-
Notifications
You must be signed in to change notification settings - Fork 811
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Bug]: ClientHello's status_request extension handling issue #7275
Comments
Hi @SmallTown123 , This is Anthony Hu again. I helped you on your previous ticket. For this one, I will need some more time to confer with my colleagues. Please stay tuned. Warm regards, Anthony |
Hi @SmallTown123 , I've assigned @ejohnstown to have a look at this issue. Please stay tuned. |
Can you send a Wireshark capture of this happening? The code is expecting there to be a length of the responder_id_list right after the CertificateStatusType value. I modified the client side code to put only the CertificateStatusType value in the extension, and I'm getting the error I expect. |
This is a wireshark screenshot of our test data. status_request is supposed to be 5, but when its length is changed to 1, the last 4 bytes 00 00 00 00 are parsed as server_name without affecting the normal parsing of subsequent extensions. But should the server respond with an Alert in this case? |
It should and as far as I can tell it does. First, I tested this running TLSv1.2 and I get the alert. I modified the function Today I'm running with TLSv1.3 with the master branch of our repository. I use the same modification to I switched back to v5.5.1, and I'm getting the same result. The server is sending an alert. I configured wolfSSL with the option |
Hi, @ejohnstown: I configured wolfSSL with the option --enable-ecc --enable-dsa --enable-aesccm --enable-curve25519 --enable-tls13. In fact, the content of the status_request extension should have been Not sure if your test data is consistent with that situation? |
I should have asked for your configuration first. Given your configuration, that extension isn't supported and it is skipped, not even looked at. Add |
And I'm actually shortening the extension rather than lying about the length in my test. I see the difference now. I'll take another look at this. |
I recreated your failure mode. Looking at a solution. |
Thanks for your reply, I was wondering if it would be possible to parse the extension while considering a safety check on the extension length. |
Im waiting too .-) |
@anhu or @ejohnstown any update on this? |
Hi @SmallTown123, This PR #7602 should add the required checks that you brought up. |
Contact Details
small_town_123@163.com
Version
5.5.1
Description
The wolfssl server appears to be having some issues with the status_request extension of the Clienthello message, specifically:
When the length of this extension is shorter than normal, specifically 1, and the CertificateStatusType value is 0x01, i.e. OSCP, a normal response still exists from the server.
The structure of the status_request extension item as normalized in the RFC6066 document is as follows:
struct {
CertificateStatusType status_type.
select (status_type) {
case ocsp: OCSPStatusRequest.
} request;
} CertificateStatusRequest;
enum { ocsp(1), (255) } CertificateStatusType;
struct {
ResponderID responder_id_list<0..2^16-1>;
Extensions request_extensions;
} OCSPStatusRequest;
opaque ResponderID<1..2^16-1>;
opaque Extensions<0..2^16-1>;
If the CertificateStatusType is OSCP, the length of the extension must be greater than 1.
Therefore, when the above situation occurs, the server should return an Alert response instead of a normal response.
Reproduction steps
No response
Relevant log output
No response
The text was updated successfully, but these errors were encountered: