On 17 Feb 2011, at 04:40, Adam Wiggins wrote: > In the meantime, you shouldn't have any difficulties as long as you > keep your web requests short (less than about 500ms), which is good > practice anyway.
Nice to know you are working on it. I was bitten by this badly[1], because I assumed the docs where correct although my observing of the logs hinted otherwise. One scenario where you can't keep your requests under 30s is file upload w/ post processing. If for example you have a 2MB image that you need to upload, post-process 4 versions, and then upload to S3, you could be looking at 30-60s per request. Interface-wise this is acceptable (you can fake a file upload bar), but obviously it will either fail, or leave a dyno blocked but available as far as the routing mesh is concerned. If you have 5 dynos serving your app really fast (200ms), and 3 get blocked in a file upload operation for a long time (40secs), then New Relic is going to have field day sending you site downtime alerts... The rack-timeout solution could mitigate the 'false' New Relic alerts, but obviously your file uploads will fail intermittently. -christos [1] http://bit.ly/eqK79C -- You received this message because you are subscribed to the Google Groups "Heroku" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
