[users@httpd] distinguishing different flavours of 502 errors (bad gateway error)

[Ian Stuart <Ia...@ed.ac.uk>]

I'm trying to distinguish between the two error-conditions
"(146)Connection refused: proxy: HTTP: attempt to connect to .....
failed" and "(70007)The timeout specified has expired: proxy: HTTP:
attempt to connect to .... failed"

The two codes appear to be consistent for n repeats of the the same call
- even after server restarts.

To explain:
I have a couple of dozen servers, spread over two or three hosts, all
running on high port numbers, and all accessed through vhosts defined in
a central proxying server.
I want to be able to distinguish between various failure states:
- I can deal the valid upstream server but a bad filename (you get the
404 error from the running upstream host)
- I can deal with abad hostname [eg, it's not in DNS].
("REDIRECT_ERROR_NOTES" contains some text I can scan)

I just need to distinguish between two similar, but fundamentally
different error conditions:
- a host which is down
- a host which is up, but with a non-running web-server.

One problem is that the link between http://sub.domain.com/ < == >
http://domain2.com:12345/ is defined in hte http.conf file - but this is
not reflected in the environment seen by the apache CGI environment...

Any hints/tips?
(If it's significant, I'm using apache 2.0.xx in a unix environment, and
Perl to create a dynamic fragment of html to include in a customised 502
error page)
-- 

Ian Stuart.
Bibliographics and Multimedia Service Delivery team,
EDINA,
The University of Edinburgh.

http://edina.ac.uk/


---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
   "   from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org