Skip to content

Retry on aborted connections #24

@jcass77

Description

@jcass77

The radio stream stops playing when the connection is reset, and the log contains:

2015-12-08 05:29:42,998 - INFO 304 GET /musicbox_webclient/index.html (192.168.0.213) 13.42ms
2015-12-08 05:29:43,164 - INFO 304 GET /mopidy/mopidy.min.js (192.168.0.213) 11.70ms
2015-12-08 05:29:53,476 - INFO Starting new HTTP connection (1): opml.radiotime.com
2015-12-08 05:30:29,771 - INFO TuneIn API request for Tune.ashx failed: ('Connection aborted.', error(104, 'Connection reset by peer'))
2015-12-08 05:30:29,775 - ERROR Failed to tune station id s225233
2015-12-08 05:30:29,779 - WARNING Track is not playable: tunein:station:s225233
2015-12-08 05:30:39,838 - INFO Starting new HTTP connection (2): opml.radiotime.com
2015-12-08 05:30:42,126 - INFO 200 GET / (127.0.0.1) 42.55ms
2015-12-08 05:30:43,962 - WARNING Element doesn't implement handling of this stream. Please file a bug.

It might be possible to have Mopidy-TuneIn retry a couple of times before giving up, which should make it much more robust.

pydora follows a two-pronged strategy of retrying connections at least once, and also handling specific HTTP-related exceptions with the time between retries increasing exponentially.

It seems to be working well for the Mopidy-Pandora backend and I haven't had any connection issues since - not sure if it can be ported.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions