* dirmngr/dirmngr.c (parse_ldapserver_file): Add arg ignore_enoent.
(main): Use that arg for the default file.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dns.h (dns_strsection, dns_strclass)
(dns_strtype): Change APIs.
* dirmngr/dns.c (dns_p_lines): Use __dst for dns_strsection.
(dns_rr_print): Use __dst for dns_strclass and dns_strtype.
(dns_trace_dump): Likewise.
(dns_ai_print): Use __dst for dns_strtype.
(dns_strsection): Add an argument __dst for storage.
(dns_strclass, dns_strtype): Likewise.
(parse_packet): Use __dst for dns_strsection.
(send_query): Use __dst for dns_strtype.
(isection): Use __dst for dns_strsection.
(iclass): Use __dst for dns_strclass.
(itype): Use __dst for dns_strtype.
--
Backport master commit of:
30eef28bc0f5deaa1b4b7f04293a6527524280a9
GnuPG-bug-id: 4934
Reported-by: Tomas Mraz
Fixes-commit: ff7d01fc6d396fc3b8d37baa9bd4cdebc8853648
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* common/logging.c (log_printhex): Chnage order of args. Make it
printf alike. Change all callers.
* configure.ac: Add -Wno-format-zero-length
--
This makes it consistent with modern libgpgrt logging and thus eases
back porting from newer GnuPG versions which use libgpgrt logging.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/validate.c (hash_algo_from_buffer): New.
(uint_from_buffer): New.
(check_cert_sig): Support rsaPSS.
* sm/certcheck.c (gpgsm_check_cert_sig): Fix small memory leak on
error.
--
Yes, I know that there is a lot of code duplication. In fact some of
the code is ugly and it would be better if we enhance Libgcrypt to
guarantee that returned memory buffers via gcry_sexp_extract_param are
allways Nul terminated and we should also enhance that function to
directly extract into an unsigned int or char *.
GnuPG-bug-id: 4538
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/certcache.c (find_cert_bysubject): Add better debug output
and try to locate by keyid.
--
This change was suggested in T4536
but we do not have any test cases for this.
GnuPG-bug-id: 4536
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 4699e294cc9e59f35262adca26ca291927acca9e)
The bug report meanwhile has a test description but I have not done
the testing yet. I port this back to 2.2 anyway given that no
regression have been reported for master in nearly a year.
* common/util.h (EXTERN_UNLESS_MAIN_MODULE): Add the definion only
here but now without the Norcroft-C. Change all other places where it
gets defined.
* common/iobuf.h (iobuf_debug_mode): Declare unconditionally as
extern.
* common/iobuf.c (iobuf_debug_mode): Define it here.
* agent/gpg-agent.c (INCLUDED_BY_MAIN_MODULE): Define here and also in
all main modules of all other programs.
* g10/main.h: Put util.h before the local header files.
--
This change is required for use with gcc/ld's LTO feature which does
not allow common blocks. Further gcc 10 will make -fno-common the
default and thus this chnage is always needed. What a pitty.
Co-authored-by: Tomáš Mráz
GnuPG-bug-id: 4831
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/server.c (proc_wkd_get): Print new NOTE status
"wkd_cached_result".
* g10/call-dirmngr.c (ks_status_cb): Detect this and print a not ein
verbose mode.
--
This little patch is helpful to see why a WKD change still does not
work after it has been updated on the server.
Signed-off-by: Werner Koch <wk@gnupg.org>
* doc/DETAILS: Specify new status code "NOTE".
* dirmngr/ks-engine-http.c (ks_http_fetch): Print a NOTE status for a
bad TLS certificate.
* g10/call-dirmngr.c (ks_status_cb): Detect this status.
--
For example a
gpg -v --locate-external-keys dd9jn@posteo.net
now yields
gpg: Note: server uses an invalid certificate
gpg: (further info: bad cert for 'posteo.net': \
Hostname does not match the certificate)
gpg: error retrieving 'dd9jn@posteo.net' via WKD: Wrong name
gpg: error reading key: Wrong name
(without -v the "further info" line is not shown). Note that even
after years Posteo is not able to provide a valid certificate for
their .net addresses. Anyway, this help to show the feature.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/http.c: Include dirmngr-status.h
(http_prepare_redirect): Emit WARNING status lines for redirection
problems.
* dirmngr/http.h: Include fwddecl.h.
(struct http_redir_info_s): Add field ctrl.
* dirmngr/ks-engine-hkp.c (send_request): Set it.
* dirmngr/ks-engine-http.c (ks_http_fetch): Set it.
* g10/call-dirmngr.c (ks_status_cb): Detect the two new warnings.
--
This should make it easier to diagnose problems with bad WKD servers.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dirmngr-status.h: New.
