![\"2Miners](\"https:\/\/2miners.com\/blog\/wp-content\/uploads\/2026\/09\/2miners-august-2026-work-progress-report.jpg\")
<\/p>\nAugust belonged to Ravencoin. A consensus flaw in KAWPOW header validation let an attacker forge blocks for almost nothing, and the network spent a week deciding which chain was real. We shipped the emergency patch that the Ravencoin project later published as v4.8.0, and together with the other pools we rebuilt the chain from the last clean block. Pearl went through a hard fork at block 99,000, Kaspa cut its block reward again, and we updated our Ergo and Quai nodes.<\/p>\n
<\/p>\n
<\/p>\n
This was the most serious incident on any of our pools this year, so here is the whole thing in order.<\/p>\n
The flaw sits in Ravencoin’s KAWPOW block header validation. The nHeight<\/strong> field inside the KAWPOW header was never checked against the block’s actual position in the chain. An attacker who manipulates that field can get a vulnerable node to accept a supplied mix hash without ever doing the real ProgPoW work – no DAG, no memory-hard hashing, no GPU. Forging a block that way is orders of magnitude cheaper than mining one honestly.<\/p>\n The bug had two ugly side effects beyond the forged blocks themselves. Affected nodes could crash on restart, and nodes trying to sync ran into broken header sequences and stalled.<\/p>\n Scanning a 2,089-block window from 4,489,527<\/strong> to 4,491,615<\/strong>, we identified 96 forged blocks<\/strong>. Everything confirmed after 4,487,775 on the exploited branch was at risk, which is roughly three days of chain history.<\/p>\n 2Miners and the other pools started building a competing chain from block 4,487,776 that excludes the branch carrying the invalid blocks. Together we hold the majority of Ravencoin’s hash rate, so the clean chain grew faster than the forged one and overtook it. It is now thousands of blocks ahead, peers have reconnected, and the split is settled in favour of the honest chain.<\/p>\n Every RVN mined before August 7 was never in question. Our RVN pool<\/a> and RVN SOLO<\/a> have been running on the clean chain throughout.<\/p>\n If you mine RVN with us and do not run your own node, there is nothing to change on your side. KAWPOW mining and payouts continue as normal.<\/p>\n Pearl activated a hard fork at mainnet block 99,000<\/strong> on August 11<\/strong>. The change salts the noise seed used by PearlHash, which means shares produced by pre-fork mining software are simply invalid after the fork block.<\/p>\n Please pay attention!<\/strong> This failure mode is quiet. An outdated miner keeps reporting a healthy hashrate on your screen while nothing it submits gets credited. If your PRL earnings flatlined in mid-August, this is why.<\/p>\n Mining software with fork support:<\/p>\n On the node side the fork ships in Pearl 1.4.1<\/strong>, and that is what our node runs. Earlier in the month we also took Pearl 1.3.1<\/strong> (commit b6d7e5e), a follow-up to the 1.3.0 rank-penalty soft fork released on August 5. Our pools were ready before the fork block and kept producing. Mine on PRL POOL<\/a> and PRL SOLO<\/a>, and check profitability on 2CryptoCalc<\/a>.<\/p>\n As defined by the Kaspa Tokenomics<\/a>, the Crescendo Hard Fork, and the official emission schedule, Kaspa underwent another scheduled block reward reduction in August.<\/p>\n Please pay attention!<\/strong> Kaspa mining profitability has changed. We have already updated 2CryptoCalc<\/a> to reflect the new reward values. Mining continues on the KAS pool<\/a> and KAS SOLO<\/a>.<\/p>\n Released August 18. Most of this release is API hardening and test work, but three items matter if you run an Ergo node or care about block propagation:<\/p>\n The rest is housekeeping: duplicate IDs are now prevented in the ordered transaction pool, several REST and wallet routes validate their parameters properly, and the testnet version 2 activation height was corrected. Autolykos mining is unchanged. Mine on the ERG pool<\/a> and ERG SOLO<\/a>.<\/p>\n Tagged August 11. The headline for miners is that go-quai now carries a built-in stratum proxy for solo mining<\/strong>, and the release collects a long run of work on it:<\/p>\n Nothing changes in your miner configuration on our side. Mine on QUAI KawPow<\/a>, QUAI KawPow SOLO<\/a>, QUAI SHA-256<\/a>, and QUAI SHA-256 SOLO<\/a>.<\/p>\n Follow us on X (Twitter)<\/a> and join our Telegram chat<\/a> to stay informed about new developments, miner alerts, and network upgrades.<\/p>\n","protected":false},"excerpt":{"rendered":" August belonged to Ravencoin. A consensus flaw in KAWPOW header validation let an attacker forge blocks for almost nothing, and the network spent a week deciding which chain was real. We shipped the emergency patch that the Ravencoin project later published as v4.8.0, and together with the other pools we rebuilt the chain from the […]<\/p>\n","protected":false},"author":4,"featured_media":5384,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-5380","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"yoast_head":"\nTimeline<\/h3>\n
\n
How big the damage was<\/h3>\n
The recovery chain<\/h3>\n
What you need to do<\/h3>\n
\n
Pearl (PRL) Salted Noise-Seed Hard Fork at Block 99,000<\/h2>\n
\n
Kaspa (KAS) Block Reward Reduction<\/h2>\n
\n
Node Updates<\/h2>\n
Ergo – 6.0.4<\/h3>\n
\n
Quai – v0.55.1<\/h3>\n
\n
Stay Updated<\/h2>\n