Dockerized wdb server doesn't react to SIGTERM #136
Comments
yajo
added a commit
to Tecnativa/doodba-scaffolding
that referenced
this issue
Nov 18, 2019
Nowadays wdb has an official image which is up to date, so there's no need to use my own. Also I added a workaround for Kozea/wdb#136 to avoid having to wait so long for wdb container to stop. I hope I can revert that part soon.
yajo
added a commit
to Tecnativa/doodba-scaffolding
that referenced
this issue
Nov 18, 2019
Nowadays wdb has an official image which is up to date, so there's no need to use my own. Also I added a workaround for Kozea/wdb#136 to avoid having to wait so long for wdb container to stop. I hope I can revert that part soon.
|
I finally found a good workaorund: wdb:
image: kozea/wdb
# HACK https://github.com/Kozea/wdb/issues/136
init: trueThis avoids the problem of |
yajo
pushed a commit
to Tecnativa/doodba-copier-template
that referenced
this issue
Aug 6, 2020
Previous hack made it quick to stop the `wdb.server.py` process, which doesn't react properly to `SIGTERM` as explained in Kozea/wdb#136. However, if this process was still alive, when powering off the computer it took forever to kill it. Thus, using this alternative hack is better, because it uses docker's built-in `tini` to reap child processes. It is still fast, but it also helps the host's process reaper to work reliably. BTW I update MQT submodule.
yajo
pushed a commit
to Tecnativa/doodba-copier-template
that referenced
this issue
Aug 11, 2020
Previous hack made it quick to stop the `wdb.server.py` process, which doesn't react properly to `SIGTERM` as explained in Kozea/wdb#136. However, if this process was still alive, when powering off the computer it took forever to kill it. Thus, using this alternative hack is better, because it uses docker's built-in `tini` to reap child processes. It is still fast, but it also helps the host's process reaper to work reliably. BTW I update MQT submodule.
github-actions bot
pushed a commit
to Tecnativa/doodba-copier-template
that referenced
this issue
Aug 11, 2020
Previous hack made it quick to stop the `wdb.server.py` process, which doesn't react properly to `SIGTERM` as explained in Kozea/wdb#136. However, if this process was still alive, when powering off the computer it took forever to kill it. Thus, using this alternative hack is better, because it uses docker's built-in `tini` to reap child processes. It is still fast, but it also helps the host's process reaper to work reliably. BTW I update MQT submodule.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I use wdb-server in a
docker-compose.ymlfile, and when stopping services, it's always the latest. I investigated, and it seems it doesn't react toSIGTERM:The text was updated successfully, but these errors were encountered: