100% CPU utilization

[captnbp]

Hi,

When launching supervisord (last Github version), it seems to work fine, but the supervisord process uses 100% of my CPU without getting lower.

I have a RedHat 6.5, with Python 3.4.

[mnaberez]

Does it behave differently on the latest stable release version (3.1.3)? Can you give us some way to reproduce it (e.g. sample config file, specific steps to follow)?

[captnbp]

Hi,

I can't test 3.1.3 as I have Python 3.4, not Python 2.7.

For the config files I have this first one :

file=/data/hyperic/tmp/supervisor.sock   ; (the path to the socket file)
chmod=0700                 ; socket file mode (default 0700)

[supervisord]
logfile=/data/hyperic/var/log/supervisord.log ; (main log file;default $CWD/supervisord.log)
logfile_maxbytes=50MB        ; (max main logfile bytes b4 rotation;default 50MB)
logfile_backups=10           ; (num of main logfile rotation backups;default 10)
loglevel=info                ; (log level;default info; others: debug,warn,trace)
pidfile=/data/hyperic/var/run/supervisord.pid ; (supervisord pidfile;default supervisord.pid)
nodaemon=false               ; (start in foreground if true;default false)
minfds=1024                  ; (min. avail startup file descriptors;default 1024)
minprocs=200                 ; (min. avail process descriptors;default 200)
childlogdir=/data/hyperic/var/log            ; ('AUTO' child log dir, default $TEMP)

[rpcinterface:supervisor]
supervisor.rpcinterface_factory = supervisor.rpcinterface:make_main_rpcinterface

[supervisorctl]
serverurl=unix:///data/hyperic/tmp/supervisor.sock ; use a unix:// URL  for a unix socket

[include]
files = /data/hyperic/etc/supervisord/*.ini

If I don't provide any .ini file, so nothing to start the CPU is fine. If I provide one or several like this one, the CPU is fully loaded by the supervisord process only. All my services (Nginx, PHP-FPM, uWSGI are working fine) :

command=nginx
numprocs=1                    ; number of processes copies to start (def 1)
autostart=true                ; start at supervisord start (default: true)
autorestart=unexpected        ; whether/when to restart (default: unexpected)
startsecs=5                   ; number of secs prog must stay running (def. 1)
startretries=3                ; max # of serial start failures (default 3)
exitcodes=0,2                 ; 'expected' exit codes for process (default 0,2)
stdout_logfile=%(ENV_SVCMOUNT)s/var/log/nginx/nginx.log
stdout_logfile_maxbytes=10MB   ; max # logfile bytes b4 rotation (default 50MB)
stdout_logfile_backups=10     ; # of stdout logfile backups (default 10)
stderr_logfile=%(ENV_SVCMOUNT)s/var/log/nginx/nginx_error.log
stderr_logfile_maxbytes=10MB   ; max # logfile bytes b4 rotation (default 50MB)
stderr_logfile_backups=10     ; # of stderr logfile backups (default 10)

Here are my logs :

2015-02-17 14:32:05,147 WARN Included extra file "/data/hyperic/etc/supervisord/php-fpm.ini" during parsing
2015-02-17 14:32:05,161 INFO RPC interface 'supervisor' initialized
2015-02-17 14:32:05,161 CRIT Server 'unix_http_server' running without any HTTP authentication checking
2015-02-17 14:32:05,162 INFO daemonizing the supervisord process
2015-02-17 14:32:05,163 INFO supervisord started with pid 10435
2015-02-17 14:32:06,167 INFO spawned: 'php-fpm' with pid 10436
2015-02-17 14:32:16,165 INFO success: php-fpm entered RUNNING state, process has stayed up for > than 10 seconds (startsecs)
2015-02-17 14:32:24,889 INFO waiting for php-fpm to die
2015-02-17 14:32:24,892 INFO stopped: php-fpm (exit status 0)

[mnaberez]

I can't test 3.1.3 as I have Python 3.4, not Python 2.7.

The latest stable version (3.1.3) works on Python 2.5 - 2.7 if one of those is an option. Python 3 support has not been released yet. We appreciate you testing it and will try to look into this issue, but please keep the README.rst disclaimer in mind, it's probably not a good idea to run it in production yet.

[captnbp]

I'm aware of the disclamer, no problem for me, production is for end of 2015 on my side, so I have time to test and help you debugging. I only use it for Test purpose. I'll install a sandbox with Python 2.7 and reproduce the same context to see how Supervisor behaves.

[auvipy]

i also have plan to use supervisor with python 3.4x on ubuntu test production server

[mnaberez]

This should be fixed by #589.

[msabramo]

@captnbp: Did this solve your problem?

[captnbp]

Yes perfect ! No more CPU overhead. Thanks a lot !

Benoît Pourre
Tel: (+33)6 81 36 78 92

2015-03-06 5:04 GMT+01:00 Marc Abramowitz [email protected]:

@captnbp https://github.com/captnbp: Did this solve your problem?

—
Reply to this email directly or view it on GitHub
#581 (comment)
.

[RyanBalfanz]

I'm seeing this now on a Mac, installed via Homebrew, but only after enabling the web UI like so:

[inet_http_server]         ; inet (TCP) server disabled by default
port=127.0.0.1:9001        ; (ip_address:port specifier, *:port for all iface)

➜  ~ brew info supervisor
supervisor: stable 3.3.1 (bottled)
Process Control System
http://supervisord.org/
/usr/local/Cellar/supervisor/3.3.1 (587 files, 7.4M) *
  Poured from bottle on 2016-08-09 at 23:31:13
From: https://github.com/Homebrew/homebrew-core/blob/master/Formula/supervisor.rb
==> Caveats
To have launchd start supervisor now and restart at login:
  brew services start supervisor
Or, if you don't want/need a background service you can just run:
  supervisord -c /usr/local/etc/supervisord.ini

2016-08-09 23:50:50,876 INFO supervisord started with pid 4382
2016-08-09 23:51:27,008 WARN received SIGTERM indicating exit request
2016-08-09 23:51:27,234 INFO Increased RLIMIT_NOFILE limit to 1024
2016-08-09 23:51:27,245 INFO RPC interface 'supervisor' initialized
2016-08-09 23:51:27,245 CRIT Server 'inet_http_server' running without any HTTP authentication checking
2016-08-09 23:51:27,245 INFO RPC interface 'supervisor' initialized
2016-08-09 23:51:27,245 CRIT Server 'unix_http_server' running without any HTTP authentication checking
2016-08-09 23:51:27,246 INFO supervisord started with pid 4487

[nicolastobo]

I have the same issue than @RyanBalfanz , installed via brew on MacOs

$ brew info supervisor                                                                                                                                                                                                          
supervisor: stable 3.3.1 (bottled)
Process Control System
http://supervisord.org/
/usr/local/Cellar/supervisor/3.3.1 (592 files, 7.6M) *
  Poured from bottle on 2016-11-03 at 11:42:18
From: https://github.com/Homebrew/homebrew-core/blob/master/Formula/supervisor.rb
==> Caveats
To have launchd start supervisor now and restart at login:
  brew services start supervisor
Or, if you don't want/need a background service you can just run:
  supervisord -c /usr/local/etc/supervisord.ini

[hehongbo]

experienced same thing like @RyanBalfanz @nicolastobo .
macOS Sierra 10.11.2, installed supervisor via easy_install
Activity Monitor report high CPU Usage of Python process when i open Web UI（:9001）
restart supervisord will make it come back normal.