rc_auth_proxy(3)
β¨ rc_auth_proxy β the manual, but make it make sense β¨
okay bestie, sit down, hydrate, because we're talking about rc_auth_proxy(3) from the radcli Radius client library and it's genuinely not that deep once you see it. it's a section 3 library function, it lives in the radcli world, and it's the one you reach for when you're proxying an auth request. let's go π
πΈ NAME πΈ
rc_auth_proxy β builds an authentication request for proxying, no cap.
π SYNOPSIS π
#include <radcli/radcli.h>
int rc_auth_proxy(
rc_handle *rh,
VALUE_PAIR *send,
VALUE_PAIR **received,
char *msg
);
π¦ DESCRIPTION π¦
So here's the vibe: this function builds an authentication request for proxying.
It builds an authentication request using the value_pairs in send and submits it to a server. That's the whole plot. β¨
But it's the proxy version, and that changes things, so listen up:
- it works for a proxy,
- it does not add an IP address,
- and it does not rely on the config file.
That's it. That's the energy. lowkey elegant tbh π«Ά
π PARAMETERS π
rhβ a handle to parsed configuration. your little config bestie.sendβ aVALUE_PAIRarray of values (e.g.,PW_USER_NAME). this is what you're sending out into the world πreceivedβ an allocated array of received values. where the goodies come back to you.msgβ must be an array ofPW_MAX_MSG_SIZEorNULL. it will contain the concatenation of anyPW_REPLY_MESSAGEreceived. so if the server has something to say, it says it here. π¬
not me reminding you thatmsgis either aPW_MAX_MSG_SIZEarray or literallyNULLβ those are the only two options, bestie, there is no third one. π
π§ STRUCTURES π§
avoid using this structure directly. use the rc_avpair_get_ functions. fr fr, this is the warning, don't skip it.
struct rc_value_pair {
char name[RC_NAME_LENGTH+1]; // attribute name if known.
uint64_t attribute; // attribute numeric value of type rc_attr_id including vendor; use VENDOR() and ATTRID() to separate.
rc_attr_type type; // attribute type.
uint32_t lvalue; // attribute value if type is PW_TYPE_INTEGER, PW_TYPE_DATE or PW_TYPE_IPADDR.
char strvalue[AUTH_STRING_LEN+1]; // contains attribute value in other cases.
struct rc_value_pair *next;
char pad[32]; // unused pad
};
and yes, the same structure shows up a second time in this section, word for word β same fields, same comments, same everything. so here it is again, because the manpage said it twice and I'm not about to edit the source material:
avoid using this structure directly. use the rc_avpair_get_ functions. π
struct rc_value_pair {
char name[RC_NAME_LENGTH+1]; // attribute name if known.
uint64_t attribute; // attribute numeric value of type rc_attr_id including vendor; use VENDOR() and ATTRID() to separate.
rc_attr_type type; // attribute type.
uint32_t lvalue; // attribute value if type is PW_TYPE_INTEGER, PW_TYPE_DATE or PW_TYPE_IPADDR.
char strvalue[AUTH_STRING_LEN+1]; // contains attribute value in other cases.
struct rc_value_pair *next;
char pad[32]; // unused pad
};
twinning. iconic behaviour honestly. π―
β RETURN VALUE β
here's the scoreboard, bestie:
- received value_pairs land in
receivedβ - messages from the server land in
msg(if it's non-NULL) β OK_RC(0) on success β slay πCHALLENGE_RC(3) when an Access-Challenge is received β okay so it's not a fail, it's a challenge, keep up π- negative on failure β that's your return value, and yeah, negative means it went bad π
π SEE ALSO π
the whole friend group, in alphabetical order, exactly as listed:
radcli.h(3), rc_aaa(3), rc_aaa_ctx(3), rc_aaa_ctx_free(3), rc_aaa_ctx_get_secret(3), rc_aaa_ctx_get_vector(3), rc_aaa_ctx_server(3), rc_acct(3), rc_acct_proxy(3), rc_add_config(3), rc_apply_config(3), rc_auth(3), rc_auth_proxy(3), rc_avpair_add(3), rc_avpair_assign(3), rc_avpair_copy(3), rc_avpair_free(3), rc_avpair_gen(3), rc_avpair_get(3), rc_avpair_get_attr(3), rc_avpair_get_in6(3), rc_avpair_get_raw(3), rc_avpair_get_uint32(3), rc_avpair_insert(3), rc_avpair_log(3), rc_avpair_new(3), rc_avpair_next(3), rc_avpair_parse(3), rc_avpair_remove(3), rc_avpair_tostr(3), rc_buildreq(3), rc_check(3), rc_check_tls(3), rc_conf_int(3), rc_conf_srv(3), rc_conf_str(3), rc_config_free(3), rc_config_init(3), rc_destroy(3), rc_dict_addattr(3), rc_dict_addval(3), rc_dict_addvend(3), rc_dict_findattr(3), rc_dict_findval(3), rc_dict_findvend(3), rc_dict_free(3), rc_dict_getattr(3), rc_dict_getval(3), rc_dict_getvend(3), rc_find_server_addr(3), rc_get_socket_type(3), rc_get_srcaddr(3), rc_getport(3), rc_mksid(3), rc_new(3), rc_openlog(3), rc_own_hostname(3), rc_read_config(3), rc_read_dictionary(3), rc_read_dictionary_from_buffer(3), rc_send_server(3), rc_setdebug(3), rc_test_config(3), rc_tls_fd(3)
and that's the whole page, bestie. you read all of it, you understood all of it, you're literally the main character of the radius client library now. π«Άβ¨
this little love letter to rc_auth_proxy came from radcli and it's dated 2023-01-29. take care of her, she's iconic. πππΈ
π¦ where this page comes from
- alpine 3.24: radcli-doc 1.4.0-r0 Β· BSD-2-Clause AND MIT
- epel 10: radcli-devel 1.5.2-3.el10_3 Β· BSD-2-Clause AND UMich-Merit AND HPND-Fenneberg-Livingston
- fedora 43: radcli-devel 1.4.0-5.fc43 Β· BSD-2-Clause AND UMich-Merit AND HPND-Fenneberg-Livingston
- fedora 43: radcli-devel 1.4.0-5.fc43 Β· BSD-2-Clause AND UMich-Merit AND HPND-Fenneberg-Livingston
- fedora 44: radcli-devel 1.4.0-6.fc44 Β· BSD-2-Clause AND UMich-Merit AND HPND-Fenneberg-Livingston
- fedora 44: radcli-devel 1.4.0-6.fc44 Β· BSD-2-Clause AND UMich-Merit AND HPND-Fenneberg-Livingston
- fedora 44: radcli-devel 1.5.2-3.fc44 Β· BSD-2-Clause AND UMich-Merit AND HPND-Fenneberg-Livingston
- fedora 44: radcli-devel 1.5.2-3.fc44 Β· BSD-2-Clause AND UMich-Merit AND HPND-Fenneberg-Livingston
- nixos 26.05: radcli 1.5.1
- nixpkgs unstable: radcli 1.5.2
- opensuse-leap 16.0: radcli-devel 1.4.0-bp160.1.11 Β· BSD-2-Clause AND MIT
- opensuse-leap 16.0: radcli-devel 1.4.0-bp160.1.11 Β· BSD-2-Clause AND MIT
- opensuse-leap 16.0: radcli-devel 1.4.0-bp160.1.11 Β· BSD-2-Clause AND MIT
- opensuse-leap 16.0: radcli-devel 1.4.0-bp160.1.11 Β· BSD-2-Clause AND MIT
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms