am making some stupid mistakes, or if I should test it >>> with an older haproxy version or how to debug the issue further. So any >>> pointers are very much appreciated. >> Can you share the configuration you attempted to use? > Sorry, I forgot to mention that the configuration used is the example > configuration from haproxy repository. But here it is, to ensure that > there were no changes in the meantime: Unfortunately I can't comment on the SPOA functionality, because I never used it. Hopefully the information is sufficient for someone more proficient to tell what's going wrong there. Best regards Tim Düsterhus