* dirmngr/dirmngr.h: Include dirmngr-status.h and move some prototypes
to that file.
* dirmngr/t-support.c: New.
* dirmngr/Makefile.am (t_common_src): Add new file.
--
This helps to backport changes from master.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dns-stuff.c (cached_inet_support): New variable.
(dns_stuff_housekeeping): New.
(check_inet_support): New.
* dirmngr/http.c (connect_server): Use only detected interfaces.
* dirmngr/dirmngr.c (housekeeping_thread): Flush the new cache.
--
This currently works only for Windows but that is where users really
ran into problems. The old workaround was to configure disable-ipv4
or disable-ipv6.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (send_request): Reinitialize HTTP session when
following a HTTP redirection.
--
inspired by patch from Damien Goutte-Gattat <dgouttegattat@incenp.org>
GnuPG-Bug_id: 4566
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
* dirmngr/ks-engine-hkp.c (SEND_REQUEST_EXTRA_RETRIES): New.
(handle_send_request_error): Use it for 503 and 504.
(ks_hkp_search, ks_hkp_get, ks_hkp_put): Pass a new var for
extra_tries.
--
This is a pretty stupid fix but one which works without much risk of
regressions. We could have used the existing TRIES but in that case
the fallback to other host would have been too limited. With the used
value we can have several fallbacks to other hosts. Note that the
TRIES is still cumulative and not per host.
GnuPG-bug-id: 4600
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 8b113bb148f273524682252233b3c65954e1419e)
* dirmngr/http.c (same_host_p): Consider certain subdomains to be the
same.
--
GnuPG-bug-id: 4603
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 37f0c55c7be3fc4912237f2bc72466aef6f8aa36)
* dirmngr/ocsp.c (do_ocsp_request): Remove arg md. Add args r_sigval,
r_produced_at, and r_md. Get the hash algo from the signature and
create the context here.
(check_signature): Allow any hash algo. Print a diagnostic if the
signature does not verify.
--
GnuPG-bug-id: 3966
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/domaininfo.c (struct domaininfo_s): Add field keepmark.
(insert_or_update): Implement new update algorithm.
--
The old algorithm limited the length of a bucket chain by purging the
last 50% or the entries. Thus the first domains entered into the
cache were never purged. The new algorithm is a bit better: It also
limits the chain length on overflow to 50% but tries to keep the
entries indicating that a WKD is available in the cache. If there is
still space to keep more, those which clearly do not support WKD are
also kept.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (send_request): New case for 413.
* dirmngr/ks-engine-http.c (ks_http_fetch): Ditto.
* dirmngr/ocsp.c (do_ocsp_request): Ditto.
--
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dns.h (dns_quietinit): Remove.
(dns_hints_i_new): Remove.
--
Even before our change, dns_quietinit was questionable macro; There
was no place in dns.c which requires overrides in initializer list.
Only redundant zero were.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
(cherry picked from commit 371ae25f8f6f2d1ac030bf984bca479393a5ed43)
* dirmngr/dns.h (DNS_OPTS_INIT, dns_opts): Remove.
* dirmngr/dns-stuff.c (libdns_res_open): Use zero-ed, and initialized
automatic variable for opts.
* dirmngr/dns.c (send_query, resolve_query, resolve_addrinfo):
Likewise.
--
In fact, DNS_OPTS_INIT was only needed when args are none. With
partially specified initialization, C99 guarantees zero-ed other
members just like static object.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
(cherry picked from commit d661acd483236d34720a4959fc816d05f89c2cb7)
* dirmngr/dns.h (dns_rr_foreach): Don't use dns_rr_i_new.
Call dns_rr_grep with NULL.
* dirmngr/dns.c (dns_rr_grep): Support NULL for error_.
--
Here we still use C99 feature of struct member initialization in
dns_rr_foreach, for struct dns_rr_i. Note that in C99, it guarantees
non-specified member fields are initialized by zero. So, there's no
need to use dns_rr_i_new at all.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
(cherry picked from commit a1ccfe2b37847cce0db2fb94a7365c9fa501eda4)
* dirmngr/dns.c (dns_inet_pton, dns_so_tcp_keep): Use automatic
variables.
(dns_poll, dns_send_nopipe): Likewise, adding const qualifier.
--
Compound literals is a feature of C99. Because we only use C90 plus
some limited features, in the project, it's better to avoid it.
Besides, we make sure when it's read-only.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
(cherry picked from commit 1c405499388fd5bed0968ab5c6c5d1b3373537b9)
* dirmngr/http.c (same_host_p): Add exception table.
--
Please: Adding entries to this table shall be an exception and not the
rule.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 134c3c16523b1a267ebdd2df6339240fd9e1e3b3)
* dirmngr/dns.c: Include gpgrt.h. Silence -Warray-bounds also gcc.
