Blockops is a platform that makes it extremely easy for builders, solo-stakers and node-operators to deploy, maintain and manage their nodes ANYWHERE with complete observability.
This is a proposal to extend Blockops Support for Lido CSM and integrating OBOL and SSV DVT for seamless Node Setup, we currently have an integration for Obol Network in our beta version and will like to integrate Lido CSM and SDVTM for Obol and SSV.
Key Features Blockops currently offers:
One Click Node Deployment ANYWHERE (only GCP is active in our Beta version)
Node Fleet Management support (client upgrades, upscaling and downscaling of nodes with NO Slashing and Zero Downtime)
Customizable Alerts to multiple notification channels
Upstream and Downstream Observability with Telescope: this allows operators to gain full visibility into the performance of their nodes and also allows DVT gains full observability into the performance of the node operators and validators running their client
By Tier-0 I mean the baseline integration that focuses on making it easy and seamless for NO to setup their EL, CL on any infra. and also that other future functionalities and features will be based on
Also I see some inaccuracies in references to different modules (CSM and SDVTM), I’m happy to chat with you to help to make the proposal a bit more clear!
For clarity, this is basically a “Validators as a Service” proposal right (i.e. the infrastructure (nodes+software) would be run by Blockops)? In this case you (currently) utilize GCP (a public cloud provider) for the infrastructure, but the management of the node software would be done by you?
TL;DR: No, The nodes will be completely managed by the user on their own environment
Additional Context:
The user has the autonomy to choose how and where to run their nodes: We have the Self Manage Node and Managed Node Feature
Self Managed Node: the infrastructure will be run and setup by the user on their env (anywhere: GCP, AWS or Bare Metal), they will be able to create, manage and monitor their nodes seamlessly from our platform, ideal for solo-stakers and validator runners
Managed: We had the feature for managed nodes for appchain bootstraping but from the feedback from our customers, they prefer to be in control of their nodes, they just want to abstract away the complexities of creating and managing it, and also because some of our users had cloud credits so they just want to make use of it without dealing with the technicalities. This is outside of the scope of our proposal, its a feature that had already existed on our platform
Can you be more specific about what the software actually is, where it sits, and how it enables this? E.g. if I want to use the proposed software to run nodes on a bare metal server I control, what would that look like? Where does this software sit? Is this an application that someone runs locally on their own machine? Is it open source (or source available)? Is this a web-app that will run on blockops’ infra? So then a user would upload details about their servers (and e.g. private keys for validators)?
I see in the proposal the examples are basically attaching services files from the cloud provider, so the software is basically a frontend to an IaC approach to manage remote nodes? e.g. if a user wants to update a CL node, that’s a part of the Lido Simple DVT package they’ve decided to use (how would that even work, would it be separate packages for different CL+EL combos, or would the user need to set up every node separate and then somehow connect them?), how would that work?
Blockops is a Node Setup and Management Platform, Our focus is Seamless NodeOps f& Observability for NO & Validator runners…it currently sits on our infrastructure
Blockops utilizes cloud access keys e.g services account file for GCP (with least privilege) for us to set up and manage their node.
Users upload a service account key file from a cloud provider, to authenticate and authorize node setup. (We will update our naming to be more descriptive)
User is in full control of Blockops access to their nodes and can revoke at any time
The service account access allows the user:
- Update dvt client versions with zero downtime
- Scale up and scale down their nodes with Zero Downtime & Avoid Slashing
Our boostraped node deployment has vault service deployed alongside and its able to integrate with vault for secret management for managing users secrets and credentials. We do not control or have access to Private Key being generated by the user, we only provide direction and guidance for users to do that seamlessly
The source code is available upon request, we have code audit and other compliance audits ready to be done as part of our roadmap before launch
We have two categories under Node Management: Node Deployment and Appchain Deployment (or Node Fleet Deployment), Node deployment deploys just one single instance of a node type for the network, while appchain deployment, deploys the entire stack needed to get a network up and running
In the case of Lido SDVTM taking Obol as an example:
Appchain: Users will be able to configure the deployment of their EL or CL nodes, as well as MEVBoost and other components needed to standup and Obol DVT Node, then blockops will deploy to users environment and displays all the information they need to access it
For Bare Metal Servers, our ssh module will allows users to connect to their server via ssh and they can configure a node deployment remotely from it
Hi @haroldsphinx and thank you for this application!
Regrettably, the LEGO council has prioritized other RFP applicants for the SDVTM integration as more promising. But let’s stay in touch, maybe we can revisit it in the Q2.
Hi Blockops! I would like to inform you that the LEGO committee voted on the CSM integration application, and decided to not proceed with the proposal at this time, as it wanted to focus on own-hardware based integrations for the time being, in spirit of community staking.
As the scope of the CSM increases and its target share is increased, it may make sense to also have integrations that leverage infrastructure as code for robust node deployment to cloud or even bare metal, and at that point the integration can be re-considered.
Would it matter to the committee that we’re already working on converting our Node Deployment feature that achieves all we proposed as an electron app that can be installed on Linux and MacOS?, so this way users can decide to run it on their own hardware or use our PaaS on portal.blockops.network.
We can still meet up with the proposed deadline if re-considered, otherwise, we’re happy to keep building and hope for another opportunity.
