Piki

2.2K posts

Piki banner
Piki

Piki

@Piki_eth

full-time dev, part-time bag holder-

Katılım Eylül 2022
543 Takip Edilen1.2K Takipçiler
Piki
Piki@Piki_eth·
What happened to Drift doesn’t look like a normal “found a bug in the contract and drained funds” exploit. This looks way more like a long-game infiltration attack. The scary part? The attack may have started months before the exploit. Not with code. Not with wallets. Not with a multisig. With people. A group allegedly presented themselves as a legit trading firm. Built relationships. Met contributors in person. Asked smart questions. Acted normal. Even moved real capital into the ecosystem. That’s what made it dangerous. Because once trust is built, sending a repo, a tool, or an app no longer feels suspicious. It feels like routine work. And that’s where the real damage can start. Simple version of the attack flow: fake but believable identities months of trust-building normal-looking business relationship malicious repo / app / tooling shared contributor device gets compromised attacker reaches sensitive systems protocol and funds get hit So no, this is not “just a smart contract hack.” It’s a reminder that in crypto, the attack surface is not only onchain. It’s also: contributors laptops internal workflows integration processes signer environments every device that touches privileged access This is what a modern crypto attack looks like: patient, social, operational, and highly organized. The biggest lesson for every team: You can audit contracts for months, but if an attacker compromises the human layer first, they may never need to beat the code directly.
Drift@DriftProtocol

x.com/i/article/2040…

English
0
0
1
30
Piki
Piki@Piki_eth·
I like this dead chain
Piki tweet media
English
0
0
0
17
Piki
Piki@Piki_eth·
increasing the validator set to 250 is huge this opens the door for more operators to join if you're setting up a validator: don’t cluster in already saturated regions deploy in underrepresented geos better for decentralization better for the network long term @monad
Piki@Piki_eth

looking at Monad validator distribution: 200 validators across 26 countries and 56 cities that level of geographic spread this early is a strong signal there are still regional concentrations (normal for testnet) but the network is clearly expanding globally this is what healthy decentralization looks like in early stages #monad monad-scope-piki-nodes.com/validator-map

