summaryrefslogtreecommitdiffstats
path: root/proxy/src
Commit message (Collapse)AuthorAgeFilesLines
* Add mechglue functions declarationsSimo Sorce2012-06-261-0/+114
|
* Add simple functions to map errorsSimo Sorce2012-06-262-0/+26
| | | | | | | | | | The mechglue stores a map of errors/mech oids, this means that we should never return the same error we got from a mechanism after re-entering the mechglue as we then may get the mechglue confused and prevent us from asking an interposed mech for the error. Also we want to try to aqvoid collisions from errors returned from the proxy, as they could end up fetching errors from the wrong mechanism. For now just make a very simple mapping by always adding a special error base.
* Add function to special-filter OID setsSimo Sorce2012-06-262-0/+54
|
* Add function to return a special mechSimo Sorce2012-06-263-0/+62
| | | | | | | When the interposer wants to call the mechglue and have it call a real mechanism it does so by providing a speecial mechanism oid. This is an invalid oid that the mechglue transforms back into a real OID before selecting the appropriate mechanism.
* Add loop avoidance in proxy daemon and gssapi pluginSimo Sorce2012-06-262-0/+11
|
* Add initialization codeSimo Sorce2012-06-262-0/+201
| | | | | | | For now return fixed list of mechanisms. Later on we can try to fetch this list from the proxy. Also split RPC client code from actual plugin
* Add macro to suppress const warningsSimo Sorce2012-06-261-0/+2
|
* Save internal errors in init_sec_contextSimo Sorce2012-06-261-0/+2
|
* Add function to report internal client errorsSimo Sorce2012-06-262-0/+19
| | | | | | | This is to allow gss_display_status to return a meaningful error if there is an internal client error as opposed to a returned proxy server error. If we do not do this a call to gpm_display_status() after a failure would return a NULL string and give no clue about the error.
* Move release_name to the proper placeSimo Sorce2012-06-262-11/+10
|
* Fix gpm_display_statusSimo Sorce2012-06-261-19/+13
| | | | | | | Fix some assumptions in gpm_display_status and make it clearly state no error is available, when we do not have one. This is so that the plugin code later on will know when to properly fall back to ask to the local provider.
* Add gss_inquire_name to mechglueSimo Sorce2012-06-262-0/+64
|
* Fix DEBUG message formatSimo Sorce2012-06-261-1/+2
|
* memset default input token for init_sec_contextSimo Sorce2012-06-261-3/+2
|
* Fix handling deleg_creds in accept_sec_contextSimo Sorce2012-06-261-13/+10
| | | | We would segfault if the user didn't want delegated redentials.
* Fix segfaults when NULL buffer is passed in.Simo Sorce2012-06-262-5/+11
|
* Use gp_find_and_free_cred() in gp_release_handle().Günther Deschner2012-06-251-4/+1
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Add gp_find_and_free_cred().Günther Deschner2012-06-252-0/+30
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Add free_config() and call it when server shuts down.Günther Deschner2012-06-253-0/+27
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Finally use ring buffer in import and export calls.Günther Deschner2012-06-251-8/+84
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Make gp_import_gssx_cred() static, it is not used outside this file.Günther Deschner2012-06-252-2/+1
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Pass down struct gp_service to import and export credential calls.Günther Deschner2012-06-256-7/+15
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Allow to set per service ring buffer, with configurable size.Günther Deschner2012-06-251-0/+45
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Set pointer to one of the existing ring buffers in service loading.Günther Deschner2012-06-251-0/+8
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Add gp_service_get_ring_buffer to retrieve buffer from a service.Günther Deschner2012-06-252-0/+5
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Add gp_ring_buffer pointer to struct gp_service.Günther Deschner2012-06-251-0/+4
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Add gp_read_gss_creds_from_ring_buffer/gp_write_gss_creds_to_ring_buffer.Günther Deschner2012-06-251-0/+93
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Add two ring_buffers to gp_config struct.Günther Deschner2012-06-252-0/+36
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Add gp_init_ring_buffer/gp_free_ring_buffer.Günther Deschner2012-06-252-1/+143
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Mention socket filename in debug output in init_unix_socket().Günther Deschner2012-06-211-1/+2
| | | | Guenther
* Fix option_is_set() function.Günther Deschner2012-06-211-1/+1
| | | | Guenther
* use option_is_set() config parser helper.Günther Deschner2012-05-311-7/+2
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Fix popt usage message for -d option.Günther Deschner2012-05-311-1/+1
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* add option_is_set() helper function for config parsing.Günther Deschner2012-05-311-0/+12
| | | | | | Guenther Signed-off-by: Simo Sorce <simo@redhat.com>
* Replace direct syslog calls with logging macrosSimo Sorce2012-04-154-23/+15
|
* Add logging helpersSimo Sorce2012-04-154-4/+71
|
* Add some socket debuggingSimo Sorce2012-04-151-1/+14
|
* gp_export: Add facility to export creds as gssx optionsSimo Sorce2012-04-153-0/+268
| | | | Add Initial test implementation of the Linux Kernel NFS creds structure.
* Change accept/init_sec_context to use the new export context functionsSimo Sorce2012-04-153-13/+23
|
* Add support for linux kernel lucid ctx buffer formatSimo Sorce2012-04-151-11/+130
|
* Move ctx_id function and add export type optionSimo Sorce2012-04-154-107/+152
| | | | | Move the function to export a context to gssx_ctx or import into gp_export.c Add function to check for exported_context_type option in call_ctx
* Use gp_add_krb5_creds until cred store extensions are available.Simo Sorce2012-04-152-35/+88
|
* creds: add code to import krb5 credentials based on configuration.Simo Sorce2012-04-0515-15/+433
|
* Depend on UCRED, we cannot work otherwise.Simo Sorce2012-04-051-4/+0
|
* config: Rework configuration syntaxSimo Sorce2012-04-052-139/+77
| | | | | | | Keeping 2 separate sections for credentials and services seem to just make things really confusing. The off chance of reusing a 'credential' section is dwarfed by the confusion cause by keeping them separate. Having to copy a full service section is not a big deal so KISS wins here.
* 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.
* 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.