credential: clear expired c->credential, unify secret clearing
PATCH v3 1/1 credential: clear expired c->credential, unify secret clearing
When a struct credential expires, credential_fill() clears c->password so that clients don't try to use it later. However, a struct cred that uses an alternate authtype won't have a password, but might have a credential stored in c->credential.
This is a problem, for example, when an OAuth2 bearer token is used. In the system I'm using, the OAuth2 configuration generates and caches a bearer token that is valid for an hour. After the token expires, git needs to call back into the credential helper to use a stored refresh token to get a new bearer token. But if c->credential is still non-NULL, git will instead try to use the expired token and fail with an error:
fatal: Authentication failed for 'https://<oauth2-enabled-server>/repository'
And on the server:
[auth_openidc:error] [client <ip>:34012] oidc_proto_validate_exp: "exp" validation failure (1717522989): JWT expired 224 seconds ago
Fix this by clearing both c->password and c->credential for an expired struct credential. While we're at it, use credential_clear_secrets() wherever both c->password and c->credential are being cleared.
Update comments in credential.h to mention the new struct fields.
---
v3: I reverted the behavior change to credential_reject() and just unified everything to use credential_clear_secrets() instead. We can rework credential_reject() in a later change if we decide to. So the only behavior change now should be the expiration case in credential_fill()
I also updated some of the comments in credential.h to mention the new struct fields.
Thanks for your patience with this series, everyone!
credential.c | 16 ++++++++++------ credential.h | 34 ++++++++++++++++++---------------- 2 files changed, 28 insertions(+), 22 deletions(-)
credential.c +10 -6
@@ -20,12 +20,11 @@ void credential_init(struct credential *c)void credential_clear (struct credential *c) { +credential_clear_secrets (c);free (c->protocol);free (c->host);free (c->path);free (c->username); -free (c->password); -free (c->credential);free (c->oauth_refresh_token);free (c->authtype);string_list_clear (&c->helpers,0 );
@@ -479,9 +478,15 @@ void credential_fill(struct credential *c, int all_capabilities)for (i =0 ; i < c->helpers.nr; i++) {credential_do (c, c->helpers.items[i].string,"get" ); +if (c->password_expiry_utc <time (NULL )) { -/* Discard expired password */ -FREE_AND_NULL (c->password); +/* +* Don't use credential_clear() here: callers such as +* cmd_credential() expect to still be able to call +* credential_write() on a struct credential whose +* secrets have expired. +*/ +credential_clear_secrets (c);/* Reset expiry to maintain consistency */ c->password_expiry_utc =TIME_MAX ; }
@@ -528,9 +533,8 @@ void credential_reject(struct credential *c)for (i =0 ; i < c->helpers.nr; i++)credential_do (c, c->helpers.items[i].string,"erase" ); +credential_clear_secrets (c);FREE_AND_NULL (c->username); -FREE_AND_NULL (c->password); -FREE_AND_NULL (c->credential);FREE_AND_NULL (c->oauth_refresh_token); c->password_expiry_utc =TIME_MAX ; c->approved =0 ;
credential.h +18 -16
@@ -5,8 +5,8 @@#include "strvec.h" /** -* The credentials API provides an abstracted way of gathering username and -* password credentials from the user. +* The credentials API provides an abstracted way of gathering +* authentication credentials from the user. * * Typical setup * -------------
@@ -116,11 +116,12 @@ struct credential_capability { };/** -* This struct represents a single username/password combination -* along with any associated context. All string fields should be -* heap-allocated (or NULL if they are not known or not applicable). -* The meaning of the individual context fields is the same as -* their counterparts in the helper protocol. +* This struct represents a single login credential (typically a +* username/password combination) along with any associated +* context. All string fields should be heap-allocated (or NULL if +* they are not known or not applicable). The meaning of the +* individual context fields is the same as their counterparts in +* the helper protocol. * * This struct should always be initialized with `CREDENTIAL_INIT` or * `credential_init`.
@@ -207,11 +208,12 @@ void credential_clear(struct credential *);/** * Instruct the credential subsystem to fill the username and -* password fields of the passed credential struct by first -* consulting helpers, then asking the user. After this function -* returns, the username and password fields of the credential are -* guaranteed to be non-NULL. If an error occurs, the function will -* die(). +* password (or authtype and credential) fields of the passed +* credential struct by first consulting helpers, then asking the +* user. After this function returns, either the username and +* password fields or the credential field of the credential are +* guaranteed to be non-NULL. If an error occurs, the function +* will die(). * * If all_capabilities is set, this is an internal user that is prepared * to deal with all known capabilities, and we should advertise that fact.
@@ -232,10 +234,10 @@ void credential_approve(struct credential *);* have been rejected. This will cause the credential subsystem to * notify any helpers of the rejection (which allows them, for * example, to purge the invalid credentials from storage). It -* will also free() the username and password fields of the -* credential and set them to NULL (readying the credential for -* another call to `credential_fill`). Any errors from helpers are -* ignored. +* will also free() the username, password, and credential fields +* of the credential and set them to NULL (readying the credential +* for another call to `credential_fill`). Any errors from helpers +* are ignored. */ void credential_reject (struct credential *);
Signature · 1 line
2.45.2.409.g7b0defb391
Quoted 4 lines · Aaron Plattner
v3: I reverted the behavior change to
credential_reject()and just unified everything to usecredential_clear_secrets()instead. We can reworkcredential_reject()in a later change if we decide to. So the only behavior change now should be the expiration case incredential_fill()
Looks good.
Quoted 3 lines
I also updated some of the comments in
credential.hto mention the new struct fields.
Thanks for paying attention to such details. I very much like these updated comments.
Will queue.
Quoted 7 lines · Aaron Plattner
v3: I reverted the behavior change to
credential_reject()and just unified everything to usecredential_clear_secrets()instead. We can reworkcredential_reject()in a later change if we decide to. So the only behavior change now should be the expiration case incredential_fill()I also updated some of the comments in
credential.hto mention the new struct fields.
Thanks, this one looks great to me.
Quoted 1 line
Thanks for your patience with this series, everyone!
Likewise!
-Peff