Talk:List of HTTP status codes/Archive 3

Official?
Which is REALLY the official source of HTTP response codes? w3c? (as linked to in the article) Or is it the RFC? (http://www.ietf.org/rfc/rfc2616.txt)


 * HTTP status codes need to be registered, IANA maintains the registry at http://www.iana.org/assignments/http-status-codes/http-status-codes.xml. Reschke (talk) 08:18, 19 August 2012 (UTC)


 * Well, the W3C link explicitely says "rfc2616" in it. And the lead for this article notes the official (excluding non-standardized ones) are from the RFC.  So the RFC it is, right?  --Interiot 20:17, 25 May 2006 (UTC)
 * RFC's are official I belive, as the w3c only standardizes (X)HTML, CSS and stuff like that. the actual protocol is all done in RFC's I think. Bawolff 22:33, 4 March 2007 (UTC)


 * RFC's are NOT official unless they are actually approved and accepted into the standards. They are actually proposals, thus their name "request for comments."  Granted, many if not most RFCs do become part of some standard.  A good analogy is RFCs are to standards as regulations are to statutes (i.e. laws). 71.106.210.230 (talk) 22:49, 10 July 2010 (UTC)


 * To become an IETF RFC, a document has been approved by the IESG. It's as official as you get with IETF specs (but there are some differences still, see RFC 2026). Reschke (talk) 08:17, 19 August 2012 (UTC)

Possible / Impossible?
I am looking for some information that may be of use for citations on this page.

We have a bunch of status codes listed, but it isn't clear which ones can actually occur on widely-used web servers (there is a list here). For example, we know that Apache can generate a 200 or a 404 and that there exists no error (internal or in the client request) that will cause Apache to return a 402 or 451, but is there a list somewhere of all the status codes that Apache is capable of returning?

I have been looking at various Apache .htaccess examples and tutorials, and a surprising number of them have an ErrorDocument set for 402, and a lot of them have an ErrorDocument set for every status code Wikipedia lists, including my personal favorite, 418.

TLDR: looking for a list of status codes that Apache, IIS, nginx. etc. are capable of returning. --Guy Macon (talk) 14:37, 1 January 2013 (UTC)


 * (Sound of Crickets...) --Guy Macon (talk) 04:09, 18 June 2013 (UTC)

Code 420
The Twitter API implements code 420 as "you are being rate-limited [on trends and searches]". This seems like a fairly valid use for HTTP response codes, but it is not a formal extension. However, they also named it "Enhance Your Calm" and the choice of number seems to be a reference to 420 in cannabis culture. Should this be added to the list of codes? I note that we have a number of unofficial Microsoft extensions to the response code block listed. 86.14.76.99 (talk) 14:44, 12 April 2010 (UTC)

If this were to be added in some future standard, I would suggest that the official definition be merely a "local definition"; similarly with 419 and 421 (below). In other words, it should be valid but its precise meaning (beyond "error") may be left up to the individual server. I use "420" as an error response to detected malicious web server hack attempts at my server, and on my error list web page, document it as "Malicious Web Robot Response (What are you smoking?)", with the same implication of the meaning of "420" as in the comment above. 71.106.210.230 (talk) 22:36, 10 July 2010 (UTC)

Why was the edit to change 419 and 420, and add 421, to say "Local definition" reversed? As there is no official definition in any RFC but these codes are in use at certain sites, "Local definition" seems to be the most reasonable explanation. 71.105.105.145 (talk) 22:04, 30 July 2013 (UTC)

In Popular Culture
Could we get an "In Popular Culture" section in this article? — Preceding unsigned comment added by 216.175.84.179 (talk) 06:28, 18 January 2014 (UTC)

Explanation of groupings
Scalhotrod has now twice removed an addition to the lead that explains the codes being grouped by overall meaning, thus also giving an explanation for why there are section headings here. It also explained that a minimal web client must process error codes, but may only work from this leading digit.

