Submitted by dhank, also found by KlosMitSoss
Safe cannot reduce  expirationPeriod to a newExpirationPeriod when
where id is the hash of updateExpirationPeriod() and timestamp[id] is the timestamp when the id can be executed.
Safe should be able to update the expirationPeriod to any values >= MIN_DELAY by scheduling the updateExpirationPeriod() and later execute from timelock when the operation is ready (before the expiry).
But the protocol has overlooked the situation and added an reduntant  check inside _afterCall() which is executed at the end of _execute().
Here the isOperationReady(id) will be executed with the newExpirationPeriod.
code
There it is checking whether the currentTimestamp is less than the timestamp + updated EpirationPeriod instead of the actual expirationPeriod.
forge test match-test testDUpdateExpirationPeriodRevert -vvv
Updating to the new expirationPeriod  will revert in this case.
This can affect the protocols core design features.
ElliotFriedman (Kleidi) confirmed, but disagreed with severity and commented:
Alex the Entreprenerd (judge) commented:
Alex the Entreprenerd (judge) decreased severity to Low/Non-Critical and commented:
Alex the Entreprenerd (judge) commented:
ElliotFriedman (Kleidi) commented:
Alex the Entreprenerd (judge) increased severity to Medium and commented:
For this audit, 11 reports were submitted by wardens detailing low risk and non-critical issues. The report highlighted below by 0xAkira received the top score from the judge.
The following wardens also submitted reports: KlosMitSoss, Sathish9098, 0x37, Rhaydden, Guardians, 0xSecuri, Sparrow, DemoreX, aariiif, and ZanyBonzy.
