summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Document build requirements a little better.HEADmasterGünther Deschner2012-04-051-0/+20
| | | | | | resolves ticket: #37 Guenther
* Debug: Add debugging macro and config optionsSimo Sorce2012-04-025-0/+38
| | | | Ticket #43
* Avoid double free on error.Simo Sorce2012-04-021-10/+0
| | | | | There is no need to expliciltly free here, as gp_rpc_process_call already always frees the res structure. By doing this we cause double frees on errors.
* Official IANA Assigned RPC Program NumberSimo Sorce2012-03-302-2/+2
|
* conv: Fix typo causing bad output.Simo Sorce2012-03-221-1/+1
|
* gp_socket: properly handle fagment bitSimo Sorce2012-03-222-2/+20
|
* gp_socket: Optimize sending data with writevSimo Sorce2012-03-221-21/+22
| | | | | | Turns out the Linux kernel expects to get the whole reply in a single operation. This optimizes the code to call one less syscall and makes it also more compact. So let's please the kernel and get better code in place.
* Add option to request (or not) delegated credentials backSimo Sorce2012-03-224-2/+11
|
* gp_conv: Fix conversion functions after change to gss_proxy.xSimo Sorce2012-03-211-8/+2
|
* gss-proxy.x: We do not need pointers in gssx_ctxSimo Sorce2012-03-133-6/+6
| | | | There is no need to distinguish between empty buffer and no buffer.
* Update xdr files after changes to master gss_proxy.x fileSimo Sorce2012-03-092-194/+119
|
* Get rid of extensions type and replace with optionsNicolas Williams2012-03-091-55/+34
| | | | | | | And make the typed-hole type field a string (URN) instead of OID. Signed-off-by: Simo Sorce <simo@redhat.com>
* config: parse credential/service config sectionsSimo Sorce2012-02-233-1/+291
|
* gp_conv: use maj/min err codes with gp_conv_ctx_id_to_gssx tooSimo Sorce2012-02-234-17/+18
|
* server: better handle return statusSimo Sorce2012-02-2310-124/+181
| | | | Ticket #33
* Tidyup gp_accept_sec_context()Simo Sorce2012-02-231-5/+2
|
* workers: Fix handling of workers query handlingSimo Sorce2012-02-231-15/+13
| | | | | | Chekcing for query == NULL outside of the mutex was wrong as it may make us end up restarting operation on the value of q we just handed back as if it was a new query. We just need to always go through the conditional lock.
* Test init_sec_contextSimo Sorce2012-02-091-22/+25
|
* There is no need anymore to load the server config in the client test program.Simo Sorce2012-02-092-14/+1
|
* Retrieve the buffer when needed, right before accept_sec_contextSimo Sorce2012-02-091-6/+6
|
* Remove useless connection to the proxySimo Sorce2012-02-091-37/+0
| | | | Connecting to the proxy is handled internally by the client mechglue library.
* Rename test program.Simo Sorce2012-02-093-6/+6
| | | | | The test program is now testing almost all functions so change name to reflect reaility.
* Implement init_sec_context mechglue functionSimo Sorce2012-02-093-0/+186
|
* Implement server side init_sec_context function.Simo Sorce2012-02-093-4/+140
|
* gp_conv: Add function to import sec contextSimo Sorce2012-02-092-0/+17
|
* gp_conv: return GSS_C_NO_OID if input is null or zero length.Simo Sorce2012-02-091-0/+5
| | | | Fixes useless allocation in gp_conv_gssx_to_oid_alloc()
* gp_conv: Use correct type to calculate size of allocationSimo Sorce2012-02-091-1/+1
|
* gp_conv: Allow null oids as input in conversion functionsSimo Sorce2012-02-091-0/+12
|
* Ignore more built filesSimo Sorce2012-02-071-0/+5
|
* Add reference to codying style guideSimo Sorce2012-02-071-0/+10
|
* Add import_and_canon_name function family testsSimo Sorce2012-02-071-8/+45
|
* Add more helper functions to gp_convSimo Sorce2012-02-072-15/+53
|
* Implement import_and_canon_name mechglue functions familySimo Sorce2012-02-073-0/+286
|
* Implement server side import_and_canon_name()Simo Sorce2012-02-075-17/+164
| | | | | Also fix name conversion functions, to properly handle exporting/importing names.
* gssx_name -> display_name should not be a pointerSimo Sorce2012-02-074-21/+10
|
* Add tests for indicate_mechs functionsSimo Sorce2012-02-052-0/+80
|
* Add more functions of the indicate_mechs() familySimo Sorce2012-02-052-14/+466
|
* Fix attr description set in gp_indicate_mechs()Simo Sorce2012-02-051-0/+29
|
* Fix accept_sec_context to properly return src_nameSimo Sorce2012-02-053-0/+57
|
* Add gpm_indicate_mechs to test programSimo Sorce2012-02-031-1/+9
|
* Add gpm_indicate_mechs functions to mechglue librarySimo Sorce2012-02-033-0/+310
|
* Add server implementation of gssx_indicate_mechsSimo Sorce2012-02-033-4/+253
|
* Allow conversion of empty strings.Simo Sorce2012-02-031-0/+11
|
* Change gssx_mechinfoSimo Sorce2012-02-033-14/+10
| | | | | | | | We do not need provider names and paths apssed to clients. Clients have no use for such info and can be argued it is something admins do not want to disclose anyway. Add saslname extensions that were missing from the first draft.
* Fix loop in case of errors and use gp_debugSimo Sorce2012-02-031-23/+6
| | | | | | | We need to have 2 variables to hold the array index otherwise we will leave holes and loose elements that come after a failure. Use gp_log_failure() to make code more readable
* Add simple debugging helperSimo Sorce2012-02-033-0/+87
|
* Fix gp_conv_gssx_to_oid_allocSimo Sorce2012-02-031-0/+2
|
* Temporary workaround for MIT gssapi bugsSimo Sorce2012-02-011-0/+26
| | | | | gss_iniquire_cred_by_mech() doesn't work on all mechanisms returned by gss_inquire_cred(), this is a but in MIT libraries that MIT is going to fix.
* Modify accept_context test to use acquire_credSimo Sorce2012-02-011-3/+17
|
* Add gpm_acquire_cred functions to mechglue librarySimo Sorce2012-02-013-0/+307
|
generated by cgit (git 2.34.1) at 2024-05-05 01:19:54 +0000