500 Errors (DeadlineExceededError) on module imports and other random bits of code

[PandaSuit]

Anyone else seeing a large amount of 500s (DeadlineExceededError)
lately on module imports?

Over the last few days parts of my application that have never
generated errors (and have not changed recently) have been throwing
500 errors and timing out. Often it is the first import of a module
that times out or it may time out in the middle of an SDK function. It
is usually never the same line of code from error to error either.

[Atif]

Hi Panda

Yes, we are also facing the same issue on most of our servers hosted on app-engine. Same thing was happened about 16 hours ago as well. 

It is very strange that no google officials normally respond to such emails.

regards,

Atif 

[Dan]

I've seen the same problems. It actually forced my to start paying
for the Always-On feature. Otherwise my users would be waiting for
20-30 seconds for a request if it worked at all.

It might be related to issue 1695: http://code.google.com/p/googleappengine/issues/detail?id=1695

Dan

[radomir]

We also see deadline exceptions and Memcache was throwing exceptions
for some time. What's interesting is that in those cases capability
API was returning DISABLED for all checks.

[John Wheeler]

And make the mail.Send() calls too. We've been getting these consistently when we never did before:

The API call mail.Send() took too long to respond and was cancelled.

[John Wheeler]

Came to the newsgroup for this exact issue as a matter of fact.

[Ronoaldo José de Lana Pereira]

I'm facing this too recently.

[tempy]

Me three, its been bad the last couple of days for me. I just got a
deadlineexceeded on the Users API, which, obviously, breaks my app
completely.

On Jun 10, 1:15 pm, Ronoaldo José de Lana Pereira

[PandaSuit]

The errors have been less frequent lately but I am still getting them.

I submitted a high priority production issue to the issue tracker on
the 7th and have not heard anything back from Google.

Please star the issue and add your information and maybe we can get
some attention:
http://code.google.com/p/googleappengine/issues/detail?id=5158

[Brandon Thomson]

I have been experimenting with deploying a new version and switching the default when this happens; seems to fix the problem for a while. YMMV

[Tom Phillips]

I've been seeing more and more of these DeadlineExceededError
exceptions over time especially in warmup requests.

My /_ah/warmup should finish in about 6s (PersistenceManagerFactory,
Jaxb). And I'm seeing these exceptions on user requests too even
though the average latency is usually ~50ms.

Java, Master/Slave.

Not sure with these problems how my app will scale beyond the Always-
on instances.

[Kenneth]

I'm seeing a lot of these recently as well.  I assume we're all on m/s datastore and that's where the problem lies.  

Come on Google, give us a least a hint that you know there's a problem.  What happened to those better hr migration tools we were promised?

[pinhopro]

I also stared to get import errors. 

It started yesterday around  10pm ( EST ) 

See the logs

2011-06-16 08:32:33.467

<type 'exceptions.ImportError'>: cannot import name service_handlers
Traceback (most recent call last):
  File "/base/data/home/apps/bookcalc/1.351177451425526630/services.py", line 12, in <module> 

from protorpc import service_handlers 

It is not a deadline or timeout, it is a import error.

I redeployed the application, and didn't help.

I built the application using google protorpc, but now, nothing is working.  

[Soshi HANAMURA]

I'm getting same errors today.
Errors have obviously increased about 2 hours ago, June 30 22:00
(UTC).
My Applications are using python and M/S datastore.

Anyone else?