WARNING WARNING WARNING: This is the Digital Rebar v2 product documentation. The v2 version is EOL as of September 2017. Please refer to the new Digital Rebar Provision v3 documentation: http://provision.readthedocs.io/en/tip/README.html
There are several possible failures when running Workload scripts.
5.1. No Response From Server¶
Digital Rebar requires HTTPS and accepting the self-signed certificate.
5.2. SSH Key¶
If having trouble getting the SSH keys to work correctly, include the
--clean-ids flag in the script.
html Rebar creates a unique SSH key for each node created; consequently, Rebar SSH IDs may sometimes be orphaned with the provider. These can be safely removed with the Provider UX or CLI.
5.3. Provider Failure¶
Provider node allocation sometimes fails. In these cases, the annealer will be stuck waiting for a node that never completes. For worker nodes, simply``destroy`` the node and processing will continue. If the node was a master node, the whole deployment might have to be destroyed and rebuilt.
See Troubleshooting Tips for details about testing provider configuration.
5.4. Script Failure¶
If the script fails to complete after it has created the Digital Rebar Admin node and before it completed Admin configuration, retry the script by providing the
--device-id instead of
--deploy-admin. This will tell the script retry the Rebar Admin setup.