[Masterwow3]

An SSH/SFTP connection loss during upload currently fails the entire backup run, even when the archive has already been created and the containers have restarted. The keepalive added in #863 helps prevent idle disconnects but does not reconnect after a transport failure.

This change retries the existing archive over a fresh SSH/SFTP connection. Each attempt starts at byte zero and truncates the destination file. It does not recreate the archive or run container lifecycle hooks again.

• Add SSH_UPLOAD_RETRIES (default: 2; 0 disables retries) and SSH_UPLOAD_RETRY_DELAY (default: 5s).
• Retry connection errors only. Local source errors, permission errors, and other permanent SFTP errors still fail immediately.
• Close the failed session and its keepalive before reconnecting. Cleanup and pruning use the current session; a previous session's close error cannot turn a recovered upload into a failed run.
• Log upload success only after the remote file has closed successfully and the transferred byte count matches the source size.

Validation:

• go test ./...
• go vet ./...
• GOOS=linux GOARCH=amd64 go build ./...
• In-process SSH/SFTP regression test that forcibly closes the TCP connection during upload, verifies a new connection, compares the complete remote archive byte for byte, and checks pruning and cleanup on the replacement session.
• Unit coverage for retry limits, failed reconnects, disabled retries, retry delays, permanent errors, remote close failures, size mismatches, and configuration defaults/validation.

The Docker Compose integration suite was not run because no Docker daemon was available in the development environment. The SSH/SFTP regression test runs as part of the Go test suite without Docker.

Scope: this change does not retry initial backend setup or pruning, retain temporary archives after retries are exhausted, or publish remote files by atomic rename. An exhausted upload can still leave a partial remote file, as before. Existing SSH connection/handshake timeout behavior is unchanged.

[m90]

Thanks for this. I think this is a nice feature to have, but considering stability, I'd prefer starting with a default value of 0 and letting users opt into this behavior. See the comment inline. Also, there is a linting error in the test.

Let me know if there are any questions.

[Masterwow3]

Thanks for the review! Addressed the feedback in a94c676: retries now default to 0, the redundant retry-count check is removed, and the lint error is fixed.

[m90]

Thank you once more, I appreciate the thorough tests on top of the feature. This looks good to me, but I'd like to defer releasing it for a week or so so that SSH users can catch up with the required migration from https://github.com/offen/docker-volume-backup/releases/tag/v2.49.2. Otherwise, I'd be afraid the release notes get lost and people are thinking this feature broke their setup. I'll let you know when this is released.