rdelaage
/
gmid
mirror of https://github.com/omar-polo/gmid.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
938 Commits 4 Branches 28 Tags 5.4 MiB
Commit Graph

164 Commits

Author SHA1 Message Date
Omar Polo 8e09ee1d50 typo 2021-03-20 10:48:11 +00:00
Omar Polo b9c9123b8e fix signal handling so it works on linux too 
it seems that linux calls the signal handlers even when we're waiting
on sigwait for that signal.  Work around that.
 2021-03-20 10:43:23 +00:00
Omar Polo d632468df5 fix correct shutdown after SIG{INT,TERM} 2021-03-20 09:23:53 +00:00
Omar Polo 1d3eb470b0 quit every process cleanly when receiving SIGINT or SIGTERM 2021-03-20 08:51:17 +00:00
Omar Polo bc99d868bc refactoring: imsg everywhere 
use imsg to handle ALL kinds of IPC in gmid.  This simplifies and shorten the
code, and  makes everything more uniform too.
 2021-03-19 19:21:29 +00:00
Omar Polo dbe262a45d split usage string into two lines 2021-03-03 17:51:42 +00:00
Omar Polo 2c3e53dac6 give each server process its own socket for the executor 
this fixes a bug introduced with the prefork mechanics: every server
process shared the same socket, and this would cause a race condition
when multiple server processes asked for a script cgi being executed.

This gives each server process its own socket to talk to the executor,
so the race cannot happen.
 2021-03-03 17:22:01 +00:00
Omar Polo 376a540764
move log_init & vars to gmid.c, retain logger_main in log.c 
this is to let the regression suite compile
 2021-02-23 13:44:20 +01:00
Omar Polo d278a0c3c5
moving logging to its own process 2021-02-23 13:40:59 +01:00
Omar Polo 52418c8d82 fix various compilation errors 
Include gmid.h as first header in every file, as it then includes
config.h (that defines _GNU_SOURCE for instance).

Fix also a warning about unsigned vs signed const char pointers in
openssl.
 2021-02-12 12:47:20 +00:00
Omar Polo aa37287565 add newline after usage 2021-02-10 17:58:43 +00:00
Omar Polo df58efff26 fix seccomp for the new event loop 
add/remove syscalls from the BPF filter and move sandbox() after
libevent initialisation
 2021-02-08 12:46:46 +00:00
Omar Polo 3abf91b0b4 improve logs management 2021-02-07 15:30:28 +00:00
Omar Polo a709ddf5eb added prefork option 2021-02-07 12:05:32 +00:00
Omar Polo e3ddf39095 add the ``entrypoint'' option 2021-02-06 18:28:43 +00:00
Omar Polo 6abda252e9 added ``block return'' and ``strip'' options 2021-02-06 17:22:37 +00:00
Omar Polo 102d2e9cce missing argument for LOGI 2021-02-04 18:09:42 +00:00
Omar Polo 9543e3fbaf avoid race-condition 
what if we receive a SIGHUP right after unblock_signal (or during the
whole block_signals...unblock_signals) but *before* the wait_sighup?
Yeah.
 2021-02-04 18:09:30 +00:00
Omar Polo ca21e10043 reload configuration on SIGHUP 2021-02-04 13:23:15 +00:00
