I'm seeing this in Python too, not sure when it started. I'm chaining
tasks together to loop over a large set of data so there is only one
task executing at a time. Some execute right after the previous one
completes (as expected and as it worked in the past) while others take
between 15-30 seconds to start. There are no TransientErrors being
throw and all the tasks have unique names.

Ikai - is capacity defined for each App or for GAE as a whole?

I starred the issue. IMO, this is not a small problem

thanks
brian

On Mar 7, 1:34 am, kazunori_279 <[email protected]> wrote:
> Hi,
>
> I've experienced the same phenomenon so that I added a new issue for
> this:
>
> http://code.google.com/p/googleappengine/issues/detail?id=2924
>
> Thanks,
>
> Kaz
>
> On Feb 26, 8:05 am, Jerome <[email protected]> wrote:
>
>
>
> > Hello,
>
> > We have been usingtaskqueues for a few months without problem. We
> > handle actions that must execute right away.
>
> > At the time of the release of 1.3.1, we noticed that sometimes, some
> > tasks will take 1 to 2 minutes to start, but in thetaskqueue, the
> > ETA shows as a date in the past (like 0:01:12 ago), usually 1 second
> > or so after the creation time.
>
> > This is not a consistent issue. I might works perfectly (taskbeing
> > kicked in within 1 second of creation) for several times and suddenly,
> > the nexttasktakes 1.5 minutes to get started.
>
> > Ourtaskqueueis empty when we see this issue.
> > Ourqueueis setup with a max rate of 5/s and a bucket size of 1.
>
> > Any idea of what we can do to work around or solve this new issue?
>
> > Jerome

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine for Java" 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-java?hl=en.

Reply via email to