summaryrefslogtreecommitdiff
path: root/src/usr.bin (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Display TLSv1.3 extension type with openssl(1) -tlsextdebuginoguchi2020-04-261-7/+49
| | | | | | | | Add TLSv1.3 extension type, and sort by the definition order in tls1.h. This helps that openssl(1) s_server and s_client with -tlsextdebug displays the TLS extension type instead of "unknown". ok beck@ jsing@ tb@
* s_client: fix use of possibly uninitialized valuesinoguchi2020-04-261-2/+2
| | | | | | | | | | Set initial value to variable 'p' and 'pending'. Reported and fix requested from leonklingele by GitHub pull request. https://github.com/libressl-portable/portable/issues/577 https://github.com/libressl-portable/openbsd/pull/114 ok bcook@ jsing@ tb@
* In s_server.c rev. 1.33, jsing added support for "openssl s_server -groups";schwarze2020-04-251-6/+18
| | | | | | | | document it and deprecate "openssl s_server -named_curve". While here, fix the error in the synopsis for "openssl s_client -groups" and use unified argument naming and similar wording like in SSL_CTX_set1_groups_list(3). OK jsing@
* Add -groups option to openssl(1) s_server.jsing2020-04-192-35/+31
| | | | | | | | | | This allows supported EC groups to be configured, which will also control which TLSv1.3 key shares we'll accept. While here, deprecate the rather useless -named_curve option, which is effectively the same as -groups with a single group. Also stop setting a single default group of P-256 via SSL_CTX_set_tmp_ecdh() - use the library defaults instead. ok beck@ inoguchi@
* Tag subcommand sectionskn2020-02-191-2/+44
| | | | | | | | | Allow semantic lookup by manually tagging sections (.Sh) for which no automatic tagging has been implemented; this thereby also eliminates false positives such that ":tx509<Enter>" now jumps to the X509 section. feedback remi inoguchi schwarze OK tb
* Add -tls1_3 and -notls1_3 options to openssl(1) s_client.jsing2020-02-162-23/+37
| | | | | | | | | Also stop using version pinned methods, instead setting the min and max protocol versions. Requested by inoguchi@ ok inoguchi@ tb@
* Standardize argument naming for "sourceaddr" and unify the wording a bit,schwarze2020-02-122-10/+9
| | | | similar to what deraadt@ recently did in other manual pages.
* revert previous nc loop refactor from 1.211, breaks bluhm's stuffbeck2020-01-261-64/+36
| | | | will attempt again later, now that there is new regress
* add a couple of XXX for future cleanuptb2020-01-251-1/+4
|
* Teach openssl s_client a bit about TLSv1.3.tb2020-01-241-2/+17
| | | | ok beck jsing
* Make -peekaboo mode also use SSL_pending after peeking, to ensurebeck2020-01-231-2/+9
| | | | | | SSL_pending implementation is correct. annoying jsing@
* Simplify the peekaboo code.jsing2020-01-221-35/+6
| | | | ok beck@
* Add -peekaboo option to s_client, to test SSL_peekbeck2020-01-221-4/+66
| | | | | | peeks data before reading, compares to subsequent read. ok jsing@
* If the client provides a TLS certificate and the user specifies abluhm2020-01-071-5/+6
| | | | | | | | | | hash value on the nc(1) server command line, the netcat server must use the TLS context of the accepted socket for verification. As the listening socket was used instead, the verification was always successful. If the peer provides a certificate, there must be a hash. Make the hash verification fail safe. OK tb@
* The unveil(2) for nc -U -u -l was wrong. The server cannot unveilbluhm2020-01-061-4/+23
| | | | | | | the file system as it has to connect to the UNIX domain client socket. The path of the latter is determined dynamically. Instead add a restrictive pledge(2) after connect(2). OK tb@
* When using UNIX domain sockets, always call report_sock() with thebluhm2020-01-061-6/+10
| | | | | | path name of the socket. This avoids bad errors from getnameinfo(3). Use the same error check for both calls to getnameinfo(3). OK millert@ tb@
* Check CMS API return value in openssl(1) cmsinoguchi2020-01-041-11/+21
| | | | ok jsing@
* use "Currently" in the doc for "openssl enc" when talking about defaultsthen2019-12-181-2/+2
| | | | | md, to hint that it might not always be the case (e.g. if dealing with files from a different version of the tool). ok tb@
* In January, the default digest used in the openssl enc command wastb2019-12-181-7/+4
| | | | | | | | changed from md5 to sha256. Update manual to reflect that. From Fabio Scotoni ok jmc
* Add manual for openssl(1) cmsinoguchi2019-11-281-2/+518
| | | | ok and comments jmc@
* More return value check in openssl(1) cmsinoguchi2019-11-191-23/+29
| | | | | | Checking return value of sk_.*_new_null(). ok beck@ jsing@
* Add manual descriptions for openssl(1) req -addextinoguchi2019-11-191-2/+10
| | | | ok jmc@
* Remove typedef and check sk_push return value in openssl(1) cmsinoguchi2019-11-181-23/+33
