One thing that might help people offer suggestions is to include the whole 'log header', ie the part that lists the latency, cpu ms, api ms, loading request indicator, etc.... If possible you might also include log entries from before the problem started too (preferably for the same request).
Are you guys only seeing increased latency on loading requests or does this also happen on other requests? Do you see 'pending ms' values in the logs? Is the cpu time increased or only the latency? Have you tried using Appstats to profile requests? Have you tried adding a simple handler that outputs 'hello world' as a separate 'app' in your app.yaml? As Denis pointed out, if you're using something like django you should try to see what's getting loaded up / reduce the amount of junk it loads as much as possible. Also, it is very difficult to compare performance on your local machine to performance on production. The SDK uses stubs to mock services, it is single threaded, your CPU is probably much faster, you've (hopefully) got more RAM, not all restrictions are enforced 100% accurately, different version of Python, etc.., etc... All that said, I've had an app that suddenly started experiencing strange times durring startup before. I use webapp as the base framework; I was not have requests failures just unexplainable large latencies loading the app. It sucks, and it is hard to diagnose. Your best bet is to collect as much information (ie log entries) from before / after the problem began so that you've got something showing the issue. For me it just magically cleared itself up after a couple weeks of pain. Robert On Tue, May 24, 2011 at 13:01, Mars <[email protected]> wrote: > I'm seeing a lot of DeadlineExceededError on random import for my app > as well for the last couple of days. Wondering if it's a GAE issue? > > On May 24, 9:41 am, Brandon Thomson <[email protected]> wrote: >> Here is some example logging output, it usually hits DeadlineExceeded on >> some random import statement: >> >> 2011-05-24 09:16:23.034 / 500 488641ms 910cpu_ms 0kb Mozilla/5.0 (Windows NT >> 6.1; WOW64) AppleWebKit/534.24 (KHTML, like Gecko) Chrome/11.0.696.68 >> Safari/534.24,gzip(gfe),gzip(gfe),gzip(gfe) >> E 2011-05-24 09:16:23.024 Traceback (most recent call last): >> E 2011-05-24 09:16:23.024 File >> "/base/python_runtime/python_dist/lib/python2.5/wsgiref/handlers.py", line >> 92, in run >> E 2011-05-24 09:16:23.024 self.result = application(self.environ, >> self.start_response) >> E 2011-05-24 09:16:23.024 File "distlib/tipfy/__init__.py", line 390, in >> __call__ >> E 2011-05-24 09:16:23.024 return self.wsgi_app(environ, start_response) >> E 2011-05-24 09:16:23.025 File "distlib/tipfy/__init__.py", line 430, in >> wsgi_app >> E 2011-05-24 09:16:23.025 rv = self.dispatch(request) >> E 2011-05-24 09:16:23.025 File "distlib/tipfy/__init__.py", line 559, in >> dispatch >> E 2011-05-24 09:16:23.025 self.handlers[handler] = import_string(handler) >> E 2011-05-24 09:16:23.025 File "distlib/werkzeug/utils.py", line 526, in >> import_string >> E 2011-05-24 09:16:23.025 return getattr(__import__(module, None, None, >> [obj]), obj) >> E 2011-05-24 09:16:23.025 File >> "/base/data/home/apps/space-trade/179.350623500077407841/apps/ip/handlers.p >> y", >> line 3, in <module> >> E 2011-05-24 09:16:23.025 import tipfy >> E 2011-05-24 09:16:23.026 DeadlineExceededError > > -- > 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. > > -- 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.
