Why is my gunicorn process ignoring the log-level setting with Django?

Viewed 22502

I have Nginx, Gunicorn, and Django all running on the same Ubuntu EC2 instance. I have a fairly conventional setup and would like to log all gunicorn errors to a particular file.

My configuration for Gunicorn is:

#!/bin/bash

NAME="server"
GUNICORNDIR=/ebs/env/bin
DJANGODIR=/ebs/server/
SOCKFILE=/tmp/gunicorn.sock
LOGFILE=/var/log/gunicorn/gunicorn.error
USER=ubuntu
GROUP=ubuntu
NUM_WORKERS=5
TIMEOUT=60
DJANGO_SETTINGS_MODULE=settings
DJANGO_WSGI_MODULE=wsgi

echo "Starting $NAME"

RUNDIR=$(dirname $SOCKFILE)
test -d $RUNDIR || mkdir -p $RUNDIR

exec $GUNICORNDIR/gunicorn ${DJANGO_WSGI_MODULE}:application \
  --name $NAME \
  --workers $NUM_WORKERS \
  --timeout=$TIMEOUT \
  --user=$USER --group=$GROUP \
  --log-level=error --log-file=$LOGFILE \
  --bind=unix:$SOCKFILE

However with this configuration I am getting all logs from DEBUG and above written to the file. My log-level parameter appears to be getting ignored.

What I am looking for is to only have these types of log messages written:

2014-01-02 13:54:53 [3327] [CRITICAL] WORKER TIMEOUT (pid:3338)
2014-01-02 13:54:53 [3327] [CRITICAL] WORKER TIMEOUT (pid:3338)

I thought that the Django logging config specified in my settings.py might be interfering so I added a handler and a logger to try and target gunicorn but that did not work.

'handlers': {
'gunicorn': {
    'level': 'ERROR',
    'class': 'logging.handlers.RotatingFileHandler',
    'formatter': 'verbose',
    'filename': '/ebs/log/gunicorn.error',
    'maxBytes': 1024 * 1024 * 100,
},
}
'loggers': {
'gunicorn.errors': {
'level': 'ERROR',
'handlers': ['gunicorn'],
'propagate': False,
},

Here are the versions that I am running

Django 1.5.4 Nginx nginx/1.1.19 Gunicorn 18.0

Any thoughts on what is wrong here?

** Update **

Here is what my django logging config looks like:

LOGGING = {
'version': 1,
'disable_existing_loggers': True,
'root': {
    'level': 'WARNING',
    'handlers': ['sentry'],
},
'formatters': {
    'verbose': {
        'format': '%(levelname)s %(asctime)s %(module)s %(process)d %(thread)d %(message)s'
    },
    'simple': {
        'format': '%(levelname)s %(asctime)s -- %(message)s'
    }
},
'handlers': {
    'sentry': {
        'level': 'ERROR',
        'class': 'raven.contrib.django.raven_compat.handlers.SentryHandler',
    },
    'sentry_file': {
        'level': 'ERROR',
        'class': 'logging.handlers.RotatingFileHandler',
        'formatter': 'verbose',
        'filename': '/ebs/log/sentry_log.txt',
        'maxBytes': 1024 * 1024 * 100,  # 100 mb
    },
    'celery': {
        'level': 'DEBUG',
        'class': 'logging.handlers.RotatingFileHandler',
        'filename': '/ebs/log/celery/celery.log',
        'formatter': 'verbose',
        'maxBytes': 1024 * 1024 * 100,
    },
    'apps': {
        'level': 'DEBUG',
        'class': 'logging.handlers.RotatingFileHandler',
        'formatter': 'verbose',
        'filename': '/ebs/log/apps.log',
        'maxBytes': 1024 * 1024 * 100,
    },
    'apps.dss': {
        'level': 'DEBUG',
        'class': 'logging.handlers.RotatingFileHandler',
        'formatter': 'verbose',
        'filename': '/ebs/log/dss_apps.log',
        'maxBytes': 1024 * 1024 * 100,
    },

},
'loggers': {
    'django.db.backends': {
        'level': 'DEBUG',
        'handlers': ['sentry'],
        'propagate': False,
    },
    'sentry': {
        'level': 'DEBUG',
        'handlers': ['sentry'],
        'propagate': False,
    },
    'sentry.errors': {
        'level': 'ERROR',
        'handlers': ['sentry_file', 'sentry'],
        'propagate': False,
    },
    'celery': {
        'level': 'INFO',
        'handlers': ['sentry', 'celery'],
        'propagate': False
    },
    'apps': {
        'level': 'DEBUG',
        'handlers': ['apps', 'sentry'],
        'propagate': False
    },
    'apps.dss' : {
        'level': 'DEBUG',
        'handlers': ['apps.dss', 'sentry'],
        'propagate': False,
    },
},
}
1 Answers
Related