Sudden redirect to Turing anti-bot test during remote api session

87 views
Skip to first unread message

Denis Volokhovskiy

unread,
Jan 25, 2011, 11:15:53 AM1/25/11
to Google App Engine
Hi Google Team,

I have performed datastore maintenance from the remote api console for
python app (likeourselvesapp),
much records were processed in cycle by single requests (I think about
100),
and suddenly have encountered automatic redirect (HTTP 302) to Turing
test page (with captcha entering)
right from the remote shell.
Then I exited and tried to deploy, and here traces for the same
denial:

Application: likeourselvesapp; version: v132-2.
Server: appengine.google.com.
Scanning files on local disk.
Scanned 500 files.
Initiating update.
Error 302: --- begin server output ---
<HTML><HEAD><meta http-equiv="content-type" content="text/
html;charset=utf-8">
<TITLE>302 Moved</TITLE></HEAD><BODY>
<H1>302 Moved</H1>
The document has moved
<A HREF="http://sorry.google.com/sorry/?continue=http://
appengine.google.com/api/appversion/create%3Fversion%3Dv132-2%26app_id
%3Dlikeourselvesapp">here</A>.
</BODY></HTML>
--- end server output ---

I tried to open specified URL with Turing test and entering captcha -
but it still abandoned my further requests.

After several minutes problem gone,
but such delays is harmful for normal maintenance from remote console.

We are billing-enabled app, and I'm registered developer for a long
time.
Why such problem happens, and is any solution to avoid it?

Thank you

Denis Volokhovskiy

unread,
Jan 25, 2011, 2:06:17 PM1/25/11
to Google App Engine
After some time, have retried the same session - without this issue.

Looks like it is rare case.






On Jan 25, 6:15 pm, Denis Volokhovskiy <altitudebre...@gmail.com>
wrote:

Waleed Abdulla

unread,
Jan 27, 2011, 4:26:05 AM1/27/11
to google-a...@googlegroups.com
I just started getting the same exact issue about 15 minutes ago and it brought down my app. 

I make a lot of API calls from my main server to my app engine app, and now the app engine is rejecting all these requests. Issue can be replicated from the command line as follows:


Which returns:

[1] 22560
[root@web1 temp]# --2011-01-27 03:20:53--  http://api.networkedblogs.com/api/posts.getBlogPosts?bid=343457
Resolving api.networkedblogs.com... 74.125.53.121
Connecting to api.networkedblogs.com|74.125.53.121|:80... connected.
HTTP request sent, awaiting response... 302 Found
Resolving sorry.google.com... 74.125.53.105, 74.125.53.106, 74.125.53.147, ...
Connecting to sorry.google.com|74.125.53.105|:80... connected.
HTTP request sent, awaiting response... 503 Service Unavailable
2011-01-27 03:20:57 ERROR 503: Service Unavailable.

The same command above succeeds if I run it from my local machine, but it fails if I run it from my Web server. The app engine seems to be falsely considering my other server as a DoS attempt and blocking all requests from it.

This is a blocker issue. Any help from the App Engine team is greatly appreciated. 

Regards,
Waleed





--
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-a...@googlegroups.com.
To unsubscribe from this group, send email to google-appengi...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/google-appengine?hl=en.


Waleed Abdulla

unread,
Jan 27, 2011, 5:13:32 AM1/27/11
to google-a...@googlegroups.com
Update: I put my php app in maintenance mode so it's not accessible and doesn't make any calls to my GAE app, and left it for 10 minutes, then enabled it again, and now GAE is accepting requests from the PHP server. Odd, but worked for me. 

Tim Hoffman

unread,
Jan 27, 2011, 9:07:38 AM1/27/11
to google-a...@googlegroups.com
Hi

I ran into this a long time ago and I found it occurred if I wasn't reusing the remote_api handle.

I had been recreating the remote_api handle on each transaction, which is inefficient and after 200 transactions I would be
redirected to a recaptcha service so I could prove I wasn't a bot ;-)

Rgds

T

Waleed Abdulla

unread,
Feb 15, 2011, 5:07:41 AM2/15/11
to google-a...@googlegroups.com
I'm getting the same issue again today. It's happening more often recently, which is worrying. I put my other server in maintenance more so it doesn't call App Engine hoping that after a few minutes GAE would stop blocking my IP, but so far that's not helping either. I'll try waiting a longer time. 

Tim,
   Thanks for the tip, but in my case my calls don't require authentication. I'm calling public APIs that I host on the App Engine over http and without any kind of authentication. So I guess that issue in the thread you referred to doesn't apply. 

    Any way to white-list my server's IP address so it doesn't get blocked like this? 

Regards,
Waleed





T

--

Tim Hoffman

unread,
Feb 15, 2011, 5:19:09 AM2/15/11
to google-a...@googlegroups.com
Hi 

Maybe you should require auth, we never get the turing bot now that I fixed the connection

T

Waleed Abdulla

unread,
Feb 15, 2011, 5:24:50 AM2/15/11
to google-a...@googlegroups.com
Good point. I'll give that a shot. 

Regarding the current issue, I put my other server on hold for a while and tried again, and GAE kept rejecting my calls. I tried that a few times, waiting for more than 10 minutes between tests and it still failed. Finally, I tried uploading my app to GAE again, and that fixed it. I'm guessing uploading a new version restarts the counter for whatever system was falsely considering me like a bot. 

Regards,
Waleed





T

--

Nickolas Daskalou

unread,
Feb 15, 2011, 6:49:14 AM2/15/11
to google-a...@googlegroups.com
Hi Waleed,

Approximately how many requests/minute was your other server making to your GAE app before it was blocked? Would be interesting to know what the cut-off point is.

Nick
Reply all
Reply to author
Forward
0 new messages