* tests/gpgscm/scheme.c: Include gpgrt.h.
(Eval_Cycle): Ignore -Wimplicit-fallthrough.
--
The funny use of case and labels in the CASE macro seems confuse the
fallthrough detection.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (handle_send_request_error): Add arg
http_status and handle it.
(ks_hkp_search): Get http_status froms end_request and pass on to
handle_send_request_error.
(ks_hkp_get): Ditto.
(ks_hkp_put): Ditto.
--
GnuPG-bug-id: 4175
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 05ef6282784495a77f4faf76c0de5bc85dfecf06)
* dirmngr/http.c (http_status2string): New.
--
Right now only the standard 5xx codes.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit dc61f4ecea5c9815cb00aeb25439978337c1fd64)
* dirmngr/http.h (parsed_uri_s): Add fields off_host and off_path.
(http_redir_info_t): New.
* dirmngr/http.c (do_parse_uri): Set new fields.
(same_host_p): New.
(http_prepare_redirect): New.
* dirmngr/t-http-basic.c: New test.
* dirmngr/ks-engine-hkp.c (send_request): Use http_prepare_redirect
instead of the open code.
* dirmngr/ks-engine-http.c (ks_http_fetch): Ditto.
--
With this change a http query will not follow a redirect unless the
Location header gives the same host. If the host is different only
the host and port is taken from the Location header and the original
path and query parts are kept.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit fa1b1eaa4241ff3f0634c8bdf8591cbc7c464144)
Summary:
* dirmngr/crlcache.c (crl_cache_flush): Also deinit the cache.
* dirmngr/server.c (hlp_flushcrls, cmd_flushcrls): New.
(register_commands): Add FLUSHCRLS.
--
This allows it to flush the CRL cache of a running dirmngr
server. This can be useful to debug / analyze CRL issues.
GnuPG-Bug-Id: T3967
Differential Revision: https://dev.gnupg.org/D469
Signed-off-by: Andre Heinecke <aheinecke@intevation.de>
* dirmngr/ks-engine-hkp.c (ks_hkp_search): Send SOURCE status also on
NO DATA error.
(ks_hkp_get): Ditto.
* g10/call-dirmngr.c (gpg_dirmngr_ks_search): Print "data source" info
also on error.
(gpg_dirmngr_ks_get): Ditto.
--
If a keyserver does not return any data it can be useful to know which
keyserver out of the pool answered.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit bee65edfbc8cc2c369e5941cc9d1a01a0519b388)
* dirmngr/misc.c (host_and_port_from_url): Fix bad port parsing and a
segv for a missing slash after the host name.
--
Reportted-by: Tomas Mraz
GnuPG-bug-id: 4230
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit a3a5a2451924640588e5ecc03a1d4ba6a6ba94a5)
* dirmngr/ks-engine-ldap.c (extract_attributes): Don't use a variabale
out of scope and cleanup the entire pgpKeySize block.
--
GnuPG-bug-id: 4229
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 2b57a8159cdc3b212a4efc68787b40cafcd91ebe)
* dirmngr/dirmngr.c (parse_ldapserver_file): Only print info
for ENOENT.
--
It is very common that there is no ldapserver file. In that
case users are confused by an error printed at dirmngr startup.
* dirmngr/server.c (proc_wkd_get): Tack the raw local address to the
request.
--
We append the raw non-canonicalized local address part to the hash.
Servers who serve the requests from static files will ignore the
parameters and a test with posteo shows that also services using a
database ignore the parameter. The general idea is that service
providers may use their own canonicalization rules. The problem is
that we currently filter the returned key for the full mail address
and thus we will never see a key if the service did a different
canonicalization than we. So consider this to be an experiment.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 256a280c51f9ea862e4bfb0bb530c2a96f9088f9)
* common/asshelp.c (wait_for_sock): Round SECSLEFT.
* dirmngr/dirmngr.c (main): Take care of --debug-wait also in dameon
mode.
* common/sysutils.c (gnupg_usleep) [HAVE_NANOSLEEP]: Fix nanosleep use.
--
Without the rounding we saw in verbose mose
[...]to come up ... (5s)
[...]to come up ... (4s)
immediately without the expected one second delay. Waiting for the
next seconds did not work if nanosleep was used due to improper passed
parameters in gnupg_usleep.
Adding --debug-wait for dirmngr in daemon mode is required to test
this change.
GnuPG-bug-id: 3490
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 0cfdd3b57d592fb6baa7dafe8fde124e8a6c7798)
Fixes-commit: 1189df2cd7d4b6896ba22aa204c159ff2a425ead
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/server.c (proc_wkd_get): Check the returned SRV record names
to mitigate rogue DNS servers.
