nginx-common nginx.service comments use wrong signal
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
nginx (Debian) |
New
|
Unknown
|
|||
nginx (Ubuntu) |
Fix Released
|
Low
|
Thomas Ward |
Bug Description
lsb_release -rd:
Description: Ubuntu 20.04.1 LTS
Release: 20.04
apt-cache policy nginx-common:
nginx-common:
Installed: 1.18.0-0ubuntu1
Candidate: 1.18.0-0ubuntu1
Version table:
*** 1.18.0-0ubuntu1 500
500 http://
100 /var/lib/
1.
500 http://
In ubuntu 20.04, nginx use systemd to control status of nginx, the following is part of the /lib/systemd/
# ExecStop sends SIGSTOP (graceful stop) to the nginx process.
# If, after 5s (--retry QUIT/5) nginx is still running, systemd takes control
# and sends SIGTERM (fast shutdown) to the main process.
# After another 5s (TimeoutStopSec=5), and if nginx is alive, systemd sends
# SIGKILL to all the remaining processes in the process group (KillMode=mixed).
[Service]
Type=forking
PIDFile=
ExecStartPr
ExecStart=
ExecReload=
ExecStop=
TimeoutStop
As comment said, execstop will send SIGSTOP, and SIGTERM after 5 seconds if process don't stop, but actually the command below use SIGQUIT instead of SIGSTOP to gracefully shutdown the process.
then I looked for the nginx documentation http://
also, refrring to the linux signal(7) https:/
The signals SIGKILL and SIGSTOP cannot be caught, blocked, or ignored.
nginx shouldn't be able to capture SIGSTOP and take time to gracefully shutdown, SIGQUIT can
the comments above should change to
ExecStop sends SIGQUIT (graceful stop) to the nginx process.
If there's anything I missed it, pls let me know, thanks
Changed in nginx (Ubuntu): | |
status: | Confirmed → Triaged |
Changed in nginx (Debian): | |
status: | Unknown → New |
Hi Lisbeth,
Thank you for taking the time to report this bug and helping to make Ubuntu better.
your report is absolutely right and I checked the latest Ubuntu 21.04 where this is still the case.
OTOH while correct this is also of a low severity.
I'm assigning Thomas who usually does the nginx uploads to fix this on one of the next uploads he does.
But on the other hand this issues isn't a Ubuntu-specific one, so you might even consider filing the very same report in the Debian bug tracker would then fix it for both at once - and then Ubuntu will pick it up on the next merge.
Would you mind filing a bug with Debian please?