Omar Polo e824d03efa drop unnecessary check around close 2021-02-03 16:53:34 +00:00
Omar Polo 4e2e2ab1d3 refactor executor_main 
now it's symmetrical to listener_main().
 2021-02-03 16:37:53 +00:00
Omar Polo d672b8fba1 refactoring startup logic 2021-02-03 16:28:00 +00:00
Omar Polo 9edb828251 drop privileges after the fork 2021-02-03 14:13:32 +00:00
Omar Polo 3c680bddab configless: fixing the case of the implicit "." 
I got bitten by the scope visibility rules.  After the end of the
block, the path variable is no longer valid, and in fact later
load_vhosts fails to open that (because the buffer gets invalidated)
 2021-02-02 09:30:24 +00:00
Omar Polo bcf5d929e6 ensure absolute paths in config-less mode 2021-02-01 11:07:57 +00:00
Omar Polo 8503a1431d set cloexec status on the socketpair fds 
the executor forks to spawn the cgi scripts, and they inherit the
socket for communication with the listener process.  Make that
impossible.
 2021-01-31 11:07:12 +00:00
Omar Polo ef0f2ae88c remove unused function 2021-01-28 14:21:57 +00:00
Omar Polo 8904fa0e64 -v for verbose logging 2021-01-27 23:14:16 +00:00
Omar Polo 13ed2fb63f improve configuration parsing error 2021-01-27 16:45:45 +00:00
Omar Polo b736ca35f2 drop unused variable 2021-01-27 15:35:55 +00:00
Omar Polo 44ee1bac8b use starts_with in puny.c 2021-01-27 15:35:09 +00:00
Omar Polo 72bbed9136 make configless mode works again. also accept every host 2021-01-27 15:00:55 +00:00
Omar Polo 46af8c6cfe drop the daemon config in favour of the -f flag 
Now it daemonize by default when running with a config, unless the -f
flag is given.
 2021-01-27 12:04:37 +00:00
Omar Polo 3300cbe06a initial punycode support 2021-01-27 10:47:49 +00:00
Omar Polo 390a611893 sane defaults for the config-less mode 2021-01-27 10:43:55 +00:00
Omar Polo 8443bff77a rework the configless mode: change flags and generate certs 2021-01-25 14:08:31 +00:00
Omar Polo ae08ec7da5 chroot & drop privileges 2021-01-25 10:30:07 +00:00
Omar Polo 2030e31486 order 2021-01-25 09:45:09 +00:00
Omar Polo 252908e6bb added support for location blocks 2021-01-24 18:53:26 +00:00
Omar Polo 501e489c90 initialize config 2021-01-24 12:45:22 +00:00
Omar Polo 8d6ae384be move configuration init stuff in its own function 2021-01-24 10:39:48 +00:00
Omar Polo f77a8c867e add ends_with 2021-01-21 22:45:49 +00:00
Omar Polo ea28a44931 const-ify parameter to strtonum 2021-01-21 16:15:10 +00:00
Omar Polo a5a09e44b2 use strtonum 2021-01-21 16:07:12 +00:00
Omar Polo b2a6b61371 rename mimes to mime ; pass config esplicitly to mime* functions 2021-01-21 15:01:37 +00:00
Omar Polo 0ab65593e2 don't crash on wrong vhost or missing SNI 
the new logging code was crashing if the client didn't support SNI or
if required an unknown vhost: this because we short-circuit in
handle_handshake to an error, so c->iri isn't populated yet (we don't
even read the request).

