Skip to content

Conversation

jimmykarily
Copy link
Contributor

Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
since we implemented the getRawTPM which takes a different struct as
input and nobody needs to call getTPMDevice outside this package.

Signed-off-by: Dimitris Karakasilis <[email protected]>
Every request that is prone to replay attacks, should include a nonce.
We now send the nonce only on the "proof" request along with the proof
data.

Signed-off-by: Dimitris Karakasilis <[email protected]>
@jimmykarily jimmykarily force-pushed the 2988-remote-attestation branch from 10dfc4d to de0e2ba Compare September 17, 2025 11:16
Signed-off-by: Dimitris Karakasilis <[email protected]>
which is what the legacy getAttestationData() method was doing, which is
enough (the server only needs the public key not the whole parameters
struct)

Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
where no nonce is needed. Also use the full attestation parameters as
there are more fields reuquired to produce the challenge response

Signed-off-by: Dimitris Karakasilis <[email protected]>
@jimmykarily jimmykarily force-pushed the 2988-remote-attestation branch from a00e87c to 08bcc4f Compare September 19, 2025 11:17
Signed-off-by: Dimitris Karakasilis <[email protected]>
so that we can choose which PCR to include in the tpm Quote in the
kcrypt-challenger repo. The old version included all PCRs and didn't let
us choose while now we can:

https://github.com/google/go-attestation/blob/4f3c3b0fe5706286182530cd798be833ad0c4a74/attest/tpm.go#L325

Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
since they are internal implementation

Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
because swtpm is in the official repos:

https://launchpad.net/ubuntu/noble/+package/swtpm

Signed-off-by: Dimitris Karakasilis <[email protected]>
Signed-off-by: Dimitris Karakasilis <[email protected]>
jimmykarily added a commit to kairos-io/kcrypt-discovery-challenger that referenced this pull request Sep 24, 2025
to allow kcrypt-challenger to cleanup the NV storage (e.g. to reset the
local passphrase)

Signed-off-by: Dimitris Karakasilis <[email protected]>
because enrollement happens in a later step

Signed-off-by: Dimitris Karakasilis <[email protected]>
jimmykarily added a commit to kairos-io/kcrypt-discovery-challenger that referenced this pull request Sep 30, 2025
because during initramfs stage, there is no persistent storage available
to write a file to (except maybe the encrypted partitions which is what
we are trying to decrypt in the first place).

Signed-off-by: Dimitris Karakasilis <[email protected]>
@jimmykarily jimmykarily force-pushed the 2988-remote-attestation branch from 662b294 to 229b783 Compare September 30, 2025 07:35
because we don't need to verify the AK (we verify the EK instead) so we
don't have the problem of storing the AK permanently, thus fixing the error:

TPM_RC_SIZE (parameter 2): structure is the wrong size

when trying to store the whole AK into NV index

Signed-off-by: Dimitris Karakasilis <[email protected]>
@jimmykarily jimmykarily force-pushed the 2988-remote-attestation branch from 229b783 to 96deb0f Compare September 30, 2025 10:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: In Progress 🏃

Development

Successfully merging this pull request may close these issues.

1 participant