With a fresh build of 684eb478e my SCGI setup with nginx 1.5.12 on Ubuntu Server 13.10 64bit appears to be working.
Fossil is started via upstart using: exec /usr/local/sbin/fossil server --localhost --scgi /srv/fossils/partner Nice catch Joe! http://www.mail-archive.com/fossil-users@lists.fossil-scm.org/msg15001.html On 3/5/14, Jon <jon.for...@gmail.com> wrote: >>> I'll tinker with scgi tomorrow, but even if I get it working, I don't >>> yet >>> see how scgi ensures fossil generates correct links to it's gui and >>> route back user url's similar to >>> >>> http://somewhere.mydomain.com/partner/<repo-name>/... >>> >>> ...SNIP... >>> >> I've only used nginx once, and that was to test the SCGI code of Fossil >> when I was writing it. I didn't have to do anything special, other than >> setting the SCRIPT_NAME scgi_param. The links that the GUI generates are >> either relative, or else they are use the HTTP_HOST scgi_param, depending >> on context. >> >> http://www.fossil-scm.org/fossil/doc/tip/www/server.wiki#scgi shows my >> complete setup. It just works. Don't try to over-think it. > > Here's a summary of an scgi config that rewards me with a 502 from nginx. > > How do you run `fossil server` in verbose/debug mode and/or have it > log out detailed info? Any obvious config mistakes or other debugging > ideas? > > > $ /usr/local/sbin/fossil version > This is fossil version 1.28 [3d49f04587] 2014-01-27 17:33:44 UTC > > $ nginx -v > nginx version: nginx/1.5.11 > > # start fossil server like so (or via `exec` in an upstart script) > $ sudo /usr/local/sbin/fossil server --scgi /srv/fossils/partner > > # relevant /etc/nginx/conf.d/default.conf entry > location ~ ^/partner/ { > include scgi_params; > scgi_pass 127.0.0.1:8080; > scgi_param SCRIPT_NAME "/partner"; > } > > # fossil repos live in this structure > # have used both `root` and `nginx` ownership; no difference wrt error > $ ll /srv/fossils > total 4 > drwxr-xr-x 2 root root 4096 2014-03-02 10:05:12 partner/ > > $ ll /srv/fossils/partner > total 88 > -rw-r--r-- 1 root root 89088 2014-03-02 10:05:12 logicalmaps.fossil > > # client requests get HTTP 502 errors > curl -I http://stimpy/partner/logicalmaps/ > HTTP/1.1 502 Bad Gateway > Server: nginx > Date: Thu, 06 Mar 2014 01:25:08 GMT > Content-Type: text/html > Content-Length: 537 > Connection: keep-alive > ETag: "5315ce33-219" > > # nginx error message corresponding to HTTP 502 error to client request > $ tail -1 /var/log/nginx/errors.log > 2014/03/05 20:57:43 [error] 2054#0: *1 upstream prematurely closed > connection while reading response header from upstream, client: > 10.0.2.2, server: stimpy, request: "GET /partner/logicalmaps/ > HTTP/1.1", upstream: "scgi://127.0.0.1:8080", host: "stimpy" > _______________________________________________ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
