Hi Willy, list, I just encountered a segfault with 1.8.19 at startup and thought you might want to check it out. I put together a small config with which I can reproduce the crash every time:
----- global nbthread 2 listen test bind *:8080 mode tcp option tcp-check tcp-check connect port 1234 server test nonexistent.example.com:1234 check init-addr libc,none ----- I also managed to get a stack trace with the current haproxy-1.8 (a3cfe8f4): #0 tcpcheck_main (check=check@entry=0x555555746608) at src/checks.c:2774 #1 0x000055555561c464 in connect_conn_chk (t=0x55555575d1f0) at src/checks.c:1549 #2 process_chk_conn (t=0x55555575d1f0) at src/checks.c:2136 #3 process_chk (t=0x55555575d1f0) at src/checks.c:2282 #4 0x0000555555674182 in process_runnable_tasks () at src/task.c:320 #5 0x0000555555621fec in run_poll_loop () at src/haproxy.c:2419 #6 run_thread_poll_loop (data=<optimized out>) at src/haproxy.c:2485 #7 0x00007ffff7d39164 in start_thread (arg=<optimized out>) at pthread_create.c:486 #8 0x00007ffff7659def in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95 It does not happen with "nbthread 1". Please let me know if you need more debugging info. Best. Leo