Should we include an explanation of the overall grouping of HTTP response codes? Andy Dingley (talk) 18:03, 27 October 2014 (UTC)


 * Your list is no different than the Table of Contents which is in roughly the same spot. The TOC has the added benefit of being usefully linked to each section. Please explain why the article needs two TOCs? --SChotrod - Just your average banjo playing, drag racing, cowboy... (Talk) ☮ღ☺ 01:59, 28 October 2014 (UTC)

Code 443
I've removed this as I couldn't find even an unreliable source anywhere. The only references in google to "http error 443" were from fake SEO sites that create bogus response pages to any search term featuring the word "error". I think the original entry must be based on a confusion with TCP port number 443, which is often shown in error messages of the format "(DNS name or IP address):443 uses an invalid certificate". 82.6.102.118 (talk) 08:46, 10 June 2012 (UTC)

I don't know what this code is but this comes from a real error. I typed xxx to remove the real values (confidentiality). Indeed, the 443 is from port [ERROR]  The project xxx:1.0 (/xxx/pom.xml) has 1 error [ERROR]    Non-resolvable parent POM: Could not transfer artifact xxx.xxx.xxx.maven:xxx-master:pom:1.0 from/to central (https://repo.maven.apache.org/maven2): Connect to repo.maven.apache.org:443 [repo.maven.apache.org/23.235.43.215] failed: Connection timed out and 'parent.relativePath' points at wrong local POM @ line 4, column 10 -> [Help 2]

Caperutxa (talk) 12:07, 26 February 2015 (UTC) Caperutxa

Undiscussed deletions
Recently, there have been attempts to delete a Twitter-specific status code while retaining various Nginx-specific and Microsoft-specific status codes. I can see an argument for deleting all status codes that are not found in the RFCs, but I see no justification for only deleting the non-standard status codes used by organizations that you don't like. --Guy Macon (talk) 17:07, 7 August 2013 (UTC)


 * I would guess that the reason behind removing "420 Enhance Your Calm" would be it's not generated by the server itself; rather, by twitter's framework. Should any response codes be listed here, or just ones generated by http servers(regardless of RFC compliance)? - akoimeexx  &laquo;talk&raquo; 20:09, 17 April 2014 (UTC)


 * The reason it should be removed (I've tried myself, but people always foolishly restore it) is that it is not an HTTP status code (as conventionally understood), but something invented by one specific proprietary Web site. 86.159.197.174 (talk) 03:08, 4 August 2014 (UTC)


 * And of course, any Web programmer can return any old custom status they like; it would be ridiculous and futile to try to list them all. This is just a bias by editors who are Twitter fans, I suppose. 86.179.191.90 (talk) 23:45, 27 February 2015 (UTC)

452-463, and 551
A recent edit introduced several new entries to the list, 452-463, and 551, with no mention of source or reference. A brief search for any instances of any of these being used revealed nothing, does anybody have a source for these? — Preceding unsigned comment added by 109.156.231.62 (talk) 20:10, 6 March 2015 (UTC)

I have removed those. Reschke (talk) 21:30, 6 March 2015 (UTC)

New RFCs for HTTP/1.1
Given that the update to RFC 2616 is split into 5 RFCs


 * RFC 7231
 * RFC 7232
 * RFC 7233
 * RFC 7234
 * RFC 7235

the line "Unless otherwise stated, the status code is part of the HTTP/1.1 standard (RFC 7231)." is now incorrect.

RFC7231 defines and links to a status code registry. This appears to be a better reference document.

Furthermore, the following status coded are not defined in RFC 7231, yet the rest of the article does not make this fact known.

RFC 7232 4.1. 304 Not Modified ..........................................18 4.2. 412 Precondition Failed ...................................19

RFC 7233 4.1. 206 Partial Content .......................................10 4.4. 416 Range Not Satisfiable .................................15

RFC 7235 3.1. 401 Unauthorized ...........................................6 3.2. 407 Proxy Authentication Required ..........................6

I'm happy to make the edits to reflect this but unfortunately I could not see how RFC 7321 was linked in the introductory text. Any help would be appreciated.

Darrhiggs (talk) 09:50, 28 April 2015 (UTC)
 * The phrase "RFC XXXX" will automatically generate a link to the RFC page; you don't need to do anything other than specify the name. So you can just type it out and not worry about the linking :) —me_and 10:11, 28 April 2015 (UTC)
 * Attempted edit here, but was reverted. Any thoughts? Darrhiggs (talk) 18:55, 30 April 2015 (UTC)
 * Not immediate thoughts. You're probably better asking the editor who reverted the change. —me_and 19:05, 30 April 2015 (UTC)
 * is there a reason for non-acceptance of the aforementioned edit?

Proprietary 9xx from EZproxy
Hi, you might want to append a pile of EZproxy 9xx codes (or mention the range in total) as described on oclc.org/support/services/ezproxy.

Greetings --PerfektesChaos (talk) 08:34, 9 August 2015 (UTC)