Delayed due to the fact that DevOps subgovernments still in the jury selection stage. While you wait you can participate in this contest: Free TON DevOps Jury Contest
you’re very welcome
Hi folks!
After contest discussion on the DevOps subgov weekly call (join as if you still not in the group - > https://t.me/freetondevops) we have two suggestions:
Extend contest rewards pool but add vesting in equal parts over a 12-month period with monthly payouts. To receive all 12 distributions, the solution should be supported for 1 year
The scripts should be updated within a reasonable timeframe - 1 month maximum, and according to rust node changes
Major GitHub Issues should be resolved within a reasonable timeframe - 1 month maximum.
Winners receive 30% of their rewards when contest results are evaluated. I propose this reward distribution table:
1st prize — 40,000 Tons (12 000 + 28 000 by vesting)
2nd prize — 30,000 Tons (9 000 + 21 000 by vesting)
3rd prize — 20,000 Tons (6 000 + 14 000 by vesting)
4th prize — 5,000 Tons
5th prize — 3,000 Tons
Next 10 runners up — 1,000 Tons each
Add minimum of completed tasks (task 1, and task 4 or 5) (without them submission can be rejected) to simplify evaluation creteria and voting. Otherwise juries will potentially compare different types of works (some will be focus on monitoring/alerting, some on installation/validation scripts) which will hard and incorrect to compare (because of different purpose)
Just double-checked, according to the DevOps SG proposal (voted and passed), the reward for this contest is supposed to be 200’000, not 103’000. Compared to the first C++ DevOps node contest, minimum number of tasks in this one is mandated (i. e. more work required). For the C++ node DevOps node contest, the rough equivalent of task 3 has been done by TonLabs (scripts - validator_msig.sh, validator_depool.sh), and task 4 hasn’t been there at all (would’ve been something like automating staking through depool using liteclient). Other than that, requiring MVP (minimum number of tasks completed) and vesting for the first 3 places is a good idea.
Monitoring, Alerting, Scripts and Tooling for RUST implementation of Free TON Validator
Type
Contest
Dates
Submissions - Starts: 18.01.2021 at 00:00 UTC, Ends: 31.01.2021 at 23:59 UTC
Voting - Starts: 01.02.2021 at 00:00 UTC, Ends: 14.02.2021 at 23:59 UTC
Motivation
Validators need tools and automation to insure efficient scalable and secure operations to support Free TON blockchain
General requirements:
Documentation should be provided for each tool
Provided tools should operate exactly according to its documentation
Motivation and goal for each tool should be described
All software must be released with licences supported by Free Software Foundation
DevOps Skills required:
Linux
Unix Shell scripting, javascript
Infrastructure as a code
docker/docker-compose/kubernetes
Desirable: Ansible, logstash, Prometheus/Grafana
Tasks:
Enhanced node deployment infrastructure
Need to develop enhanced deployment infrastructure for Rust validator nodes based on modern tools and techniques (docker/docker-compose, Ansible, Terraform, etc.).
Based on infrastructure as a code. Take into account deployment from scratch, upgrade, cleanup scenarios on various set of Linux distros as well as other OS (e.g. MacOS).
Intellectual log dashboard with analytics:
Need to collect log files into centralised database (e.g. logstash).
Make base analytics with some predefined patterns which can simplify failure analysis and monitoring.
Extended metrics for node monitoring
Analyse available node/validator/elector metrics and define a subset useful for further analytics and monitoring tasks.
Develop visualisation concept to simplify the analysis.
Develop alert rules for daily monitoring
Automate staking through depool (and/or direct staking to the elector) based on tonos-cli or SDK (use TON-SDK core library or any existing binding)
Write validator script for elections using tonos-cli/SDK
Use tonos-cli/SDK to run elector get-methods.
Use tonos-cli/SDK to query global configuration parameters.
Use tonos-cli/SDK to prepare, sign and send validator query to depool/elector smart contract.
Automate confirmation of multisig transactions.
Improve custodian notification mechanism for multisig transactions: notify about submitted transactions via user-friendly mechanism (e-mail, sms, etc.).
Implement tools for staking monitoring and alerting
Automate staking through depool (and/or direct staking to the elector) using R-Node Console
Write a script to manage overall staking procedure (prepare elector request, sign, send, manage bids, etc.)
Automate confirmation of multisig transactions
Implement tools for staking monitoring and alerting
Evaluation criteria and Winning conditions:
Number of completed tasks. Tasks 1 (at least for one OS) and optionally 4 or 5 are necessary. Without them submission will be rejected
Correctness of operations of the provided tool
Quality of code and description
Work as advertised, perform its stated functions
The Jury
The Jury will be selected by a sub-governance which will support this contest.
Jury rewards:
An amount equal to 5% of the sum total of all total tokens actually awarded to winners of this contest will be divided equally between all jurors who vote and provide feedback. Both voting and feedback are mandatory in order to collect this reward.
Winners:
The winners of all 5 tasks will recieve the following:
Winners receive 40% of their rewards when contest results are evaluated.
Another 60% will be vested over a 12-month period with monthly payouts.
Vesting details
Vesting is devided into 12 equal parts. The first vesting payout is paid one month after the end of the contest. To receive all 12 distributions, the solution should be supported at least for 1 year. Scripts should be updated within a reasonable timeframe - 1 month maximum, and according to rust node changes. Major GitHub Issues should be resolved within a reasonable timeframe - 1 month maximum.
Validator node doesn’t have any extra ports exposed
Every deployment can be scaled independently and whenever is required
Very flexible in controlling costs - Validator, Controller and Logstash are deployed via Docker (backed-up with docker-compose) either to bare-metal machine or VM. At the same time monitoring can be either custom solution or one of SaaS solutions with pay-as-you-go subscriptions. The same applies for message-queue (either custom deployment or SaaS).
Pub/Sub layer provides good abstraction and allows to inject many type of notifications and ways to control validator(s), including safe for the validator web interfaces.
It is easy to integrate any kind of alerting and automatic response to those alerts.
Has following features:
Deploy, execute and monitor validator node(s) using infrastructure-as-a-code approach. So, sharing setup is seamless and easy across team members.
Fully automatic participation in elections (direct or depool), including automatic confirmations for custodians.
Can participate in elections with check of potential win in the elections. So, for example, if a node has small chance of being elected (stake that can be made is too small to be among participating nodes), then election can be skipped. All thresholds are configurable.
Reporting telemetry via logstash to any configured ELK stack service. Can be use with such SaaS as Bonsai.io, no need to deploy own solution (but for sure possible).
All scripts are in the active development stage and will be updated regularly. Feel free to contact as on github thought issues or directly on Telegram. Scripts done by validators for validators.
