* dirmngr/ks-engine-hkp.c (struct hostinfo_s): New field 'port'.
(create_new_hostinfo): Initialize 'port'.
(add_host): Add host parameter and update the hosttable entry.
(map_host): Return port if known, adjust calls to 'add_host'.
(make_host_part): Let 'map_host' specify the port if known.
Signed-off-by: Justus Winter <justus@g10code.com>
* dirmngr/ks-engine-hkp.c (map_host): Update the original 'hosttable'
entry instead of creating another one.
--
Previously, if a keyserver was given using its IP address
(e.g. 'hkp://127.0.0.1'), 'map_host' would create two entries in the
'hosttable', one with the name '127.0.0.1', and the second one with
the name 'localhost' and 'v4addr' being set to '127.0.0.1'. But the
first entry is the one being used for lookups, and since the 'v4' flag
is not set, the 'HTTP_FLAG_IGNORE_IPv4' is used when doing the http
request, making it fail.
Signed-off-by: Justus Winter <justus@g10code.com>
GnuPG-bug-id: 2012
* dirmngr/dns-stuff.c (is_onion_address): New.
* dirmngr/ks-engine-hkp.c (hostinfo_s): Add field "onion".
(map_host): Special case onion addresses.
(ks_hkp_print_hosttable): Print an 'O' for an onion address.
* dirmngr/http.c (connect_server): Special case onion addresses.
--
Note that this requires the latest libassuan from git. Onion addresses
are always support regardless of the --use-tor flag.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (my_getnameinfo): Change arg type to
dns_addrinfo_t.
(map_host): Replace getaddrinfo by resolve_dns_name.
--
Note that we still need to replace getnameinfo so that the PTR lookup
is either suppressed or also done via ADNS.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (map_host): Do not use original pointer
after realloc.
--
vex01 reported and debugged the problem.
GnuPG-bug-id: 2107
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dirmngr.h (opt): Add field "use_tor".
* dirmngr/dirmngr.c (oUseTor): New.
(opts): Add --use-tor.
(parse_rereadable_options): Set option.
(main): Tell gpgconf about that option.
* dirmngr/crlfetch.c (crl_fetch): Pass TOR flag to the http module and
return an error if LDAP is used in TOR mode.
(ca_cert_fetch): Return an error in TOR mode.
(start_cert_fetch): Ditto.
* dirmngr/ks-engine-finger.c (ks_finger_fetch): Pass TOR flag to the
http module.
* dirmngr/ks-engine-hkp.c (send_request): Ditto.
* dirmngr/ks-engine-http.c (ks_http_fetch): Ditto.
* dirmngr/ks-engine-ldap.c (ks_ldap_get): Return an error in TOR mode.
(ks_ldap_search): Ditto.
(ks_ldap_put): Ditto.
* dirmngr/ocsp.c (do_ocsp_request): Ditto. Also pass TOR flag to the
http module.
* dirmngr/server.c (option_handler): Add "honor-keyserver-url-used".
(cmd_dns_cert): Return an error in TOR mode.
(cmd_getinfo): Add subcommand "tor"
* tools/gpgconf-comp.c (gc_options_dirmngr): Add TOR group.
--
More work is required to actually make --use-tor useful. For now it
returns an error for almost all network access but as soon as we have
added the TOR feature to the http module some parts will start to
work.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (armor_data): Move function from here...
* dirmngr/misc.c (armor_data): ... to here and drop static qualifier.
* dirmngr/misc.h: New declaration.
--
Signed-off-by: Neal H. Walfield <neal@g10code.de>
* common/http.c (send_request): Set the requested for SNI.
* dirmngr/ks-engine-hkp.c (map_host): Return the poolname and not
the selecting a host.
--
GnuPG-bug-id: 1792
Thanks to davidw for figuring out the problem.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (map_host): Fix error return.
--
In ks-engine-hkp.c on line 509 'reftbl' is freed, but it is then
used on line 511. I'm guessing this is a missing return;.
Reported-by: Joshua Rogers <honey@internot.info>
Debian-Bug-Id: 773520
Other fixes on error added too.
* dirmngr/ks-engine-hkp.c (map_host): Change to return an gpg_error_t.
Return an error code for all dead hosts.
(make_host_part): Change to return an gpg_error_t. Change all
callers.
--
The functions used to return an error code via ERRNO. However, this
does not allow to return extra error codes in a portable way. Thus we
change the function to directly return a gpg_error_t.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (ks_hkp_help): Make use of TLS macros.
--
Only report support for the hkps scheme when GnuPG / dirmngr
has been built with a TLS library.
This helps debuging and enable the user to detect whether support
for hkps is included by doing a
`gpg-connect-agent --dirmngr 'keyserver --help' /bye`.
Currently hkps will be listed as a supported scheme but trying to
add a keyserver using it will silently fail.
As a digression, https is never listed as a valid scheme.
* dirmngr/ks-engine-hkp.c (cert_log_cb): Move to ...
* dirmngr/misc.c (cert_log_cb): here.
* dirmngr/ks-engine-http.c (ks_http_fetch): Support 307-redirection
and https.
--
Note that this requires that the root certificates are registered using
the --hkp-cacert option. Eventually we may introduce a separate
option to allow using different CAs for KS_FETCH and keyserver based
requests.
* dirmngr/ks-engine-hkp.c (hostinfo_s): Add fields cname, v4addr, and
v6addr.
(create_new_hostinfo): Clear them.
