A "timeout" parameter would be nice

[viktorbenei]

Sometimes, especially with larger DigitalOcean droplets/snapshots the rebuild and other actions might time out before the DO droplet would be ready.

A simple, optional parameter would be nice, to be able to increase the timeout.

Related code: https://github.com/smdahlen/vagrant-digitalocean/blob/d2b2b729ea676f1b1c1d17651fba3081f495423c/lib/vagrant-digitalocean/helpers/client.rb#L91 - right now the retry count & sleep times are hardcoded, at least the retry count could be exposed as a configuration option.

Context: we started to hit frequent timeouts today, sent a ticket to DO and they replied that for larger droplets created from snapshots the create&re-build actions might actually take longer than the hardcoded limit in this plugin.

[seth-reeser]

Thank you for your feedback. 120 retries times 10 seconds sleep = 1200 seconds / 60 = 20 minutes.. How long are you proposing?

[viktorbenei]

I think that's good for a default value, actually less than 20 mins should be enough; but there might be a case when someone have to modify the timeout to fit their needs. IMO it would be good to have a configuration option for this.

That said, we're no longer using DigitalOcean, specifically because of the long boot times of custom/snapshot based VMs.