--
I am not sure wether this really is very useful because the security
relies on a trustworthy DNS system anyway. However, that check is
easy enough to do.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit ebe727ef596eefebb5eff7d03a98649ffc7ae3ee)
* dirmngr/dns.c (dns_res_exec): If it's DNS_SO_SOCKS_CONN, don't
iterate to other server, but return the error immediately.
--
Cherry picked from master commit:
1aacd12471935a354cfd85ee1805edc7eb16e6c5
In the function libdns_switch_port_p in dns-stuff.c, this patch
allows to fallback using TOR_PORT2 correctly.
Fixes-commit: bcdbf8b8ebe9d61160e0b007dabe1b6462ffbc93
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dns.c (LEAVE_SELECTION_OF_PORT_TO_KERNEL): New.
(dns_socket): Don't select ephemeral port in user space.
--
Cherry picked from master commit:
861f1da0731bf29dcb9221c4f22c76b40ec15a78
There is no good reason to bind local port aggressively. It might be
some reason to do so, then, a user can specify it in /etc/resolv.conf
by the second argument of "interface" directive.
At least, it causes a problem on Windows. Binding a specified port in
user space can trigger the Firewall dialog on Windows. Since it can
be considered valid question, it is better not to bind with an
ephemeral port which is selected in user space, by default.
GnuPG-bug-id: 3610
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dns.c (dns_so_check): When EINVAL, release the association
by connect with AF_UNSPEC and try again. Also try again for
ECONNREFUSED.
(dns_res_exec): Try next nameserver when ECONNREFUSED.
--
Cherry pick from master commit:
bcdbf8b8ebe9d61160e0b007dabe1b6462ffbc93
GnuPG-bug-id: T3374
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dns.c (dns_resconf_pton): Clear SS.
(dns_resconf_setiface): Clear ->IFACE.
(dns_hints_root, send_query): Clear SS.
--
Cherry pick from master commit:
1c0b6681e4f322b88ac35d1f21c03d3cfc35fc23
POSIX requires clear the structure of struct sockaddr_in6. On macOS,
in some case like bind, it is better to clear even for struct
sockaddr_in.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dns-stuff.c (libdns_init): Initialize options.recurse.
--
Cherry pick from master commit:
5b40338f12762cd74238c2d2b3101c33dd2d0ed3
To reproduce an error, run:
./t-dns-stuff --debug --recursive-resolver www.gnupg.org
Then, it returns "No name" error. That's because there was only setup
for root servers, and no setup for recursive query in fact.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/ldap-wrapper.c (wrapper_list): Rename to reaper_list.
(ldap_reaper_thread): Protect all list modification with a mutex. Use
a condition var to wake up the reaper thread.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ldap-wrapper.c: Revamp module to make use of es_poll for
portability.
* configure.ac: Always use the ldap wrapper.
--
Since the migration from GNU Pth to nPth the ldap wrapper never worked
reliable on Windows. Our long term use of the old Window CE wrapper
thing didn't fixed this either. The new code uses the portable
es_poll function and thus code which is tested at several other
places. It Should(tm) fix the Windows issues.
GnuPG-bug-id: 3937
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dirmngr_ldap.c: Remove assert.h.
(main): Replace assert by log_assert.
* dirmngr/ldap.c (run_ldap_wrapper): Use debug options to pass
verbose options to dirmngr_ldap.
(start_cert_fetch_ldap): Ditto.
--
verbose is a pretty common option in dirmngr.conf and it would clutter
the logs with output from dirmngr_ldap. Now we require DBG_EXTPROG
or DBG_LOOKUP to make dirmngr_ldap more verbose.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (RESURRECT_INTERVAL): Decrease.
(INITIAL_HOSTTABLE_SIZE): Increase because the old values was likely
for development.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (map_host): Don't use the cname for HTTPHOST.
* dirmngr/server.c (make_keyserver_item): Map keys.gnupg.net.
--
For a description of the problem see the comment in
make_keyserver_item.
GnuPG-bug-id: 3755
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/http.c (send_request): Print the used TLS library in debug
mode.
--
We allow two different TLS libararies and thus it is useful to see
that in the debug output of bug reports.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine.h (KS_HTTP_FETCH_NOCACHE): New flag.
(KS_HTTP_FETCH_TRUST_CFG): Ditto.
(KS_HTTP_FETCH_NO_CRL): Ditto.
(KS_HTTP_FETCH_ALLOW_DOWNGRADE): Ditto.
* dirmngr/ks-engine-http.c (ks_http_fetch): Replace args send_no_cache
and extra_http_trust_flags by a new flags arg. Allow redirectiong
from https to http it KS_HTTP_FETCH_ALLOW_DOWNGRADE is set.
* dirmngr/loadswdb.c (fetch_file): Call with KS_HTTP_FETCH_NOCACHE.