(my_getnameinfo): Add args numeric and r_isnumeric.
(is_ip_address): New.
(map_host): Add arg r_host. Rewrite the code to handle pools in a
special way.
(ks_hkp_print_hosttable): Change format of help info output.
(make_host_part): Add arg optional r_httphost.
(send_request): Add arg httphost.
(ks_hkp_search, ks_hkp_get, ks_hkp_put): Get httphost and pass it to
send_request.
--
This changes quite some things on how the hostinfo is maintained.
However, it might be better to rework the data structures and have one
entry per IP address instead of this clumsy patch.
* common/http.c (http_open): Add arg httphost.
(http_open_document): Pass NULL for httphost.
(send_request): Add arg httphost. If given, use HTTPHOST instead of
SERVER. Use https with a proxy if requested.
(http_verify_server_credentials): Do not stop at the first error
message.
* dirmngr/ocsp.c (do_ocsp_request): Adjust call to http_open.
* keyserver/curl-shim.c (curl_easy_perform): Ditto.
* dirmngr/ks-engine-http.c (ks_http_fetch): Ditto.
* dirmngr/ks-engine-hkp.c (ks_hkp_help): Ditto.
* dirmngr/dirmngr.c (HOUSEKEEPING_INTERVAL): New.
(housekeeping_thread): New.
(handle_tick): Call new function.
* dirmngr/ks-engine-hkp.c (RESURRECT_INTERVAL): New.
(struct hostinfo_s): Add field died_at and set it along with the dead
flag.
(ks_hkp_print_hosttable): Print that info.
(ks_hkp_housekeeping): New.
--
The resurrection gives the host a chance to get back to life the next
time a new host is selected.
* dirmngr/ks-engine-hkp.c (ks_hkp_get): Allow exact search mode.
* g10/keyserver.c (keyserver_import_name): Implement.
(keyserver_get): Use exact mode for name based import.
(keyserver_get): Add args R_FPR and R_FPRLEN. Change all callers.
* dirmngr/ks-engine-hkp.c (ks_hkp_search, ks_hkp_get): Print SOURCE
status lines.
* g10/call-dirmngr.c (ks_status_parm_s): New.
(ks_search_parm_s): Add field stparm.
(ks_status_cb): New.
(ks_search_data_cb): Send source to the data callback.
(gpg_dirmngr_ks_search): Change callback prototope to include the
SPECIAL arg. Adjust all users. Use ks_status_cb.
(gpg_dirmngr_ks_get): Add arg r_source and use ks_status_cb.
* g10/keyserver.c (search_line_handler): Adjust callback and print
"data source" disgnostic.
(keyserver_get): Print data source diagnostic.
--
It has often been requested that the actually used IP of a keyservers
is shown in with gpg --recv-key and --search-key. This is helpful if
the keyserver is actually a pool of keyservers. This patch does this.
* dirmngr/ks-action.c (ks_action_resolve): Rename var for clarity.
(ks_action_search, ks_action_put): Ditto.
(ks_action_get): Consult only the first server which retruned some
data.
* dirmngr/ks-engine-hkp.c (SEND_REQUEST_RETRIES): New.
(map_host): Add arg CTRL and call dirmngr_tick.
(make_host_part): Add arg CTRL.
(mark_host_dead): Allow the use of an URL.
(handle_send_request_error): New.
(ks_hkp_search, ks_hkp_get, ks_hkp_put): Mark host dead and retry on
error.
* dirmngr/server.c (cmd_killdirmngr): Factor some code out to ...
(check_owner_permission): here.
(cmd_keyserver): Add options --dead and --alive.
* dirmngr/ks-engine-hkp.c (host_in_pool_p): New.
(ks_hkp_mark_host): New.
--
Also removed the warning that the widnows part has not yet been done.
AFAICS, the current mingw supports the all used socket functions.
The asymmetric quotes used by GNU in the past (`...') don't render
nicely on modern systems. We now use two \x27 characters ('...').
The proper solution would be to use the correct Unicode symmetric
quotes here. However this has the disadvantage that the system
requires Unicode support. We don't want that today. If Unicode is
available a generated po file can be used to output proper quotes. A
simple sed script like the one used for en@quote is sufficient to
change them.
The changes have been done by applying
sed -i "s/\`\([^'\`]*\)'/'\1'/g"
to most files and fixing obvious problems by hand. The msgid strings in
the po files were fixed with a similar command.
We had some debug code here which prevented it from working.
The host selection code still needs a review!
* ks-engine-http.c (ks_http_help): Do not print help for hkp.
* ks-engine-hkp.c (ks_hkp_help): Print help only for hkp.
(send_request): Remove test code.
(map_host): Use xtrymalloc.
* certcache.c (classify_pattern): Remove unused variable and make
explicit substring search work.
Since 2009-12-08 gpg was not able to find email addresses indicated
by a leading '<'. This happened when I merged the user id
classification code of gpgsm and gpg.
To better cope with round robin pooled A records like keys.gnupg.net
we need to keep some information on unresponsive hosts etc. What we
do now is to resolve the hostnames, remember them and select a random
one. If a host is dead it will be marked and a different one
selected. This is intended to solve the problem of long timeouts due
to unresponsive hosts.
The code is not yet finished but selection works.
We better do this once and for all instead of cluttering all future
commits with diffs of trailing white spaces. In the majority of cases
blank or single lines are affected and thus this change won't disturb
a git blame too much. For future commits the pre-commit scripts
checks that this won't happen again.
