# HTTP status code 499 from long running requests **URL:** \ **Category:** NGINX\ **Created:** [June 2, 2013, 3:10am UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138 "2013-06-02T03:10:10Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![addis\_a](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/c4cdca/32.png) [@addis\_a](https://www.ruby-forum.com/u/addis_a)\ **Post date:** [June 2, 2013, 3:10am UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138/1 "2013-06-02T03:10:10Z")
Hello, I have long running requests upwards of five minutes that are called via ajax (jQuery) (XHR), and hitting a backend PHP script. I am seeing very strange behavior, intermittently calling the script twice, even though in Chrome developers tools I only see a single XHR request in the network pane. I know that it is running twice because in my backend PHP script I write a log to a MySQL database, and I am seeing two entries. I took a look at the nginx access log, and seeing HTTP status code 499, which seems to be: client closed connection before nginx was able to send anything to client. What is the best way to fix this? Is there a config setting in nginx to keep the connection open? Or is the client terminating the request, and then reissuing the request, thus way I am seeing two entries? Thanks for the help with this. Posted at Nginx Forum: > **[HTTP status code 499 from long running requests](https://forum.nginx.org/read.php?2,239759,239759#msg-239759)** > > Hello, I have long running requests upwards of five minutes that are called via ajax (jQuery) (XHR), and hitting a backend PHP script. I am seeing very strange behavior, intermittently calling the script twice, even though in Chrome developers tools... ---
**Author:** ![justin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/3ab097/32.png) [@justin](https://www.ruby-forum.com/u/justin)\ **Post date:** [June 2, 2013, 1:04pm UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138/2 "2013-06-02T13:04:55Z")
Hello! On Sat, Jun 01, 2013 at 09:09:55PM -0400, justin wrote: > was able to send anything to client. > > What is the best way to fix this? Is there a config setting in nginx to keep > the connection open? Or is the client terminating the request, and then > reissuing the request, thus way I am seeing two entries? The 499 code means that _client_ closed connection, and there is more or less nothing you can do from nginx side. You have to either find a way to ask client to wait longer before it gives up, or return something before it gives up. – Maxim D. > **[nginx: donation](http://nginx.org/en/donation.html)** ---
**Author:** ![justin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/3ab097/32.png) [@justin](https://www.ruby-forum.com/u/justin)\ **Post date:** [June 4, 2013, 6:52am UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138/3 "2013-06-04T06:52:26Z")
Maxim, Does it seem reasonable that the 499 status codes are responsible for double requests? I.E. do you think the client is resending the request after the 499 status? ## Maxim D. Wrote: > > even though in Chrome developers tools I only see a single XHR > > – > Maxim D. > [nginx: donation](http://nginx.org/en/donation.html) > > * * * > > nginx mailing list > [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) > [mailman.nginx.org archive](http://mailman.nginx.org/mailman/listinfo/nginx) Posted at Nginx Forum: [http://forum.nginx.org/read.php?2,239759,239819#msg-239819](http://forum.nginx.org/read.php?2,239759,239819#msg-239819) ---
**Author:** ![justin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/3ab097/32.png) [@justin](https://www.ruby-forum.com/u/justin)\ **Post date:** [June 4, 2013, 1:28pm UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138/4 "2013-06-04T13:28:36Z")
Hello! On Tue, Jun 04, 2013 at 12:52:03AM -0400, justin wrote: > Does it seem reasonable that the 499 status codes are responsible for double > requests? I.E. do you think the client is resending the request after the > 499 status? No. But both 499 status codes (i.e. client closed connection) and double requests are likely happen because client times out. – Maxim D. > **[nginx: donation](http://nginx.org/en/donation.html)** ---
**Author:** ![justin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/3ab097/32.png) [@justin](https://www.ruby-forum.com/u/justin)\ **Post date:** [June 7, 2013, 4:42pm UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138/5 "2013-06-07T16:42:46Z")
Hello! On Fri, Jun 07, 2013 at 12:23:34AM -0400, justin wrote: > XX.XXX.XXX.253 - - [06/Jun/2013:21:11:32 -0700] “POST /actions/execute.php > HTTP/1.1” 200 673 “[https://dev.mydomain.com/execute](https://dev.mydomain.com/execute)” “Mozilla/5.0 > (Macintosh; Intel Mac OS X 10\_8\_4) AppleWebKit/537.36 (KHTML, like Gecko) > Chrome/27.0.1453.110 Safari/537.36” Just a side note: from the log there is no reason to assume these requests are duplicate. > I obviously need a solution, since I am essentially creating duplicate > transactions. This only happens when the execute.php takes a long time > though. In jQuery when I am making the AJAX request I am setting the timeout > to 15 minutes. Is there anything else I can look at or try? I would recommend reproducing the issue on a client, and looking into network traffic via tcpdump / browser details via built in tools to see why duplicate requests are sent. On the other hand, duplicate POSTs aren’t at all new - they used to happen due to users pressing “Submit” button multiple times. An obvious solution is to use some one-time form tokens. – Maxim D. > **[nginx: donation](http://nginx.org/en/donation.html)** ---
**Author:** ![justin](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/3ab097/32.png) [@justin](https://www.ruby-forum.com/u/justin)\ **Post date:** [June 7, 2013, 6:24am UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138/6 "2013-06-07T06:24:05Z")
Maxim, I dug a bit deeper and here is the nginx access log showing the 499 status request, and then a repost of the same request: XX.XX.XXX.253 - - [06/Jun/2013:21:09:08 -0700] “POST /actions/execute.php HTTP/1.1” 499 0 “[https://dev.mydomain.com/execute](https://dev.mydomain.com/execute)” “Mozilla/5.0 (Macintosh; Intel Mac OS X 10\_8\_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/27.0.1453.110 Safari/537.36” XX.XXX.XXX.253 - - [06/Jun/2013:21:11:32 -0700] “POST /actions/execute.php HTTP/1.1” 200 673 “[https://dev.mydomain.com/execute](https://dev.mydomain.com/execute)” “Mozilla/5.0 (Macintosh; Intel Mac OS X 10\_8\_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/27.0.1453.110 Safari/537.36” I obviously need a solution, since I am essentially creating duplicate transactions. This only happens when the execute.php takes a long time though. In jQuery when I am making the AJAX request I am setting the timeout to 15 minutes. Is there anything else I can look at or try? Thanks for the help. Posted at Nginx Forum: > **[Re: HTTP status code 499 from long running requests](https://forum.nginx.org/read.php?2,239759,239943#msg-239943)** > > Hello, I have long running requests upwards of five minutes that are called via ajax (jQuery) (XHR), and hitting a backend PHP script. I am seeing very strange behavior, intermittently calling the script twice, even though in Chrome developers tools... ---
**Author:** ![shabijkh](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/90db22/32.png) [@shabijkh](https://www.ruby-forum.com/u/shabijkh)\ **Post date:** [May 31, 2021, 7:43am UTC](https://www.ruby-forum.com/t/http-status-code-499-from-long-running-requests/235138/7 "2021-05-31T07:43:16Z")
Hi I have faced this problem recently. Could you solve it? I’m working on financial systems and running twice of transactions is real problem for me