* dirmngr/ks-action.c (ks_action_get): Ditto.
(ks_action_fetch): Ditto.
* dirmngr/crlfetch.c (crl_fetch): Call with the appropriate flags.
--
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/http.h (HTTP_FLAG_TRUST_CFG): New flag.
* dirmngr/http.c (http_register_cfg_ca): New.
(http_session_new) [HTTP_USE_GNUTLS]: Implement new trust flag.
* dirmngr/certcache.c (load_certs_from_dir): Call new function.
(cert_cache_deinit): Ditto.
* dirmngr/http-ntbtls.c (gnupg_http_tls_verify_cb): Ditto.
* dirmngr/ks-engine-http.c (ks_http_fetch): Add new args
'send_no_cache' and 'extra_http_trust_flags'. Change all callers to
provide the default value.
* dirmngr/crlfetch.c (crl_fetch): Rewrite to make use of
ks_http_fetch.
--
The old code simply did not use https for downloading of CRLS.
Instead it rewrote https to http under the assumption that the CRL
service was also available without encryption. Note that a CRL is
self-standing and thus it does not need to have extra authenticity as
provided by TLS. These days we should not use any unencrypted content
and thus this patch.
Be aware that cacert.org give a https CRL DP but that currently
redirects to to http! This is a downgrade attack which we detect and
don't allow. The outcome is that it is right now not possible to use
CAcert certificates.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/server.c (cmd_isvalid): Use option second arg to trigger
OCSP checkibng. Fallback to CRL if no default OCSP responder has been
configured.
* sm/call-dirmngr.c (gpgsm_dirmngr_isvalid): Adjust accordingly.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/crlcache.c (lock_db_file, crl_cache_insert): Open cache
file in binary mode.
--
CRLs on Windows would have line ending entries converted. This
did not cause problems in a surprising amount of cases but
can lead to unexpected and random parse / read errors. Especially
with large CRLs like cacert.
This bug has been around since 2004.
GnuPG-Bug-Id: T3923
Signed-off-by: Andre Heinecke <aheinecke@intevation.de>
* agent/command.c (cmd_getinfo): Add sub-command getenv.
* dirmngr/server.c (cmd_getinfo): Ditto.
--
It is sometimes helpful to be able to inspect certain envvars in a
running agent. For example "http_proxy".
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (handle_send_request_error): Handle two more
error codes.
--
Handle the two possible connect failures related to missing IPv6 support
gracefully by marking the host dead and retrying with another one.
If IPv6 is disabled via procfs, connect() will return EADDRNOTAVAIL.
If IPv6 is not compiled into the kernel, it will return EAFNOSUPPORT.
This makes it possible to use dual-stack hkp servers on hosts not having
IPv6 without random connection failures.
GnuPG-bug-id: 3331
--
The above description seems to be for Linux, so it is possible that
other systems might behave different. However, it is worth to try
this patch.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/asshelp2.c (vprint_assuan_status_strings): New.
(print_assuan_status_strings): New.
* agent/command.c (agent_write_status): Replace by call to new
function.
* dirmngr/server.c (dirmngr_status): Ditto.
* g13/server.c (g13_status): Ditto.
* g13/sh-cmd.c (g13_status): Ditto.
* sm/server.c (gpgsm_status2): Ditto.
* scd/command.c (send_status_info): Bump up N.
--
This fixes a potential overflow if LFs are passed to the status
string functions. This is actually not the case and would be wrong
because neither the truncating in libassuan or our escaping is not the
Right Thing. In any case the functions need to be more robust and
comply to the promised interface. Thus the code has been factored out
to a helper function and N has been bumped up correctly and checked in
all cases.
For some uses this changes the behaviour in the error case (i.e. CR or
LF passed): It will now always be C-escaped and not passed to
libassuan which would truncate the line at the first LF.
Reported-by: private_pers
* agent/gpg-agent.c (oListenBacklog): New const.
(opts): New option --listen-backlog.
(listen_backlog): New var.
(main): Parse new options.
(create_server_socket): Use var instead of 5.
* dirmngr/dirmngr.c: Likewise.
* scd/scdaemon.c: Likewise.
--
GnuPG-bug-id: 3473
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/domaininfo.c (insert_or_update): Copy the name.
* dirmngr/misc.c (copy_stream): Allow arg OUT to be NULL.
* dirmngr/server.c (set_error): Protect CTX.
(dirmngr_status): Protect against missing ASSUAN_CTX.
(dirmngr_status_help): Ditto.
(dirmngr_status_printf): Ditto.
(cmd_wkd_get): Factor code out to ...
(proc_wkd_get): new func. Support silent operation with no CTX.
(task_check_wkd_support): New.
