I'm also having a hard time understanding how the "Total Number of
Instances" page relates to Cold Starts.

My app is all back end processing stuff.  Every 5 minutes, a chained task
issues calls to two separate handlers. (Technically, it's a single chained
task that calls one handler.  After that task is done, it adds a separate
one-time task for a different handler that runs immediately and also
schedules the next 5 minute task.)

That's it.  No front-end or random requests.  Two handlers (one after
another) every 5 minutes.

I can check the "Instances" page right before the next task executes, and I
will see two Dynamic Instances showing.

Right after the chained task runs, I will then see 4 Dynamic Instances.
 Over the next 5 minutes, two of the instances will slowly time out, and it
will settle back to having 2 Instances showing right before the next 5
minute task runs.  (Where it will then start up 2 additional Instances and
show 4 Instances again.)

When I look at the Logs, it shows that each recent request "caused a new
process to be started for your application".

So, the question becomes, if there are indeed 2 Instances already running,
why is Appengine starting 2 additional Instances to service these requests?

Either the instances aren't really there and the Instances page is
inaccurate, or Appengine is wasting resources by starting 2 additional
Instances when it doesn't need to.

Is there any way to clarify what's going on?

On Fri, Dec 10, 2010 at 8:13 PM, Jon McAlister <jon...@google.com> wrote:

> Hi Sergio,
>
> I looked into your app, and unfortunately we had a slight
> configuration error on our part. This has now been fixed and pushed
> out. So, it should be significantly better now. Mind having another
> look?
>
> On Wed, Dec 8, 2010 at 5:49 PM, Sergio Lopes <slo...@gmail.com> wrote:
> > Hi guys
> >
> > I have AlwaysOn enabled since the release, but I'm not getting the
> > expected results. Searching my logs for "loading_request=1"  I found
> > 27 loading requests today. But only 1 of those 27 is an automated /_ah/
> > warmup request, all others are user requests suffering the startup
> > latency.
> >
> > My app (caelumcombr) doesn't have many requests. Before AlwaysOne, the
> > Instances panel always reported only 1 instance for my App. And now
> > that I pay for 3 instances (much more than needed), I'm still seeing a
> > lot of user requests causing instance startup. Isn't that wrong?
> >
> > Sérgio Lopes
> >
> > www.caelum.com.br
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> "Google App Engine" group.
> > To post to this group, send email to google-appeng...@googlegroups.com.
> > To unsubscribe from this group, send email to
> google-appengine+unsubscr...@googlegroups.com<google-appengine%2bunsubscr...@googlegroups.com>
> .
> > For more options, visit this group at
> http://groups.google.com/group/google-appengine?hl=en.
> >
> >
>
> --
> You received this message because you are subscribed to the Google Groups
> "Google App Engine" group.
> To post to this group, send email to google-appeng...@googlegroups.com.
> To unsubscribe from this group, send email to
> google-appengine+unsubscr...@googlegroups.com<google-appengine%2bunsubscr...@googlegroups.com>
> .
> For more options, visit this group at
> http://groups.google.com/group/google-appengine?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To post to this group, send email to google-appeng...@googlegroups.com.
To unsubscribe from this group, send email to 
google-appengine+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/google-appengine?hl=en.

Reply via email to