The second phase of the “Agoric” Incentivized testnet has begun. In this article, I will briefly describe what developers want from testnet users and what kind of difficulties our team faced.
Unlike the first phase, the second one turned out to be much more difficult, but much more interesting. The node and Prometheus have been updated.
Let’s start with the update! The Agoric-8 update, which was planned earlier, didn’t create much difficulties and the node was successfully updated in a short time. Immediately after updating the node, we started to work on Prometheus, but unexpectedly the developers rolled out a new, unplanned update Agoric-9. Our team spent about 7 hours installing the new updates. Finally, the node has been updated to the currect version v0.25.2, but the blocks were frozen for some reasons. Something went wrong! The team contacted the developers and we came to the conclusion that the updated node is still using the old Agoric-7 network. The team decided to roll the node back using back-up function and start updating the node again, step by step, using Agoric-8 and Agoric-9 updates. After, we re-checked the network again. It has been verified that the node is working properly now. The next step was -”getting out of jail”. After that, one of our team members with nickname “oxess” was nominated as validator. The link is below:
The Prometheus setup took a little time, but we completed this task too and now we can monitor our node remotely. The appropriate alerts and metrics have been configured. You can see them by following the links below:
- http://209.58.143.31:9464/metrics — our metrics
- http://209.58.143.31:9090/alerts — our alerts
Now we are waiting for April 19th, 2021. The Emergency Upgrade: Restart node is planned for this date.
Let’s now figure out what exactly needs to be done in the second phase of “Agoric” Incentivized testnet. First, you need to pay attention to the fact that there is a deadline. This means that all updates must be installed and all reports of progress must be sent by April 21 at 17:00 PDM(Pacific Daylight Time). This report consists of:
- The node update, which took about 7 hours of time. Thanks to the strong support from the development team via discord and the efforts of our team, we have coped with this task. (200 points)
- Install Prometheus Node Exporter. Difficult task that requires some technical knowledge and skills, but we did it, and if you followed the links above, you have an idea of how it should work.
- Keep 80% uptime during load bursts. If you have coped with the first task, and also will solve the task with Emergency Upgrade: Restart node, then gaining of 80% will not be a big problem!
- Cause your validator to be jailed and then recover. This means that it’s necessary to count the time, how quickly you will be returned back to the validators, after you have been released from jail. Here either you deliberately go to jail or during node updates. In any case, you are going to jail if your node willn’t function for 50 blocks.
- Emergency Upgrade: Restart node. Wrote about this earlier. Our team has prepared and written a number of commands in advance that should help us.
- Write & publish an Article. You need to create and write your own original Article/s. The development team will review each Article/s, so I highly recommend don’t using the plagiarism. Topics that you can write about can be found here at this link.
- Build a system performance analysis tool for validators. This is a separate task that is good for everyone who has developer skills. A separate pool will be allocated for this task, and it will be rewarded according to the results from your developments.
More details about each task can be found in the links below.
— — — — — — — — — — — — — — — —
Phases detailed information — link
Discord— link
Regards Alex (Telegram: @oxess, Discord: oxes#8647)