--
This finalizes the feature to efficiently cache WKD checks. If a
standard WKD query returns no data, we queue a test to be run after
the end of the session (so that we do not delay the calling client).
This check tests whether the server responsible for the queried
address has WKD at all enabled. The test is done by checking whether
the "policy" file exists. We do not check the "submission-address"
file because that is not necessary for the web key operation. The
policy file is now required.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit d4e2302d8f4a1ff52d56da4f8e3a5d1c6303822d)
* dirmngr/workqueue.c: New.
* dirmngr/Makefile.am (dirmngr_SOURCES): Add new file.
* dirmngr/server.c (server_local_s): New field session_id.
(cmd_wkd_get): Add a task.
(task_check_wkd_support): New stub function.
(cmd_getinfo): New sub-commands "session_id" and "workqueue".
(start_command_handler): Add arg session_id and store it in
SERVER_LOCAL.
(dirmngr_status_helpf): New.
* dirmngr/dirmngr.h (wqtask_t): New type.
* dirmngr/dirmngr.c (main): Pass 0 as session_id to
start_command_handler.
(start_connection_thread): Introduce a session_id and pass it to
start_command_handler. Run post session tasks.
(housekeeping_thread): Run global workqueue tasks.
--
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 96a4fbecd1acf946dcde20bef4752c539dae196b)
* dirmngr/domaininfo.c (MAX_DOMAINBUCKET_LEN): New.
(insert_or_update): Limit the length of a bucket chain.
(domaininfo_print_stats): Print just one summary line.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 26f08343fbccdbaa177c3507a3c5e24a5cf94a2d)
* dirmngr/domaininfo.c: New file.
* dirmngr/Makefile.am (dirmngr_SOURCES): Add file.
* dirmngr/server.c (cmd_wkd_get): Check whether the domain is already
known and tell domaininfo about the results.
--
This adds a registry for domain information to eventually avoid
useless queries for domains which do not support WKD. The missing
part is a background task to check whether a queried domain supports
WKD at all and to expire old entries.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 65038e6852185c20413d8f6602218ee636413b77)
* dirmngr/crlcache.c: Clearly document that this fucntions takes
ownership of MD.
(abort_sig_check): Allow NULL for MD.
(crl_parse_insert): Immediately set MD to NULL. Remove check for md
before a calling abort_sig_check.
--
GnuPG-bug-id: 3510
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-http.c (ks_http_fetch): Forbid redirects from
a https URI to a http URI.
--
GnuPG-bug-id: 3436
Signed-off-by: Damien Goutte-Gattat <dgouttegattat@incenp.org>
* dirmngr/certcache.c (any_cert_of_class): New var.
(put_cert): Set it.
(cert_cache_deinit): Clear it.
(cert_cache_any_in_class): New func.
* dirmngr/http-ntbtls.c (gnupg_http_tls_verify_cb): Add hack to
override empty list of HKP certs.
--
This patch carries the changes for GNUTLS from commit
7c1613d41566f7d8db116790087de323621205fe over to NTBTLS. NTBTLS works
quite different and thus we need to do it this way.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c (have_homedir_inotify): New var.
(reliable_homedir_inotify): New var.
(main): Set reliable_homedir_inotify.
(handle_tick): Call stat on the homedir.
(handle_connections): Mark availibility of the inotify watch.
* dirmngr/dirmngr.c (handle_tick): Call stat on the homedir.
(TIMERTICK_INTERVAL_SHUTDOWN): New.
(handle_connections): Depend tick interval on the shutdown state.
--
The stat call is used on systems which do not support inotify and also
when we assume that the inotify does not work reliable.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c (main): Always to the chdir.
* dirmngr/dirmngr.c (main): Ditto.
* scd/scdaemon.c (main): Ditto.
--
Note that only dirmngr did not call the chdir with --no-detach. thus
we kept it this way.
Tested gpg-agent by checking the properties shown by procexp.
Gnupg-bug-id: 2670
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/server.c (cmd_wkd_get): Print a SOURCE status.
--
This status allows to see whether the the WKD requests has been
resolved from the standard address or from a SRV record derived one.
We return only host and port and not the .well-known suffix because
that is obvious. HTTP redirects are not taken in account because they
may chnage at any time due to load balancing etc and not relevant for
gpg which may use the URL to detect changes in the WKD results.
For example my current setup returns
S SOURCE https://wkd.gnupg.org
for wk@gnupg.org. Without a SRV record
S SOURCE https://gnupg.org
would have been returned.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c [HAVE_W32_SYSTEM]: Include <direct.h>.
(main) [HAVE_W32_SYSTEM]: Change working directory to \.
* dirmngr/dirmngr.c [HAVE_W32_SYSTEM]: Include <direct.h>.
(main) [HAVE_W32_SYSTEM]: Change working directory to \.