fixes #1
 2021-01-21 08:26:21 +00:00
Omar Polo 0be51733ef improve logs 
now we log the full IRI requested (before was only the path) and the
response line (even for CGI).
 2021-01-20 18:54:44 +00:00
Omar Polo 0fbe79b33c improve mime handling 
we still have an hardcoded list, but this implements the API needed to
modify the mappings.
 2021-01-18 18:18:57 +00:00
Omar Polo 132cae8c6f removing err/warn functions with our fatal 
err/warn is not available on some systems (unfortunately!) and in any
case don't play well with our daemon mode (that closes std{in,out,err}).
Use our fatal that is daemon-aware.
 2021-01-18 18:15:58 +00:00
Omar Polo d3a08f4d17 reorganize: move bunch of functions to server.c 
cgi.c wasn't really needed; it better to group all the server related
functions together, cgi or not.  Now gmid.c contains only startup and
utility code.
 2021-01-17 23:23:58 +00:00
Omar Polo 0170ba023f re-enable daemon mode 
with the split into two process, the call to daemon was lost
 2021-01-17 11:56:50 +00:00
Omar Polo 7146dd552f missing break 2021-01-17 09:34:06 +00:00
Omar Polo d939d0f09e switch to getcwd 2021-01-17 09:33:45 +00:00
Omar Polo 881a9dd9c2 split into two processes: listener and executor 
this way, we can sandbox the listener with seccomp (todo) or capsicum
(already done) and still have CGI scripts.  When we want to exec, we
tell the executor what to do, the executor executes the scripts and
send the fd backt to the listener.
 2021-01-16 19:41:34 +00:00
Omar Polo 7edc455ab4 move cgi stuff to its own file 2021-01-15 19:08:28 +00:00
Omar Polo 5bc3c98ed4 add protocols to the config 2021-01-15 18:55:05 +00:00
Omar Polo b777bf4b2b check also that the port number matches 2021-01-15 18:24:24 +00:00
Omar Polo f7b816dc39 style 2021-01-15 15:21:51 +00:00
Omar Polo dafb57b8af sandbox also on FreeBSD with capsicum 2021-01-15 14:03:45 +00:00
Omar Polo 1590277007 conf & vhosts 
* gmid.c (main): changed behaviour: daemon off by default
(main): changed -c in -C (cert option)
(main): changed -k in -K (key option, for consistency with -C)
(main): added -c to load a configuration
(main): certs, key and doc (-C -K and -d) doesn't have a default value anymore
(handle_handshake): add vhosts support
 2021-01-15 09:17:43 +00:00
Omar Polo 9862b637c2 track handshakes 
This alter the current state machine by adding S_HANDSHAKE as the
initial state.  There, we ensure we did the handshake and we check
SNI.  ATM we simply continue in S_OPEN, but later we can add virtual
host checks there, and skip to S_INITIALIZING with an error state if
the client is accessing a wrong host.
 2021-01-13 18:40:18 +00:00
Omar Polo 0d7a38c4ce avoid __dead and return at the end of main 2021-01-11 15:28:25 +00:00
Omar Polo 7b1d979032 reject non-gemini protocols with 53 2021-01-11 14:45:57 +00:00
Omar Polo 3c1cf9d07c s/uri/iri since we accept IRIs 2021-01-11 13:08:00 +00:00
Omar Polo 28778244d6 fix build 2021-01-11 12:53:46 +00:00
Omar Polo 85dff1f9c3 fix remote_user for CGI and add -6 flag to enable ipv6 2021-01-11 12:08:50 +00:00
Omar Polo 33756bd235 accept both ipv4 and ipv6 2021-01-10 22:56:33 +00:00
Omar Polo 709d6e5ead log also the port of the client 2021-01-10 22:29:22 +00:00
Omar Polo 80bbcad5f2 make FATAL and LOG inline functions 
other functions that aren't macros anymore.  LOG went under a rename
to logs because log is a builtin (the math function), or so gcc says.
 2021-01-10 15:50:27 +00:00
Omar Polo 9b374f41ee macro reordering 
and while there replace SAFE_SETENV with an inline function.  LOG is
more difficult to transform into an inline function, given the string
concatenations it does.  The other LOG* and FATAL macros are fine as
they already are.
 2021-01-10 14:40:14 +00:00
Omar Polo 16ace04d45 simplify loop 
todo was initially there for an optimization: don't loop to MAX_USERS
when you know the upper limit is todo.
 2021-01-08 16:59:50 +00:00
Omar Polo 4c4167393a
simplify unveil/pledge calls 2021-01-01 13:03:04 +01:00
Omar Polo 6c6c7a0ed7
rename cgi_setpoll_on_* to cgi_poll_on_* 2020-12-28 18:01:06 +01:00
Omar Polo b3deee7b38
initialize error string to avoid returning garbage 2020-12-28 17:50:26 +01:00
Omar Polo 4a28dd014a
ignore also SIGHUP 
SIGHUP is sent when the tty is detached and by default kills the
process.  When we run in the background we don't care anymore about
the tty, so it should be safe for us to ignore SIGHUP.  (frankly,
I expected daemon(3) to do stuff like this for us).
 2020-12-28 09:55:56 +01:00
Omar Polo 5a40216f37
fix CGI with new IRI parser 
With new IRI, parser the old assumption of path starting with ./
is no longer valid.
 2020-12-28 09:52:39 +01:00
Omar Polo 33d32d1fd6
implement a valid RFC3986 (URI) parser 
Up until now I used a "poor man" approach: the uri parser is barely a
parser, it tries to extract the path from the request, with some minor
checking, and that's all.  This obviously is not RFC3986-compliant.

