20 hours work 4 hours sleep
For three consecutive epochs (19, 20, 21), the DApp360 node had an accelerating downtrend for all our performance and ranking metrics. As mentioned in the last two epochs, our team was more stressed and demoralized than ever. At this point, we still didn’t have a hack solution for the network instabilities. So one option, we tried one radical option. Sit by the computer terminal for 20 straight hours, and manually restart or reconfigure the node every time there was a forked or stuck node. If you visit our Telegram link in DApp360 Pool Community, you would see the 20 hours of real-time block-by-block monitoring that I provided to our delegates while I was babysitting the node. It started on January 4th, 2020, at 12:17 pm and ended at 2:00 PM on January 5, 2020. There you will find every single comment, image posts, and delegates encouragement that spurred our team to continue.
With this block-by-block monitoring, we got lucky. We created 18 blocks and missed three. This performance was a 28% improvement from our poor performance in the previous epoch (21). Our 83% performance in this epoch (22) was a huge reversal of the downtrends.
More importantly, during this epoch, I remembered the encouragement and loyalty shown by two of our very dedicated delegates Eli Taylor, and Federico Confalonieri. They were extremely generous in extending positivity, loyalty, and commitment to our frustrated team members. These guys encouraged our team during this stressful epoch. Even though we performed better, we knew that the cost to keep the node up was extremely high. We could not provide 20 hours of manual monitoring to perform at 83%. It was the consequence of a crappy network. It was not an efficient approach, but we were grateful that we had reversed the downtrends from the last three consecutive epochs.