Naked domain without any page in url takes too long to load

Question

I have 2 different servers. Old one is Win 2012, new is Win 2016. I shifted a few of domains from old server to new Win 2016 Server. Now the issue I am facing on the new server is that whenever someone tries to open the domains with any page urls its working fine e.g. http://example.com/page1.aspx or http://example.com/Default.aspx. But when naked domain url (e.g. http://example.com) is opened it takes randomly 2-4 minutes to load the page.

All the domains have Default.aspx in their root and in IIS also it is set on top priority. I've checked server logs and found nothing. I also removed all other documents types from default document list on IIS. Also tried setting it in the web.config file but could not identify the issue.

In the application startup I also tried to log data to at various steps in Global.asax but to no avail.

The same application is working fine in the old Win 2012/IIS 8.5. I wonder if there is any setting/config in Win 2016/IIS 10.0? Application targets .NET Framework 4.8.

Answer 1

One excellent tool for helping to diagnose this is curl along with a --write-out format and the -I and -L flags.

If you use the -I option curl will provide the response headers for each completed request.

The -L flag will make sure that curl follows redirects.

The -w"@/path/to/format/file.txt" flag will allow you to specify a format file that provides a plethora of information about the final request, including some fantastic timing information. This granular timing information can help you diagnose where in the request process the delay is being introduced.

• Notice that your time_appconnect time is excessively long? It could be that your webserver is having an issue with SSL stapling.
• What about a long time_namelookup? Check to ensure that dns resolution is quick.
• Is time_redirect taking forever? Maybe you should check out what's preventing redirects from working properly.
• etc etc etc

Here is an example format that you can use with curl: https://gist.github.com/manifestinteractive/ce8dec10dcb4725b8513

And here is an example of a request to google using that format and the options I explained above:

➜  ~ curl -IL -s -w"@sniff.txt" google.com
HTTP/1.1 301 Moved Permanently
Location: http://www.google.com/
Content-Type: text/html; charset=UTF-8
Date: Mon, 24 Aug 2020 01:24:17 GMT
Expires: Wed, 23 Sep 2020 01:24:17 GMT
Cache-Control: public, max-age=2592000
Server: gws
Content-Length: 219
X-XSS-Protection: 0
X-Frame-Options: SAMEORIGIN

HTTP/1.1 200 OK
Content-Type: text/html; charset=ISO-8859-1
P3P: CP="This is not a P3P policy! See g.co/p3phelp for more info."
Date: Mon, 24 Aug 2020 01:24:18 GMT
Server: gws
X-XSS-Protection: 0
X-Frame-Options: SAMEORIGIN
Transfer-Encoding: chunked
Expires: Mon, 24 Aug 2020 01:24:18 GMT
Cache-Control: private
Set-Cookie: 1P_JAR=2020-08-24-01; expires=Wed, 23-Sep-2020 01:24:18 GMT; path=/; domain=.google.com; Secure
Set-Cookie: NID=; expires=Tue, 23-Feb-2021 01:24:18 GMT; path=/; domain=.google.com; HttpOnly


=============  HOST:  ==========

           local_ip:  127.0.0.1
         local_port:  65084
          remote_ip:  redacted
        remote_port:  80

=======  CONNECTION:  ==========

       http_version:  1.1
          http_code:  200
       http_connect:  000
       num_connects:  2
      num_redirects:  1
       redirect_url:

=============  FILE:  ==========

       content_type:  text/html; charset=ISO-8859-1
 filename_effective:
     ftp_entry_path:
      size_download:  0
        size_header:  1013
       size_request:  154
        size_upload:  0
     speed_download:  0.000
       speed_upload:  0.000
  ssl_verify_result:  0
      url_effective:  http://www.google.com/

===  TIME BREAKDOWN:  ==========

    time_appconnect:  0.000000
       time_connect:  0.137004
    time_namelookup:  0.136557
   time_pretransfer:  0.137063
      time_redirect:  0.596540
 time_starttransfer:  0.409782
                      ----------
         time_total:  1.006389

For more information about what you can squeeze out of the curl format configuration, read the --write-out part of the manual page: https://curl.haxx.se/docs/manpage.html

Also of note, I'm not sure if you can do this reliably on Windows. I think it is possible to run curl on Windows, but I've never once attempted to do so so YMMV there. I very highly recommend using a Linux distro instead.