pki-bot
commented
4 years ago Comment from jmagne (@jmagne) at 2016-05-23 23:28:44
One more refinenemt:
TKS right now when we try to do a symettric key derivation by encryption, if this fails on the hsm, we catch the exception and try a fallback (which works). For efficiency reason, the reviewer would like to ahve this feature conrolled by a config parameter to avoid the initial failed try, (if we know it will fail on a certain hsm).
This issue was migrated from Pagure Issue #2337. Originally filed by jmagne (@jmagne) on 2016-05-23 23:26:33:
Ticket 801 resulted in moving much of the jni symkey functionality into either JSS itself or Java code within the TKS subsystem.
The review for that resulted in some potential refinements, which will be listed in this ticket.
JSS: Provide a "unwrapSymmetricTemporary" variant to the new call I created called "unwrapSymmetricPerm". This way the user will know which one to use more easily. Currently, the stock "unwrapSymmetric" call assumes a temporary key.
JSS: Make sure the license text on newly created files is up to date.
JSS and TKS: Make sure the code works on luna hsm. Adjust either or both pieces if needed.
TKS: Improve the efficiency of the computeSessionKey_SCP01, call right now it kind of generates all 3 keys needed and returns only one, which is called 3 time . Just call it once and return all 3 keys at once.
TKS: Rework the methods that deal with the shared secret key. Right now the use is a bit confusing.