The new RFC3986 (URI) parser should be fully compliant.  It may accept
some invalid URI, but shouldn't reject or mis-parse valid URI.  (in
particular, the rule for the path is way more relaxed in this parser
than it is in the RFC text).

A difference with RFC3986 is that we don't even try to parse the
(optional) userinfo part of a URI: following the Gemini spec we treat
it as an error.

A further caveats is that %2F in the path part of the URI is
indistinguishable from a literal '/': this is NOT conforming, but due
to the scope and use of gmid, I don't see how treat a %2F sequence in
the path (reject the URI?).
 2020-12-25 13:13:12 +01:00
Omar Polo 488f059ac4
moving declarations to header file 2020-12-24 16:48:39 +01:00
Omar Polo 51d876f030
always check absolutify_path return 2020-12-21 15:42:06 +01:00
Omar Polo 231bfcdc03
make -d handle correctly non-absolute paths 
before the -d option only accepted absolute paths, and this wasn't
documented.  Even more, with the default value of "docs" it won't
work.  Now it transforms all relative paths to absolute paths before
going on.
 2020-12-21 14:38:31 +01:00
Omar Polo dd3e110a70
reorder & fmt 2020-12-20 20:03:55 +01:00
Omar Polo d7802bb44a
logging reworked and daemonize by default 
The -l option was removed: now it logs on syslog if -f (foreground) is
not passed.
 2020-12-02 21:18:01 +01:00
Omar Polo 677afbd3f8
clients certs support for CGI 
internally, gmid doesn’t care if the client issued a certificate, but
now we pass that information to the CGI script in some new environment
variables.
 2020-12-02 15:17:19 +01:00
Omar Polo c603f1260d
fix bug introduced in last version regarding full URLs 
gemini://example.com was incorrectly parsed.
 2020-11-18 09:45:28 +01:00
Omar Polo 7cd6880b84
correctly handle also non absolute URLs 2020-11-18 09:34:42 +01:00
Omar Polo ae2ad03ec0
handle //example.com urls 2020-11-18 09:32:07 +01:00
Omar Polo 721e232529
make port number configurable 2020-11-18 09:12:27 +01:00
Omar Polo a5d310bc0d
[cgi] added support for path parameters 
enhance the CGI scripting support so that script can take path
parameters.  That is, a script at /cgi/foo is called when the request
path is /cgi/foo/bar/...

This commit also introduce some backward incompatible changes as the
default env variables set for the CGI script changed.
 2020-11-10 14:07:36 +01:00
Omar Polo 72f653b652
[cgi] execute cgi scripts only inside a specific directory 
change the meaning of the -x flag: now it takes a string and executes
CGI scripts only if they are inside a directory with the given name,
relatively to the document root.
 2020-11-06 18:05:44 +01:00
Omar Polo f65ed01841
[cgi] set QUERY_STRING only if there is a query string 2020-11-06 18:02:41 +01:00
Omar Polo 120381c93d
[debug] print number of connected clients on SIGINFO 
or SIGUSR2 for poor peoples
 2020-11-06 17:09:30 +01:00
Omar Polo aff8d19010
handle CGI concurrently 
don’t stop-the-world-until-cgi-end, but rather poll on the script, so
we can handle other requests in the meantime.
 2020-11-06 17:09:14 +01:00
Omar Polo 60ba426e7e
[cgi] setenv instead of building an env + define more env variables 2020-11-06 15:07:33 +01:00
Omar Polo 75d233f00c
pass the query parameter to start_cgi 2020-11-06 15:06:50 +01:00
Omar Polo aa0fe0cf24
extract query parameters from the URL 2020-11-06 15:04:16 +01:00
Omar Polo 3c19febb01
ensure the requested protocol is “gemini” 
…and not something else that happens to be 6-bytes long.
 2020-11-06 14:24:25 +01:00
Omar Polo e8cac16e03
mark every open file as close-on-exec 2020-11-06 14:18:17 +01:00
Omar Polo 72342dc960
implementing CGI – NOT READY YET! 
This is a first try at implementing CGI scripting.  The idea is that,
if CGI is explicitly enabled by the user, when a user requires an
executable file instead of serving it to the client, that file will be
executed and its output fed to the client.

There are various pieces that are still lacking, the firsts that comes
to mind are:

 - performance: the handle_cgi just loops ignoring the
   WANT_POLLIN/POLLOUT and blocking if the child process hasn’t
   outputted anything.
 - we don’t parse query variable (yet)
 - we need to set more variables in the child environment
   side question: it’s better to set the variables using setenv() or
   by providing an explicit environment?
 - document what environment the CGI script will get
 - improve the horrible unveil/pledge(cgi ? …)

but now I can serve “hello-world”-tier script from gmid!
 2020-11-06 13:01:31 +01:00
Omar Polo 2c3a40faf8
added option to log to a file 2020-11-06 10:58:30 +01:00
Omar Polo 932b001a3f
typo 2020-11-06 10:45:06 +01:00