Running a docker container built with your Dockerfile both server.py and sensor.py fail to restart.
Closed this issue · 4 comments
Running a docker container created with your Dockerfile fails to restart server.py and sensor.py via cron.
How To Reproduce
Build a docker container via your instructions.
**Expected behavior
When **2 1 * * * /usr/bin/pkill -f maltrail is run server.py and sensor.py should get restarted by
*/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep server.py)" ]; then : ; else python /opt/maltrail/server.py -c /opt/maltrail/maltrail.conf; fi >> /var/log/cron.log'
*/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep sensor.py)" ]; then : ; else python /opt/maltrail/sensor.py -c /opt/maltrail/maltrail.conf; fi >> /var/log/cron.log'
The problem is the following line is still detected by ps -ef:
root 1 0 0 Feb20 ? 00:00:00 /bin/sh -c bash -c "python /opt/maltrail/server.py &" && bash -c "python /opt/maltrail/sensor.py &" && cron && tail -f /var/log/cron.log
You need to add grep -v bash to the line so it ignore's the above line.
Your current Dockerfile has:
RUN (echo '/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep server.py)" ]; then : ; else python /opt/maltrail/server.py -c /opt/maltrail/maltrail.conf; fi >> /var/log/cron.log') | crontab
RUN (crontab -l ; echo '/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep sensor.py)" ]; then : ; else python /opt/maltrail/sensor.py -c /opt/maltrail/maltrail.conf; fi >> /var/log/cron.log') | crontab
I use:
RUN (echo '/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep -v bash | grep server.py)" ]; then : ; else bash -c "python /opt/maltrail/server.py -c /opt/maltrail/maltrail.conf &"; fi) | crontab
RUN (crontab -l ; echo '/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep -v bash | grep sensor.py)" ]; then : ; else bash -c "python /opt/maltrail/sensor.py -c /opt/maltrail/maltrail.conf &"; fi) | crontab
Environment:
- Device: Linux based
- OS: Devuan Linux
- Type of Maltrail installation: [e.g.
git clonecommand] - Problematic Maltrail component: server and sensor
- Maltrail version: 0.66
if i try to do this in separate terminals, it seems to me that all works as expected
p.s. sensor stopped because of no (root) permissions
This is what is running in my container shown when running ps -ef.
UID PID PPID C STIME TTY TIME CMD
root 1 0 0 Feb20 ? 00:00:00 /bin/sh -c bash -c "python /opt/maltrail/server.py &" && bash -c "python /opt/maltrail/sensor.py &" && cron && tail -f /var/log/cron.log
root 12 1 0 Feb20 ? 00:00:06 cron
root 13 1 0 Feb20 ? 00:00:28 tail -f /var/log/cron.log
root 18986 1 0 01:03 ? 00:02:00 python /opt/maltrail/sensor.py -c /opt/maltrail/maltrail.conf
root 18987 1 0 01:03 ? 00:00:09 python /opt/maltrail/server.py -c /opt/maltrail/maltrail.conf
root 26279 0 0 09:32 pts/0 00:00:00 bash
root 26292 26279 0 09:32 pts/0 00:00:00 ps -ef
The line with pid 1 is the line with the problem it has bash in it but more importantly it has server.py and sensor.py in it.
This line still exits even after the cron command:
2 1 * * * /usr/bin/pkill -f maltrail
is run.
The cron commands:
*/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep server.py)" ]; then : ; else python /opt/maltrail/server.py -c /opt/maltrail/maltrail.conf; fi >> /var/log/cron.log
*/1 * * * * if [ -n "$(ps -ef | grep -v grep | grep sensor.py)" ]; then : ; else python /opt/maltrail/sensor.py -c /opt/maltrail/maltrail.conf; fi >> /var/log/cron.log
match on that pid 1 line and think the server.py and sensor.py are still runing and doesn't restart them.
Adding grep -v bash prevents that from happening.