In my experience you can't trust the eta parameter. It's usually
pretty close but often tasks will be as many as 10 minutes late.

On Jun 11, 10:46 am, james lesorg <[email protected]> wrote:
> I really enjoyed this talk. Would make a great article.
>
> I wonder whether i understand the motivation behind "sequence" numbers
> correctly: we cannot just timestamp worker entities and then churn
> through them chronologically because instance times can be wildly out
> of sync? If that's the case, how can i trust the eta parameter when
> creating task objects?
>
> Is the possibility of a worker / task race condition really greater
> than the risk of the memcache lock disappearing? Is it at all possible
> that the memcache lock will disappear in this scenario, and if so how
> can we survive this edge case?
>
> Sorry if this is a duplicate post - It seems the question i asked
> earlier got lost in google groups weirdness.

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" 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/google-appengine?hl=en.

Reply via email to