* scd/scdaemon.c [HAVE_W32_SYSTEM]: Include <direct.h>.
(main) [HAVE_W32_SYSTEM]: Change working directory to \.
Signed-off-by: Marcus Brinkmann <mb@g10code.com>
GnuPG-bug-id: 2670
* dirmngr/dns.c (HAVE_STATIC_ASSERT, HAVE___ATOMIC_FETCH_ADD)
(DNS_HAVE_SOCKADDR_UN, HAVE_SOCK_NONBLOCK): Don't use defined
to be expanded for expression evaluation.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/http.c (send_request): If a http proxy is to be used, and we
want to use TLS, try to use the CONNECT method to get a connection to
the target server.
GnuPG-bug-id: 2940
Signed-off-by: Justus Winter <justus@g10code.com>
* dirmngr/http.c (http_wait_response): Get the 'use_tls' flag from the
write cookie, not from the URI.
Signed-off-by: Justus Winter <justus@g10code.com>
* dirmngr/http.c (send_request): Do not use the 'srvtag' intended for
the target host to connect to the http proxy.
Signed-off-by: Justus Winter <justus@g10code.com>
* dirmngr/http.c (parse_uri): Keep an unmodified copy of the URI.
* dirmngr/http.h (struct parsed_uri_s): New field 'original'.
* dirmngr/ks-action.c (ks_action_get): Properly handle http and https
URLs.
--
If a key has a http or https URL as preferred keyserver, fetch the key
from there. Previously, dirmngr unconditionally interpreted these
URLs as hkp servers.
GnuPG-bug-id: 2924
Signed-off-by: Justus Winter <justus@g10code.com>
--
This double license note was accidentally added while only wanting to
add another copyright line.
Fixes-commit: 3419a339d9c4e800bf30e9021e05982d8c1021c1
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ks-engine-hkp.c (enum ks_protocol): New type.
(struct hostinfo_s): New flags indicating whether we already did a
A lookup, or a SRV lookup per protocol. Turn 'port' into an array.
(create_new_hostinfo): Initialize new fields.
(add_host): Update the port for the given protocol.
(map_host): Simplify hosttable lookup misses. Check the SRV records
for both protocols on demand, do the A lookup just once. Return the
correct port.
--
Previously, if a host had both a SRV record for hkp and hkps, the
wrong port was used for the protocol that was used second, because the
hostinfo did not store a port per protocol, and the hosttable does not
discriminate between hosts using the protocol.
Fix this by querying the SRV records on demand, storing a port per
protocol, and returning the right port.
GnuPG-bug-id: 3033
Signed-off-by: Justus Winter <justus@g10code.com>
* dirmngr/ks-engine-hkp.c (struct hostinfo_s): Add explicit length and
size fields.
(MAX_POOL_SIZE): New macro.
(create_new_hostinfo): Initialize new fields.
(host_in_pool_p): Adapt.
(select_random_host): Likewise.
(add_host): Likewise. Move the resizing logic here.
(hostinfo_sort_pool): New function.
(map_host): Simplify. Move the resizing logic away from here.
(ks_hkp_mark_host): Adapt.
(ks_hkp_print_hosttable): Likewise.
--
The current code assumes that the pool array is only filled when the
hostinfo object is created. This patch removes that limitation.
GnuPG-bug-id: 3033
Signed-off-by: Justus Winter <justus@g10code.com>
* dirmngr/dns.c (dns_so_check): Reinitialize sockets on address family
mismatch.
(enum dns_res_state): New states for querying over IPv6.
(dns_res_exec): Implement the new states by copying and modifying the
IPv4 variants. Branch to their respective counterparts if the current
list of resolvers using the current address family is exhausted.
--
This allows dirmngr to resolve names on systems where the nameservers
are only reachable via IPv6.
GnuPG-bug-id: 2990
Signed-off-by: Justus Winter <justus@g10code.com>
* dirmngr/dirmngr.c (oConnectTimeout, oConnectQuickTimeout): New
enums.
(opts): New options --connect-timeout and --connect-quick-timeout.
(DEFAULT_CONNECT_TIMEOUT): New.
(DEFAULT_CONNECT_QUICK_TIMEOUT): New.
(parse_rereadable_options): Handle new options.
(post_option_parsing): New. Use instead of direct calls to
set_debug() and set_tor_mode ().
(main): Setup default timeouts.
(dirmngr_init_default_ctrl): Set standard connect timeout.
* dirmngr/dirmngr.h (opt): New fields connect_timeout and
connect_quick_timeout.
(server_control_s): New field timeout.
* dirmngr/ks-engine-finger.c (ks_finger_fetch): Pass timeout to
http_raw_connect.
