The vulnerability arises from Zebra's P2SH sigop counter using a pure-Rust code path that short-circuits on disabled opcodes, returning a partial sigop count of zero for any sigops following the disabled opcode. In contrast, the reference implementation zcashd counts through disabled opcodes during static sigop analysis. This discrepancy leads to consensus divergence when the MAX_BLOCK_SIGOPS threshold (20,000) is crossed on one side but not the other. An attacker can exploit this by broadcasting transactions spending P2SH outputs with redeem scripts containing disabled opcodes followed by multiple OP_CHECKMULTISIG opcodes, causing Zebra to undercount sigops and accept blocks that zcashd rejects. The vulnerability affects all default configurations of zebrad up to v4.4.1 on any network shared with zcashd nodes. The issue is fixed in Zebra 4.4.2 by changing the P2SH sigop counting to use the C++ FFI path consistent with legacy sigop counting.

This vulnerability can cause a chain split between Zebra and zcashd validators on networks where both participate. Zebra nodes may accept blocks containing transactions with malicious redeem scripts that zcashd nodes reject due to exceeding the sigop limit. Approximately 30% of the current network hashrate is estimated to be Zebra miners, so this can lead to significant network divergence. The attacker does not require mining capability or special privileges; only the ability to broadcast crafted transactions. The cost to the attacker is limited to transaction fees. This undermines network consensus and stability.

A patch is available in Zebra version 4.4.2 that fixes the P2SH sigop counting to use the C++ FFI path consistent with the reference implementation. There is no configuration-level workaround. Operators should upgrade to Zebra 4.4.2 or later as soon as possible to prevent chain splits and consensus divergence.