<aside> 🛠 This section is about the issues tackled by the development team for this week, and the changes which won't be included in the next release right away.
</aside>
Last Friday, we released v100103 as a hotfix and arbitrarily restricted our miners to take blocks on the 12 second hour, but the problem of 0tx (1tx to be exact) blocks was intermittently reproduced.
According to the investigations so far, this problem is presumed to occur in Libplanet 0.27 or later, and it is a bug that puts you in a situation where you cannot receive new Tx after starting the node in at least an hour and at most 4 hours.
Currently, to solve this problem, the hotfix to downgrade Libplanet to 0.26.3 is first applied to the nodes of our cluster including the minor, and will be distributed next Monday along with the other minor changes.
Also, in order to solve this problem in the next version of Libplanet (0.28+), the engine development team is analyzing the related logs and exploring the cause.
A few RPC nodes we're running for fast play have been spotted processing tens of petabytes of requests and responses. There is a possibility that it is inter-node communication, but there is also a possibility that it may be spamming or other attacks on external services provided by gRPC or GraphQL, so we are investigating the matter in detail.
We apologize for the continuing delay. As with last week, we're focused on fixing the miner's 1tx issue, and we'll start public testing as soon as that issue is resolved.
Due to the prevalence of laptops equipped with Apple Silicon, also known as M1, there were internal requests to support the arm64v8 architecture even for the Docker image of the 9c headless we provide. In addition to that, as the demand for Graviton instances has recently increased for optimizing AWS operating infrastructure, from v100110, a build has been added to provide an arm64v8 image as well.