Under some conditions, the process can go down without any apparent reason. There is typically no core file to analyze why the process goes down. In my case, I tried to show how to find sighup signal for a process. Also, you can check your process’ activity with truss command. Please check “man truss”.
Step 1: Create a dummy process to apply test action
#sleep 1000 & [2] 16177 # ps -ef|grep 16177 testuser 16177 22161 0 09:57:38 pts/23 0:00 sleep 1000
NAME | PID |
Sleep Dumy Process | 16177 |
SSHD Process | 22161 |
Step 2:How to start truss for a specific Process
# truss -o /tmp/truss.out -sall -p 16177
Step 3: After starting truss, we can send “sighup” with close terminal.
“Break your SSH connection terminal, which “Sleep” command running.I meant disconnect session or close SSH session terminal”.
Step 4:Check Truss output file.
# cat /tmp/syslog.truss.out nanosleep(0xFB53A628, 0xFB53A620) (sleeping…) Received signal #1, SIGHUP, in nanosleep() [default] siginfo: SIGHUP pid=22161 uid=1548235000 >> we observed that SIGHUP signal send by SSHD process. Because we closed SSH connection which PPID of “sleep” process. nanosleep(0xFB53A628, 0xFB53A620) Err#4 EINTR