When using atomic: true with trackMode: kubedog, the --atomic flag was
passed to Helm which caused Helm to wait for resources and rollback
internally before helmfile could start kubedog tracking. This prevented
users from seeing deployment logs and errors.
Changes:
- Add RollbackRelease method to helmexec interface and implementation
- Add appendAtomicFlags function that skips --atomic when kubedog is enabled
- Add handleKubedogFailure function to manually rollback on tracking failure
- Update flagsForUpgrade to use new atomic flag handling
- Add tests for appendAtomicFlags behavior
Now when atomic: true and trackMode: kubedog are both enabled:
1. Helm upgrade runs without --atomic (no internal wait)
2. kubedog tracks resources immediately (users can see logs)
3. If kubedog tracking fails, helmfile manually executes rollback
4. Atomic semantics are preserved
Fixes: #2448
Signed-off-by: yxxhero <aiopsclub@163.com>
8e38fb02dd