This is the possible errors you may get in HTML LinkChecker. They are shortly described by selecting a line in the list and selecting ”Show Information” from the “File”-menu, but even more described here.
Not found
The most common error, indicating that a local link is incorrect. The linked file does not exist.
Not exist
This means that a local link with an anchor, like “file.html#anchor”, points to an existing file but that the anchor does not exist on the linked page. Clicking the link in a browser will take you to the linked page, but not to the right part of it. An anchor is usually defined with the tag <A NAME=nameOfAnchor>text</a>. The HTML 3.0 standard also allow other solutions, such as <P NAME=nameOfAnchor>, but HTML LinkChecker only cares about <A NAME ...>. In other words, the link may work correctly in newer browsers even if it's reported as incorrect.
204 error
This, and all the following errors, is an external error indicating that something is wrong with a link to a page outside your computer (which means, on the net). The 204 code means that the linked file exist, but that it is empty. You maybe have got this error in your browser sometimes, and like then, it is often a temporarily error. Open the “Full Information”-dialog, click “Try again”, and you'll in most cases get a different result.
301 error
This means that the page has moved. Clicking the link in a browser (not in the earliest versions though) will take you to the new page anyway, but it would be a good idea to link to the right page instead of the old one. In the “Full Information”-dialog you can find out the new URL. A common 301 error is a link to, for example, “http://www.server.com/folder” which has moved to “http://www.server.com/folder/”. No big difference, but it's still better to use the second URL instead of the first one. (To be technical, browsers that only supports HTTP/0.9 won't find the new page unless the server software is very understandable).
400 error
This is “Bad request”, a pretty rare error. 4xx errors actually means that there is something wrong in the command to the remote server, and it is probably something wrong in the command sent by HTML LinkChecker which is not understood by the server software. Probably because of a bug in either the program or in the server. Probably you should tell me about it. Probably I will fix it.
401 error
401 stands for “Unauthorized”. You're not allowed to visit the page, and probably not your visitors either.
403 error
Means “Forbidden”. The remote file is protected by a password. HTML LinkChecker cannot be configured to send usernames and passwords to different pages, so if you know that your visitors have access passwords, this is nothing to worry about.
404 error
Without any competitior, this is the most common error code. You've probably got it hundreds of times when browsing the net, and this is one of the basic reasons to use HTML LinkChecker. The linked file does not exist, and nor the program or your visitors can't find out where it is. Either it's been moved or deleted, or you have typed the wrong URL to the file. Frustating, yes, but your visitors would be even more frustated if they got the same error... If you're using a proxy-server, 404 will be returned even if other numbers was returned to the proxy. I work on a way to fix this, if possible, in future versions.
500 error
Error codes starting with the number of five means that there was something wrong in the server and not, as with 4xx errors, in the browser (which is HTML LinkChecker in this case). 500 means “Internal server error” and is usually a temporarily error. A CGI-script which is configured in a wrong way will also return this error. There is nothing you can do about it.
501 error
This says “Not implemented”, which still doesn't make the life easier for you. If you're sure about the URL, tell the webmaster of the remote page about it. (“But I can't connect to the server, how would I then be able to know the email to the webmaster?”, you may ask. If nothing else, try for example webmaster@server.com, if the remote page should be located at http://www.server.com/somefolder/file.html. Email to webmaster use to work).
503 error
“The service is unavailable”. This is usually a temporarily error. The server exists, but can be down for backup, upgrade or there is something else which is wrong. Hopefully it will soon be fixed, and if you try again, maybe it's ok then.
Con error
HTML LinkChecker opened a connection to the server, sent a request for the specified document, but the server closed the connection before HTML LinkChecker had time to check for a reply. This can be natural, or it can be caused by a very slow operation in HTML LinkChecker. Let's say, for example, you leave the program running in the background and use another application which doesn't give any processor-time to HTML Linkchecker. The program can't check for results on sent requests, and after an amount of minutes (depending on the remote server), the server doesn't want to wait anymore and closes the connection. On the other hand, if you get this error frequently even if the program seems to work fast, and if the link works fine in a browser, it's probably a bug. Please tell me about it!
DNS error
This means that the remote host does not exist, or that it is currently unavailable, or that you've set the options in the “Preferences”-dialog a bit wrong. It is returned when HTML LinkChecker fails to make an IP-number (like 192.43.16.43) of the host (like www.server.com). It can be temporaily, but more often, the server doesn't exist at all. A link to “http://www.doesnt.exist.at.all” will for example return a DNS error. Another reason is that the “DNS Timeout” is set to low in the “Preferences”-dialog, see the “Preferences”-chapter. You may increase it, but if it's slow for you, it will also be slow to contact for your visitors. DNS-errors are never returned when using a proxy-server, instead you will get 404 errors.
Ext error
Means that the connection was refused by the server. HTML LinkChecker translated the hostname to an IP-number succesfully, asked to open a connection but found it closed immediately. In other words, the server exist, but by some reason, it doesn't want anyone to open connections to it. It's probably closed temporarily and will be back again soon.
Idl error
This is returned when HTML LinkChecker have got no response to its request after the “Idle Timeout”-delay specified in the “Preferences”-dialog. The server may be slow, in most cases it will work fine after a few retries. If you get many Idl errors, you should increase the “Idle Timeout”. Default is 300 seconds (5 minutes). Since HTML LinkChecker can work while waiting, there is really no big deal if it is set to one minute or ten.
Unknown error (xxx)
You get a number, and when you open the “Full information”-dialog, it says “Unknown error”, followed by the code within parentheses. It doesn't have to mean that there is something strange, you don't need to call the news. It's just a sort of error that HTML LinkChecker is not familiar with. If you get an error like this, please let me know and I'll teach the program what it means until the next release.
Just fine
If you choose “Try again” and the result is OK, and then click “Update list”, the list will store it labled as “Just fine”. Means “Just fine”...
200 and 302 results
These are only returned when choosing “Try again” from the “Full information”-dialog. They means that the linked file exist, and that you should be just happy. The error originally reported was just temporarily, and it isn't much to care about. It's a good idea to hide the line after you've closed the “Full information”-dialog. To be exact, 200 means “ok” and 302 means “moved temporarily”. If a file is moved temporarily, it will, differently from 301 errors, move back in the near future. Nothing to care about, in other words.
Don't just read the number
An external error does not have to be what it says. An overloaded server may report everything from “503 Service is unavailable” to “404 File not found”. It's also possible that there is a temporarily error in the server. A once-a-week backup may report “404” for a few seconds, even if it would report “200 ok” in the other 604795 seconds of the week. Usually you see what's wrong, mabe you've just spelt the URL wrong, or you know that the specified URL is very popular and sometimes is overloaded.