English
0
0
0
20
Piki
Piki@Piki_eth·
looking at Monad validator distribution: 200 validators across 26 countries and 56 cities that level of geographic spread this early is a strong signal there are still regional concentrations (normal for testnet) but the network is clearly expanding globally this is what healthy decentralization looks like in early stages #monad monad-scope-piki-nodes.com/validator-map
Piki tweet media
English
0
0
2
67
BlockPro
BlockPro@BlockPro_·
We’re excited to share that BlockPro is joining Monad Mainnet through Wave 2 of the Validator Delegation Program. The Monad Foundation has allocated 52,500,000 MON to our validator. Huge thanks to @monad_dev and @socialforging for the support and for this great news. We’ll also be sharing more good news for our stakers soon, including bonuses and contests, so stay tuned. If you’re a validator and still not participating in the Monad ecosystem, now is a great time to start. MIP-9 is currently active, proposing to expand the active validator set from 200 to 250. That creates a strong opportunity for more validators to enter the active set on Monad Mainnet. More details about the program: docs.monad.xyz/node-ops/valid…
BlockPro tweet media
English
10
0
22
969
brutaldevvin
brutaldevvin@brutaldevvin·
I started to run a Monad Testnet node about two weeks ago. The docs are solid and well-written, following them properly will get you to the end but there are some quirks and proper practices you need to follow so you can run a healthy node the right way. Let's start with glancing at the hardware reqs: Monad is fast and technically impressive. This kind of workload requires specific hardware, if you look at the well documented and tested specs, you'll see some strict requirements: •16-core CPU with 4.5GHz+ base clock •64GB RAM (32GB is technically possible but not recommended) •2x 2TB NVMe SSDs, separate drives, no RAID that CPU requirement alone filters out most dedicated servers. typical datacenter chips like an EPYC 7402 run at ~2.8GHz base, completely unusable here. the CPUs that actually work are mostly workstation-class: ryzen 7950x, ryzen 7950x3d, ryzen 9950x, epyc 4584px. Storage is where setups quietly might fail. You need two separate 2TB NVMe drives. one of them is used as a raw block device for trieDB, no filesystem, no RAID. Drive selection also matters more than expected. monad requires around 1M+ random write IOPS. Samsung 990 pro, 980 pro, and PM9A1 are safe. So, it's a good idea to confirm with server provider what you might be getting and ask for some leeway during the provision. I can confirm that Interserver carries the 990 Pro but you'll options will vary depending on the location you'll choose. I went with Interserver for a 7950x3d, 96GB RAM, and 2x2TB Samsung 990 pros. Regardless of provider, confirm upfront: IPMI access, separate NVMe drives, and a static IP. The install process itself is straightforward: apt repo, user setup, trieDB partitioning, firewall, metrics, configs, key generation, snapshot import, and service start but even with everything set up correctly, operational issues show up. Luck has it, after about 14 hours of clean runtime, my node crashed with “high qc too far ahead of block tree root.” this means the consensus layer fell too far behind and can’t recover. restarting won’t fix it, you need a full reset, fresh snapshot, and state sync. To deal with it, I built a watchdog that runs every few minutes, checks all services, and decides whether to restart or perform a full reset. It uses official Monad scripts for everything, workspace reset, snapshot restore, forkpoint download, with dual snapshot provider fallback, rate limiting, and alerting. It's also a good idea to ping your server against possible hardware end failures because something like a watchdog won't help you there. For this, you can either cook up your own Telegram bot, have an agent do it or just use something like UptimeRobot which is free and have their own app that will send you a notification if something is wrong. Without monitoring, your node can sit dead for hours. Total cost for a compliant setup for me was around $190/month. Yours can be cheaper or more expensive, depending on your provider and hardware specifications. I used Interserver, there are options out there, some more expensive, some cheaper. Make sure to smooth out the edges before you buy. The official docs will get you most of the way there, but the remaining gap is where real-world friction lives, provider quirks, hardware mismatches, and recovery flows. once you account for those, the setup becomes predictable instead of frustrating. Hope what I provided here helps. @monad_dev @socialforging
brutaldevvin tweet media
English
3
0
6
140
Piki
Piki@Piki_eth·
Inter has strong pricing for the specs, but their infrastructure is heavily US-based, and your nodes appear to be in the same region (NYC). Considering Monad testnet and mainnet already has 40+ nodes concentrated in NYC, it would be more beneficial to deploy in a less saturated or geographically unique location to strengthen network decentralization.
English
1
0
1
20
Piki
Piki@Piki_eth·
💰 Bounty secured. Another one. Reported a medium severity issue → payout received ✅ Shoutout to @code4rena & @0xIntuition for the smooth process 🤝
Piki tweet media
English
0
0
7
119
Piki
Piki@Piki_eth·
Monad Scope — UI Upgrade + New Features Live Just shipped a fresh update to Monad Scope 👇 👉 monad-scope-piki-nodes.com ✨ What’s new: 🎨 New UI redesign — cleaner, faster, more intuitive ⚡ Improved performance for real-time data tracking 🔗 Expanded indexed APIs for builders & developers 📊 Enhanced network insights (TPS, finality, validator stats) @monad @socialforging
English
1
0
3
140
Jackson (expansion arc)
Jackson (expansion arc)@socialforging·
Hardfork Complete! Mainnet is now on MONAD_NINE 100% of validators online, producing and joined the canonical chain.
Jackson (expansion arc)@socialforging

REMINDER: MAINNET HARDFORK (MONAD_NINE) - v.0.13.0 ⚠️ This version results in a hard fork Thursday 19th 14:30 GMT (2026-03-19 14:30GMT) ⚠️ Instructions: docs.monad.xyz/node-ops/upgra… Changelog: #v0130-monad_nine" target="_blank" rel="nofollow noopener">docs.monad.xyz/developer-esse… A reminder to all mainnet operators, if upgrading, all nodes are required to upgrade prior to the Hard Fork to ensure their node operations continue, nodes failing to upgrade in time will cease to function. This hard fork enacts MIP3, MIP4 and MIP5 alongside some improvements in RPC processes. (MIPs linked in thread)

English
15
9
93
3.7K