[google-appengine] Re: OpenID login redirect (/_ah/login_redir?) fails with 500 server errors!
So this is getting worse - domains that used to work yesterday are now getting the 500 error. It looks like a security change that's getting rolled out across domains is causing this. The actual URL that triggers it: https://teacherdashboard.appspot.com/_ah/login_redir?claimid=www.google.com/accounts/o8/site-xrds?hd=MYDOMAIN&continue=https://teacherdashboard.appspot.com/MYDOMAIN Btw - the following doesn't work either: https://teacherdashboard.appspot.com/_ah/login_redir?claimid=www.google.com/accounts/o8/id?hd=MYDOMAIN&continue=https://teacherdashboard.appspot.com/MYDOMAIN (note /o8/site-xrds vs /o8/id) This is quite critical as our users cannot login to GAE apps. Thanks! Jan -- 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-appengine@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.
[google-appengine] OpenID login redirect (/_ah/login_redir?) fails with 500 server errors!
We're seeing consistent OpenID redirect 500 server failures for specific Google Apps domains. OpenID works for other domains just fine, but we're seeing an increasing number of domains where the redirect just dies. Fix ETA please? Jan -- 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-appengine@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.
[google-appengine] Seeing huge delays on task queue execution - 13sec+ per task
Is anyone else seeing these delays? Jan -- 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-appengine@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.
[google-appengine] Re: Task Queue REST API and push queues?
The status API works but only for pull queues... It would be hugely useful to have access info to this for all queues. @Brandon: can you explain what you mean by wrapper? What would a wrapper accomplish in this case? Jan On Feb 2, 11:21 pm, Robert Kluin wrote: > Hi Jan, > The REST API is only for pullqueues. I don't believe there's > currently any way to get info about tasks in your pushqueues. > > There is / was a status / quota API discussed at last year's IO, but > I don't think it has been released. It might be in TT status, but I > can't remember for sure. > > Robert > > On Fri, Feb 3, 2012 at 02:07, Jan Zawadzki / Hapara > > > > > > > > wrote: > > The REST API does not appear to allow access to the pushqueues. It > > would be hugely valuable to get at least the queue state info for all > >queues- is there some magic incantation that allows this, or is this > > in the works? > > > Has anyone seen examples of this that work against pushqueues? > > > Thanks! > > > Jan > > > -- > > 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-appengine@googlegroups.com. > > To unsubscribe from this group, send email to > > google-appengine+unsubscr...@googlegroups.com. > > For more options, visit this group > > athttp://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-appengine@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.
[google-appengine] Tasks in queue counts wildly inflated (1000 => 700k)
So the queue task counts seem to be pulling wrong queues, or
miscalculating the number by a huge margin at times.
We've observed queues with ~100 tasks displaying counts of 700,000+.
When the incorrect count is displayed it decrements correctly as tasks
execute. Pausing and restarting the queue has no effect.
It's not a show stopper but it's disconcerting :) ('better bump that
quota again... ugh...')
Jan
--
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-appengine@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.
[google-appengine] Task Queue REST API and push queues?
The REST API does not appear to allow access to the push queues. It would be hugely valuable to get at least the queue state info for all queues - is there some magic incantation that allows this, or is this in the works? Has anyone seen examples of this that work against push queues? Thanks! Jan -- 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-appengine@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.
[google-appengine] Deployed versioning and Chrome "this is probably not the site you are looking for"
As of quite recently the Chrome build on the Mac will not allow you to get past the red "this is probably not the site you are looking for" page. Even with phishing and malware detection disabled, Chrome displays this page, but now, only with a Back button - there is no longer a "Proceed Anyway" button on the page! This obviously is a pretty big problem for testing pre-releases from Chrome... Any ideas Jan -- 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-appengine@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.
[google-appengine] Masses of Deadline exceptions on imports
Hi - we're experiencing > 50% of requests failing with deadline errors, across a range of apps. It's effectively a service outage as far as we're concerned Would be great to get some confirmation from Google that this is being investigated? Jan -- 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-appengine@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.
