systemd `systemctl stop` aggressively kills subprocesses - redhat

I've a daemon-like process that starts two subprocesses (and one of the subprocesses starts ~10 others). When I systemctl stop my process the child subprocesses appear to be 'aggressively' killed by systemctl - which doesn't give my process a chance to clean up.
How do I get systemctl stop to quit the aggressive kill and thus to allow my process to orchestrate an orderly clean up?
I tried timeoutSec=30 to no avail.

KillMode= defaults to control-group. That means every process of your service is killed with SIGTERM.
You have two options:
Handle SIGTERM in each of your processes and shutdown within TimeoutStopSec (which defaults to 90 seconds)
If you really want to delegate the shutdown from your main process, set KillMode=mixed. SIGTERM will be sent to the main process only. Then again shutdown within TimeoutStopSec. If you do not shutdown within TimeoutStopSec, systemd will send SIGKILL to all your processes.
Note: I suggest to use KillMode=mixed in option 2 instead of KillMode=process, as the latter would send the final SIGKILL only to your main process, which means your sub-processes would not be killed if they've locked up.

A late (possible) answer, but as I googled for weeks with a similar issue, finding nothing, I figured I add my solution.
My error was that I ran the systemd unit as root and switched (using sudo) to "the correct" user in the startscript (inherited from SysVinit script).
That starts the processes in the user.slice which is killed mercilessly on shutdown. When I changed the unit file to run as the correct user (USER=myuser) and removed sudo from the start script, the processes start in the system.slice and get properly handled on shutdown.

Related

Airflow: what do `airflow webserver`, `airflow scheduler` and `airflow worker` exactly do?

I've been working with Airflow for a while now, which was set up by a colleague. Lately I run into several errors, which require me to more in dept know how to fix certain things within Airflow.
I do understand what the 3 processes are, I just don't understand the underlying things that happen when I run them. What exactly happens when I run one of the commands? Can I somewhere see afterwards that they are running? And if I run one of these commands, does this overwrite older webservers/schedulers/workers or add a new one?
Moreover, if I for example run airflow webserver, the screen shows some of the things that are happening. Can I simply get out of this by pressing CTRL + C? Because when I do this, it says things like Worker exiting and Shutting down: Master. Does this mean I'm shutting everything down? How else should I get out of the webserver screen then?
Each process does what they are built to do while they are running (webserver provides a UI, scheduler determines when things need to be run, and workers actually run the tasks).
I think your confusion is that you may be seeing them as commands that tell some sort of "Airflow service" to do something, but they are each standalone commands that start the processes to do stuff. ie. Starting from nothing, you run airflow scheduler: now you have a scheduler running. Run airflow webserver: now you have a webserver running. When you run airflow webserver, it is starting a python flask app. While that process is running, the webserver is running, if you kill command, is goes down.
All three have to be running for airflow as a whole to work (assuming you are using an executor that needs workers). You should only ever had one scheduler running, but if you were to run two processes of airflow webserver (ignoring port conflicts, you would then have two separate http servers running using the same metadata database. Workers are a little different in that you may want multiple worker processes running so you can execute more tasks concurrently. So if you create multiple airflow worker processes, you'll end up with multiple processes taking jobs from the queue, executing them, and updating the task instance with the status of the task.
When you run any of these commands you'll see the stdout and stderr output in console. If you are running them as a daemon or background process, you can check what processes are running on the server.
If you ctrl+c you are sending a signal to kill the process. Ideally for a production airflow cluster, you should have some supervisor monitoring the processes and ensuring that they are always running. Locally you can either run the commands in the foreground of separate shells, minimize them and just keep them running when you need them. Or run them in as a background daemon with the -D argument. ie airflow webserver -D.

mongod main process killed by KILL signal

One of the mongo nodes in the replica set went down today. I couldn't find what happened but when i checked the logs on the server, I saw this message 'mongod main process killed by KILL signal'. I tried googling for more information but failed. Basically i like to know what is KILL signal, who triggered it and possible causes/fixes.
Mongo version 3.2.10 on Ubuntu.
The KILL signal means that the app will be killed instantly and there is no chance left for the process to exit cleanly. It is issued by the system when something goes very wrong.
If this is the only log left, it was killed abruptly. Probably this means that your system ran out of memory (I've had this problem with other processes before). You could check if swap is configured on your machine (by using swapon -s), but perhaps you should consider adding more memory to your server, because swap would be just for it not to break, as it is very slow.
Another thing worth looking at is the free disk space left and the syslog (/var/log/syslog)

supervisord autorestart max tries?

http://supervisord.org/configuration.html#program-x-section-values says you can use autorestart=true to restart on exit, but doesn't say how to give a maximum amount of restarts (within startsecs) before giving up. Is there a way to do this? Note: I'm not talking about the first startup, but about the event that a program crashes after, say, running fine for 10 days.
According to the docs, autorestart doesn't care about startretries:
autorestart controls whether supervisord will autorestart a program if
it exits after it has successfully started up (the process is in the
RUNNING state).
supervisord has a different restart mechanism for when the process is
starting up (the process is in the STARTING state). Retries during
process startup are controlled by startsecs and startretries.
You should use startretries as well, ex of program configuration:
[program:consumer_example]
command=command example
process_name=%(program_name)s_%(process_num)02d
numprocs=1
autostart=true
autorestart=true
startretries=10
user=USERNAME
As you can see I used startretries with 10, when you not inform into program it uses the default value (3).
I think that you need is to use the startretries parameter..
http://supervisord.org/configuration.html?highlight=startretries#program-x-section-example
best regards

Will celery handle tasks gracefully if supervisorctl stop/start/restart is used?

In a case recently, I had to restart some inexplicably idle workers run by supervisord. We are thinking about adding a periodic restart, say, once or twice a day.
This could easily be done using supervisorctl, but is there any chance tasks will be lost while the restart occurs?

How to force supervisord to stop a process in BACKOFF status

When you start a process using supervisord it is in "STARTING" status then if it gets trouble it gets in "BACKOFF" status if there is an autorestart set to true.
I don't want to wait for "startretries" to be attempted, I want to stop the restarting process manually using supervisorctl. The only way I found to do so is to stop the entire supervisord service and start it again (every process go in "STOPPED" status if there is no autostart).
Is there a better way to do so (force "STOPPED" status from "BACKOFF" status) as I have other processes managed in supervisord that I don't want to stop?
If I try to stop it with
supervisorctl stop process
I get
FAILED: attempted to kill process with sig SIGTERM but it wasn't running
If I try to start it with
supervisorctl start process
I get
process: ERROR (already started)
Of course I could disable the autorestart, but it can be useful, a workaround is to limit the startretries, is there a better solution?
Hey this maybe help for you:
When an autorestarting process is in the BACKOFF state, it will be
automatically restarted by supervisord. It will switch between
STARTING and BACKOFF states until it becomes evident that it cannot be started because the number of startretries has exceeded
the maximum, at which point it will transition to the FATAL state.
Each start retry will take progressively more time.
so you don't need to stop the BACKOFF process manually. if you do not want to wait too long time, it is better to set a little number to startretries.
see more info here: http://supervisord.org/subprocess.html
GOOD LCUKY
Use the following command to force supervisor to stop a process in the BACKOFF state.
supervisorctl stop <gname>:*