| | | | | | | | - Remove typedef and use 'struct cms_key_param' instead - Check return value of sk_X509_push and sk_OPENSSL_STRING_push - Add a blank line to separate variable declarations from code comments from jsing@
* Add -keyopt opiton to openssl(1) cms subcommandinoguchi2019-11-181-8/+128
| | | | | | | This provides rsa_padding_mode:oaep for cms -encrypt, and rsa_padding_mode:pss for cms -sign. ok jsing@
* fail to usage if extra argv are presentderaadt2019-11-171-8/+4
| | | | noticed by jsing and beck, ok tedu
* refactor the nc pool loop to not shut down the socket early, andbeck2019-11-131-36/+64
| | | | | | | to handle tls_shutdown correctly if using TLS, doing tls_shutdown correctly if we are using the -N flag ok sthen@
* Check return value and remove unnecessary variableinoguchi2019-11-061-7/+7
| | | | | | | - Check NCONF_new() return value - Remove unnecessary 'i' comments from jsing@
* Add -addext option to openssl(1) req subcommandinoguchi2019-11-061-3/+133
| | | | | | First step of adding -addext option to openssl(1) req from OpenSSL 1.1.1d. ok jsing@
* Fix an out of bound read/write when using a proxy.millert2019-11-042-11/+21
| | | | From Lucas AT sexy DOT is. OK job@ kn@
* Indent labels for diffability.jsing2019-11-041-5/+5
|
* Remove spaces between * and variable names.jsing2019-11-041-25/+25
|
* Remove explicit NULL checks before *_free() calls.jsing2019-11-041-21/+14
|
* Hook openssl(1) cms back up.jsing2019-11-043-4/+8
|
* Currently we need to include pem.h before cms.h...jsing2019-11-041-2/+3
|
* Remove engine argument from load_cert() calls.jsing2019-11-041-8/+8
| | | | This was cleaned up after cms went to the attic.
* Bring openssl(1) cms back from the attic.jsing2019-11-041-0/+1142
|
* Service names are still resolved with -nkn2019-10-241-4/+4
| | | | | | | | | Just like pfctl(8)'s -N, this flag only avoid DNS; "nc -vz ::1 socks" still works. Fix documentation by copying pfctl's wording. OK deraadt
* Print IP address in verbose modejob2019-10-241-12/+34
| | | | OK kn@
* Revert previous, which works for -N case but causes regress failuresbeck2019-10-231-18/+1
| | | | | | | for tls, since the socket is shut down without calling tls_close(). Since nc appears to have a problem with this in other shutdown() cases I am simply going to bake a new diff for this. noticed by bluhm@.
* Fix -N flag to actually shut down the (entire) socket when the inputbeck2019-10-171-1/+18
| | | | | | | | | | | | | goes away. This allows for using nc in cases where the network server will no longer expect anything after eof, instead of hanging waiting for more input from our end. Additionaly, shut down if tls is in use if either side of the socket goes away, since we higher level TLS operations (tls_read and write) will require the socket to be both readable and writable as we can get TLS_WANT_POLLIN or TLS_WANT_POLLOUT on either operation. deraadt@ buying it. found by sthen@
* Avoid a path traversal bug in s_server on Windows.bcook2019-10-041-3/+3
| | | | | | | | openssl s_server has an arbitrary read vulnerability on Windows when run with the -WWW or -HTTP options, due to an incomplete path check logic. Thanks to Jobert Abma for reporting. ok tb@
* the formatting for the mini synopses in this page did not render welljmc2019-10-041-115/+179
| | | | | | | on html or groff. the solution, to replace the non-standard .nr macros with a hang list, was provided by ingo - thanks! ok schwarze
* Remove unnecessary NULL check before free function in openssl(1) dgstinoguchi2019-08-301-11/+6
|
* Wrap lines over 80 cols and put space before goto label in openssl(1) dgstinoguchi2019-08-301-20/+30
|
* Simplify checking and more readable descriptions in openssl(1) dgstinoguchi2019-08-301-8/+8
| | | | suggested from jsing@
* Convert openssl(1) dgst to the newer style of option handlinginoguchi2019-08-301-99/+216
| | | | | | | | Adapt openssl(1) dgst command to new option handling. Added dgst_options struct and option handlers, and replaced for-if-strcmp handling with options_parse(). ok bcook@ jsing@
* added /* no filesystem visibility */ above unveil("/", "") since "" is too easymestre2019-08-081-1/+2
| | | | | | to misread. as per suggestion by and OK deraadt@
* Remove -port option from s_server since it is same as -acceptinoguchi2019-08-051-4/+3
| | | | ok schwarze@
* Make proxy auth work with http 1.1 speaking webservers.benno2019-07-291-2/+3
| | | | | Diff from Alexander Koeppe format_c -AT- online -DOT- de, thanks. ok deraadt@
* Moving variables into struct in openssl(1) dgstinoguchi2019-07-291-86/+99
| | | | | | | | First step to adapt openssl(1) dgst command to new option handling. There is no functional changes by this diff, and just moving variables into dgst_config struct. ok bcook@
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-03-19 02:17:25 +0000