[google-appengine] Re: Gurantee log
Most people would expect to see a proper database record for any sort of financial tracking. Log files aren't robust enough On Oct 1, 9:02 am, Timofey Koolin wrote: > I have billing system on GAE. > It is thing to have backup of my datastore. Now I have small record count > and can read all of it for backup every day. But in future - I need any > incremental method of backup my data. > I have 3 things: > - log operations in logging system and replicate on my server. > - log operations in log kind and replicatite on my server (but it add pay > for write, read and delete operations for each logical change in datastore). > - save timestamp field in my entities and log of deleted object. > > save data small in log - cheap and faster then write it in datastore. > > 2011/9/29 Jan Zawadzki / Hapara > > > > > > > > > > > This only works if you don't have a lot of traffic. > > > As Ikai mentioned, these are limited-size FIFO logs, so if you start > > hitting any significant traffic, you will find that your INFO logs > > hold a minute's worth of data - longer than a transfer would take you. > > > Write to a DB, OR, use RPC and URL fetch to "ping" an external server > > that will store the Apache log files for you. > > > J > > > On Sep 28, 10:06 am, rekby wrote: > > > Do log record gurantee for each request - I want write into log > > instructions > > > for incremental change data and make incremental backup on my computer by > > > download and parse log messages instead of analise datastore. > > > -- > > 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-appengine@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. > > -- > С уважением, > Кулин Тимофей. > > Телефон: +7 (4852) 974793 > ICQ: 114902104 > email: timo...@koolin.ru > Blog:http://timofey.koolin.ru -- 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-appengine@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.
[google-appengine] Re: Gurantee log
This only works if you don't have a lot of traffic. As Ikai mentioned, these are limited-size FIFO logs, so if you start hitting any significant traffic, you will find that your INFO logs hold a minute's worth of data - longer than a transfer would take you. Write to a DB, OR, use RPC and URL fetch to "ping" an external server that will store the Apache log files for you. J On Sep 28, 10:06 am, rekby wrote: > Do log record gurantee for each request - I want write into log instructions > for incremental change data and make incremental backup on my computer by > download and parse log messages instead of analise datastore. -- 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-appengine@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.
[google-appengine] Re: OpenID authentication gives 500 Server Errors?
Opened a production issue if anyone else is experiencing this: http://code.google.com/p/googleappengine/issues/detail?id=5910 On Sep 16, 1:47 pm, "Jan Zawadzki / Hapara" wrote: > Hi - we're seeing 500 Server Errors on all login attempts. It dies on > the login_redir call: > > myapp/_ah/login_redir?claimid=google.com/accounts/o8/site-xrds? > hd=blahblah&continue=noyoudont > > We're seeing this across all apps at the moment...? -- 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-appengine@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.
[google-appengine] OpenID authentication gives 500 Server Errors?
Hi - we're seeing 500 Server Errors on all login attempts. It dies on the login_redir call: myapp/_ah/login_redir?claimid=google.com/accounts/o8/site-xrds? hd=blahblah&continue=noyoudont We're seeing this across all apps at the moment...? -- 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-appengine@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.
[google-appengine] Re: Python library for uploading to AppEngine
At the risk of asking the obvious, you mean other than appcfg.py? J On Sep 11, 4:48 pm, Emlyn wrote: > Is there a python library for uploading code to AppEngine? I'd like to > build an App that can upload code to other apps, has anyone done this > already? > > -- > Emlyn > > http://my.syyn.cc- Synchonise Google+, Facebook, WordPress and Google > Buzz posts, > comments and all.http://point7.wordpress.com- My blog > Find me on Facebook and Buzz -- 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-appengine@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.
[google-appengine] Re: Loads of Exceeded soft process size limit errors? (Python app)
Hmm. We don't do enough of that, plus, it's a new problem - months of incremental code changes w/ no problems and suddenly our memory usage triggers errors. I think something else changed in the GAE environment... J On Sep 2, 1:35 am, Tim Hoffman wrote: > HI Jan > > Something to watch for on the memory front, is if you are doing a lot of > string manipulation > using '+' operator. I have seen someone here last year using + rather than > join and > hitting soft and hard memory limits with a single request, and changing to > join reduced > memory use by 100 times. Its an extreme example I know. > > Also if you are holding references to entities in modules, global, etc... > things may not get garbage collected. > > Have a look at apptracehttp://pypi.python.org/pypi/apptrace/0.2.3 which > allows you to monitor memory use in appengine > > Rgds > > T -- 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-appengine@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.
[google-appengine] Re: "Enforced Rate" in task queues ruining performance
Same here Johan - I think this is system-wide. It is definitely tied to the new application settings, in particular the min. pending latency. This sort of stuff seriously needs to be documented better folks - the ones affected by it the most are people using App Engine in earnest; for us and possibly for others this is a system-down sort of event. Jan On Aug 3, 2:08 am, Johan Euphrosine wrote: > Can you both share your appid, so we can investigate deeper ? > > Also feel free to fill a production issue about it if this is directly > affecting your > operation:http://code.google.com/p/googleappengine/issues/entry?template=Produc... > > > > > > > > > > On Tue, Aug 2, 2011 at 3:08 PM, Dan Lines wrote: > > I am having the same problem. Today has been the worst performance so far > > with theenforcedratedown at 0.16/s. The application is basically non > > functional at thisrate > > > Could we please get some information on this. > > > - Dan > > > -- > > You received this message because you are subscribed to the Google Groups > > "Google App Engine" group. > > To view this discussion on the web visit > >https://groups.google.com/d/msg/google-appengine/-/6cczXBYIK6IJ. > > To post to this group, send email to google-appengine@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. > > -- > Johan Euphrosine (proppy) > Developer Programs Engineer > Google Developer Relations -- 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-appengine@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.
