How to Troubleshoot Distribution’s Redis

Description

Redis is installed and ran as a service when using Distribution. Redis is used as a message broker when distributing release bundles from the source Artifactory instance to the targeted Artifactory Edge nodes. Since Redis is such a critical component to Distribution, it’s important to have an understanding of how to troubleshoot it if the need arises.

Resolution

Distribution should start up run on port 6379 when Distribution is started. Since Redis is a systemd service, it can be managed using systemctl.
Here are some useful commands for checking service status, starting the service, and stopping the service:
systemctl status redis
systemctl start redis
systemctl stop redis

The following configuration file can be used to verify/update the Redis connection string for the Distributor/Distribution services:
$DISTRIBUTION_HOME/etc/distribution_conf.gconfig

The following commands can be used to monitor Redis latency by running a ping command to the Redis instance and measuring the time for reply. In addition, adding the “history” flag will allow viewing of the latency over a period of time:
$ redis-cli —latency
$ redis-cli —latency-history

By default, Redis dataset snapshot set to automatically get backed up at the following location: /var/opt/jfrog/redis/dump.rdb. However, the following command can be used to manually generate a dataset snapshot from Redis to a desired location:
$ redis-cli --rdb /tmp/dump.rdb