List of Log File Locations

Management

Logs from GUI, API, automated tasks (e.g. periodic backups) and from the version upgrade process can be found inside mgmt01.local VM. You can login there using ssh without a password (only) from the first compute node (first ssh to the first compute node and then ssh as root to the IP address of GUI).

Main log directory: /opt/erigones/log/

Most important logfiles: mgmt.log and main.log

If you encounter an error in GUI/API (e.g. error 500), go to these logfiles and search for an traceback or othes sort of error message.

There are also other log files with minor importance, such as access logs, auth logs, task log (logging all tasks issuet for compute nodes and version upgrade logs.

If you want to see classic logs from the VMs operating system itself, see /var/log/.

Node

Install logs

During the installation before the first reboot, all actions are logged into /var/log/prompt-config.log on a node. After the first reboot, the next script depend on whether it is the first compute node (esdc-ce-hn-*) or you are adding a new node to existing DC installation (esdc-ce-cn-*).

1. The first compute node install logfile: /var/log/headnode-install.log
2. All next compute nodes install logfile: /var/log/computenode-install.log

Runtime logs

All other Danube Cloud logs are located in /opt/erigones/log/.

Look into these if you encounter unavailable node or other misbehaving actions that involve a compute node. It is recommended to look into fast.log first because erigonesd:fast daemon is doing the most work. You can find there also update* logs from the node version upgrade process. Please note that main upgrade logs are in the mgmt01.local VM.

If you want to see classic logs from the VMs operating system itself, see /var/adm/, /var/log/ and svcs -L for system services.