* dirmngr/ks-engine-hkp.c (send_request): Call
http_session_set_timeout.
* dirmngr/ks-engine-http.c (ks_http_fetch): Ditto.
* dirmngr/server.c (cmd_wkd_get, cmd_ks_search, cmd_ks_get)
(cmd_ks_fetch): Implement --quick option.
--
The standard connect timeouts are way to long so we add a timeout to
the connect calls. Also implement the --quick option which is already
used by gpg for non-important requests (e.g. looking up a key for
verification).
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/http.c: Include fcntl.h.
(http_session_s): Add field 'connect_timeout'.
(http_session_new): Clear that.
(http_session_set_timeout): New function.
(my_wsagetlasterror) [W32]: New.
(connect_with_timeout): New function.
(connect_server): Add arg 'timeout' and call connect_with_timeout.
(send_request): Add arg 'timeout' and pass it to connect_server.
(http_raw_connect): Add arg 'timeout'.
(http_open): Pass TIMEOUT from the session to connect_server.
--
Note that the non-blocking connect we implement is traditional a
pretty non-portable thing due to slighly different semantics. The
code uses the strategy W. Richard Stevens suggested in 1998.
Hopefully current OS versions got it all right.
The code has not been tested on Windows.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dns-stuff.c: Include sys/stat.h.
(RESOLV_CONF_NAME): New macro to replace a string.
(resolv_conf_changed_p): New.
(libdns_init): Call new function
(libdns_res_open): Ditto.
--
Don't panic. This is a simple change
Suggested-by: Stefan Bühler <stbuehler@web.de>
to avoid complicated if-up.d hooks to reload resolv.conf.
Signed-off-by: Werner Koch <wk@gnupg.org>
Summary:
* dirmngr/ldap-wrapper-ce.c (outstream_cookie_s): Add buffer_read_pos.
(buffer_get_data): Use seperate read pos.
--
Using a single buffer pos for reading and writing caused the read
to return 0 as it read from the end of the buffer. Now we use
a seperate reader position.
Differential: D427
Signed-off-by: Andre Heinecke <aheinecke@intevation.de>
* dirmngr/http.c (read_server, write_server): Use assuan_fd_t.
(http_wait_response): Use FD2INT to get unsigned integer fd.
(read_server, write_server): Likewise.
(simple_cookie_read, simple_cookie_write): Use assuan_fd_t.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/http.c (simple_cookie_read, simple_cookie_write): Only
valid with HTTP_USE_NTBTLS.
(_my_socket_new): Simply cast to int since it's for debug.
(_my_socket_ref, _my_socket_unref): Likewise.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dns-stuff.c (T_CERT): Define our own.
--
T_CERT may be defined by another enum type even if the value is same.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dns-stuff.c: Don't include arpa/nameser.h.
--
It is not needed at all. T_CERT may be defined by different type of
ns_type.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dns.c (plus1_ns): New.
(dns_p_push): Use it.
--
On OpenBSD, htons and ntohs are expanded to GCC's statement
expressions where local variable is allowed. Consecutive use
of htons and ntohs causes problem of variable name.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* dirmngr/dirmngr.h (struct opt): Add field 'disable_ipv6'.
* dirmngr/dirmngr.c (oDisableIPv6): New const.
(opts): New option --disable-ipv6.
(parse_rereadable_options): Set that option.
* dirmngr/dns-stuff.c (opt_disable_ipv6): New var.
(set_dns_disable_ipv6): New.
(resolve_name_standard): Make use of it.
* dirmngr/ks-engine-finger.c (ks_finger_fetch): Take care of
OPT.DISABLE_IPV6.
* dirmngr/ks-engine-hkp.c (map_host): Ditto.
(send_request): Ditto.
* dirmngr/ks-engine-http.c (ks_http_fetch): Ditto.
* dirmngr/ocsp.c (do_ocsp_request): Ditto.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dns-stuff.c (libdns_init): No Windows specific handling of a
missing /etc/hosts.
--
My last comment on this was flawed. Windows seems to always have its
version of /etc/hosts. Only the en passant fixed bad escaping led me
assume that this was the case. Thanks to Andre for complaining about
my comment remark.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dns-stuff.c (libdns_init): Do not bail out.
--
A standard Windows installation does not have a hosts file and thus we
can't bail out here. We should also not bail out on a Unix system
because /etc/hosts is just one method in nsswitch.conf.
Fixes-commit: 88f1505f0613894d5544290a170119eb538921e5
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dns-stuff.c (libdns_init): Actually load the hosts file into
libdns.
--
Previously, connecting to key servers specified in /etc/hosts was not
possible because libdns' hosts structure was initialized, but not
filled with the content of the hosts file.
GnuPG-bug-id: 2977
Signed-off-by: Justus Winter <justus@g10code.com>