[google-appengine] Loads of Exceeded soft process size limit errors? (Python app)
Hi - we're seeing tons of Exceeded soft process size limit errors appear in the last 24 hours. I have not seen these before, ever. These kick in after anywhere from 20 to 700 requests. Should I worry? Not exactly much we can do about memory leaks here, is there? (Python app) Thanks! Jan -- 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-appengine@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.
[google-appengine] Re: Datastore admin still broken with "too many redirects"
Ping. Is it just our domain that's broken or is someone else experiencing this...? Thanks! Jan On Aug 18, 9:44 am, "Jan Zawadzki / Hapara" wrote: > Any ETA on getting this fixed? > > Breaks on every one of our apps redirecting > throughhttps://appengine.google.com/_ah/conflogin. > > Does not seem to be browser related (Chrome, FF, Safari) > > It's a bit of a pain managing mapreduce manually, would love to have > access to this functionality! > > Jan -- 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-appengine@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.
[google-appengine] Re: High-performance, high-precision sharded counters in the HRD
Memcache incr: we have upwards of 200 instances doing this with no
apparent performance penalty.
As Robert pointed out, there are occasions when a value "sticks", so
you will need a way to deal with this gracefully.
J
On Aug 21, 8:50 am, Jeff Schnitzer wrote:
> The short version boiled down to two questions:
>
> 1) What is the maximum realistic throughput for memcache increment()
> on a single counter?
>
> 2) The documentation for the HRD says that queries across entity
> groups are eventually consistent. Does this extend to __key__ queries
> as well? For example, will this be eventually or strongly consistent?
>
> SELECT * FROM Thing WHERE __key__ >= KEY('Thing', 10) AND __key__ <=
> KEY('Thing', 20)
>
> -
>
> The long version, to hopefully open a discussion:
>
> I'm selling various types of virtual goods, each of which have a fixed
> number available. The constraints for each product are:
>
> * I can sell up to exactly N items
> * I cannot under any circumstances sell N+1
> * For some products, N might be hundreds of thousands
> * The peak purchase rate is expected to be hundreds per second when a
> highly anticipated product goes on sale. Maybe thousands per second.
>
> I have thought of several strategies for building this system, but the
> best one requires a counter that can:
>
> 1) Be updated hundreds of times per second (thousands?)
> 2) Give me a transactionally safe count, or at least temporarily err
> on the side of overcounting (not undercounting!)
>
> Between memcache and the datastore, I can build this:
>
> * Memcache increment() gives me a current atomic count of purchases.
> * Sharded counters give me a scalable persistent record of purchases
> (the purchase record can be parented by the count shard and updated in
> the same transaction)
> * Memcache putIfUntouched() lets me safely initialize the memcache
> value as long as I can get a strongly consistent count from the shards
> * By ordering memcache increments *before* shard increments for
> purchases, and shard decrements *before* memcache decrements for
> refunds, I can guarantee that datastore failures will at worst cause
> overcounts, which is fine. A timeout on the memcache value means it
> will eventually become accurate again.
>
> First question: What is the practical throughput of
> memcache.increment() on a single counter? Thousands per second?
>
> Second: The key is being able to get a strongly consistent sum of the
> sharded counters. The example of a sharded counter in the appengine
> documentation uses a query to fetch the shards, which will only work
> on the M/S datastore.
>
> Since get() is strongly consistent, the obvious solution is to give
> the shards predictable names and use batch get() for XYZ-shard1,
> XYZ-shard2, ... XYZ-shard100. Ok, but it's not quite as nice as being
> able to say query for __key__ >= XYZ-shard1 and __key__ <=
> XZY-shard100 because some of the shards might not exist... or I might
> want to decrease the number of shards for a running system. Should I
> just not care about this case?
>
> I guess I've probably answered question #2... but if anyone else has
> any suggestions for how to build a system like this, I'd love to hear
> your thoughts.
>
> Thanks,
> Jeff
--
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-appengine@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.
[google-appengine] Datastore admin still broken with "too many redirects"
Any ETA on getting this fixed? Breaks on every one of our apps redirecting through https://appengine.google.com/_ah/conflogin. Does not seem to be browser related (Chrome, FF, Safari) It's a bit of a pain managing mapreduce manually, would love to have access to this functionality! Jan -- 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-appengine@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.
