Hue logs

Cloudera Manager generates standard stream logs when each Hue role instance starts and stops. The Hue service, which is built on the Django framework, generates log4j logs when the service is running.

Standard stream logs

Cloudera Manager logs the start and stop of each supervised Hue process in standard stream logs (stdout.log, stderr.log

When the Hue service restarts, Cloudera Manager generates a new directory for each supervised process of a Hue role instance:


ls -vrl /var/run/cloudera-scm-agent/process | grep HUE

It writes to a nested logs directory for each running instance:

Configuration errors are written here because they prevent Hue servers and load balancers from starting properly.

The supervisor

The supervisor is a watchdog process and supervisor.conf manages all Hue processes; its only purpose is to spawn and monitor other processes. A standard Hue installation starts and monitors the runcpserver process, which provides the core web functionality for Hue.

For each Hue role, Cloudera Manager looks to the appropriate supervisor.conf for instructions on how to start the server.

# Hue Server Process Directory
cd /var/run/cloudera-scm-agent/process/`ls -valrt /var/run/cloudera-scm-agent/process | grep HUE_SERVER | tail -1 | awk '{print $9}'`
cat supervisor.conf

[program:288-hue-HUE_SERVER]
command=cmf-redactor "/usr/lib64/cmf/service/hue/hue.sh" "runcpserver"
autostart=true
directory=/run/cloudera-scm-agent/process/288-hue-HUE_SERVER
stdout_logfile=/run/cloudera-scm-agent/process/288-hue-HUE_SERVER/logs/stdout.log
stdout_logfile_maxbytes=10MB
stdout_logfile_backups=10
stderr_logfile=/run/cloudera-scm-agent/process/288-hue-HUE_SERVER/logs/stderr.log
stderr_logfile_maxbytes=10MB
stderr_logfile_backups=10
environment= ...

# Hue Load Balancer Process Directory
cd /var/run/cloudera-scm-agent/process/`ls -valrt /var/run/cloudera-scm-agent/process | grep HUE_LOAD | tail -1 | awk '{print $9}'`
cat supervisor.conf

[program:258-hue-HUE_LOAD_BALANCER]
command=cmf-redactor "/usr/lib64/cmf/service/hue/httpd.sh"
...

If you installed other applications into your Hue instance, you may see other daemons running under the supervisor as well. Supervisor automatically restarts these processes if they fail for any reason. If they fail repeatedly in a short period of time, the supervisor itself shuts down.

Hue service Django logs

When the Hue service is running, Hue generates logs in /var/log/hue using log4j. Load balancer logs are in /var/run/httpd. You can view these logs in Hue at http://hueserver:port/logs.

Table 1. Hue service logs
Log Name	Description
`access.log`	Filtered list of successful attempts to access Hue Web UI
`audit/hue_server_audit_wal.log`	Audit log visible in Apache Atlas.
`error.log`	Filtered list of all nontrivial errors
`kt_renewer.log`	Kerberos ticket renews
`metrics-hue_server/metrics.log`	Populates charts in Cloudera Manager
`migrate.log`	Database and table migrations + First Run of Hue server
`runcpserver.log`	Hue (CherryPy) web server info (CP server runs Django core)
`hue_install.log`	Contains the log produced during installation

Enabling DEBUG

DEBUG is available for the Hue Django logs in /var/log/hue. By default, the Hue service writes INFO level messages and keeps a small buffer of log messages at all levels in memory.

There are two ways to enable DEBUG messages for all the logs in /var/log/hue :

Cloudera Manager: Go to Hue > Configuration, check Enable Django Debug Mode, and Save Changes > Restart.
Hue Web UI: Go to the Home page, select Server Logs, and check Force Debug Level. Debug is enabled on-the-fly.