tac_plus do_chpass vendors compatibility #229
Unanswered
OlexanderH
asked this question in
Q&A
Replies: 1 comment
-
|
Hi Olexander, I usually don't backport from tac_plus-ng to tac_plus, however this one looks trivial. Adding a "chpass" member to the session struct and changing the two send_authen_reply calls to what -ng does seems sufficient. session->ctx->aaa_realm->chpass isn't suitable for this check, that one just allows/denies chpass on a per-realm basis. The dedicated session->chpass from -ng sets GETPASS/GETDATA based on context (ASCII login vs. CHPASS). I'll push a patch. Thanks, Marc |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
In the tac_plus package, I noticed that when changing the password, when prompted for "New password: " in the TAC_PLUS_AUTHEN_STATUS_GETPASS mode, Juniper provides the very first password entered by the user, it does not ask the user for "New password: " and "Retype new password: " at all.
When I change the mode to TAC_PLUS_AUTHEN_STATUS_GETDATA, the dialog runs normally.
tac_plus/authen.c
found that in the tac_plus-ng package this situation is fixed by checking session->chpass ? TAC_PLUS_AUTHEN_STATUS_GETPASS : TAC_PLUS_AUTHEN_STATUS_GETDATA
tac_plus-ng/authen.c
so in tac_plus I fixed it this way and it worked
but I couldn't understand what session->ctx->aaa_realm->chpass should contain, logically it should be "true" when do_chpass, but then TAC_PLUS_AUTHEN_STATUS_GETPASS would executed, maybe it's better to leave only TAC_PLUS_AUTHEN_STATUS_GETDATA in do_chpass for compatibility reason?
Tested on Cisco, Huawei routers - works in both versions (GETDATA and GETPASS), Juniper only with GETDATA.
Beta Was this translation helpful? Give feedback.
All reactions