All Collections
EOS Support Media
Bi-Monthly EOS Node Operator Roundtable Summary [September #2]
Bi-Monthly EOS Node Operator Roundtable Summary [September #2]

Published on October 2, 2023

Markus Hinrichs avatar
Written by Markus Hinrichs
Updated over a week ago

Author: Markus Hinrichs

Editor: Randall Roland

Node operators, Antelope core developers, and community members get together each week to talk about the network and how it's developing. The primary objective of each Node Operator Roundtable is:

“…to improve the Antelope protocol (specifically) for node operators.”

Roundtables occur every Wednesday. Visit the Telegram channel for information about joining. The EOS Network Foundation provides tutorials and documentation for those who want to learn the basics of operating an EOS node.

Below is a list of the roundtables contained within this bi-monthly summary:

  • September 20: API node testing for Leap 5.0.0 Deferred Transactions are deactivated

  • September 27: Discussion centred on issues that arose during early API testing.

Look for additional meeting notes and comments on GitHub. Videos reside on the ENF’s YT.

September 20: API node testing for Leap 5 | Deferred Transactions deactivated

This succinct 15-minute Roundtable session encapsulated the following key topics:

Deferred Transactions Status:

  • Deferred transactions are disabled on Jungle and Kylin.

  • In the process of disabling on EOS mainnet.

  • Verbal confirmations were received from BPs regarding deactivation.

  • Substream confirms no more deferred transactions.

RAM Usage:

  • Keep an eye on RAM usage for potential increases.

  • An increase in RAM usage may be due to accumulated deferred transactions (queue).

API Node Testing:

  • Updates made to the API Node testing document, addressing concerns and confusion from previous meetings.

  • Encouragement to test early APIs and monitor RAM.

“Keep an eye on RAM usage if it starts going up significantly, we have to take action to clear those out.” Kevin Heifner

Timeline for Leap 5.0 Upgrade:

  • Release Candidate 1 (RC1) is the next major milestone.

  • RC1 is needed for testnets to proceed.

  • Further testing and monitoring are ongoing.

Acknowledgement of Contributions:

  • Michael from EOS USA thanks Kevin Heifner for state history fixes and contributions.

  • Gratitude was expressed for the invaluable assistance and unwavering support from the community in diligently preparing for the imminent upgrade.

September 27: Discussion on issues that arose during early API testing

This Node Operator Roundtable was notably short, lasting less than half an hour. The primary emphasis of the meeting centred around the continuous API testing efforts.

Testing Progress and Status Updates:

  • The discussion revolved around the early tests of APIs for Leap 5.0.

  • Brian Hazzard (ENF) asked if there was any feedback or questions to report after testing the API.

Feedback from Matthew Darwin:

  • Matthew Darwin reported that his first two tests failed.

  • Issues related to pin builds and the Prometheus exporter were discussed. Some GitHub issues are being created.

  • Reproducible builds are already being addressed.

  • Matthew Darwin expressed concerns about the output from Prometheus.

  • There was a discussion about configuring Prometheus to output relevant data.

New Way of Configuring Ports:

  • Kevin Heifner (OC) provided a description of the new method for configuring ports, and a link to the issue was shared, which will then be included in the release notes and final documentation.

  • Within the configuration, you now have the ability to define categories for different endpoints. You can specify addresses, host names, ports, or socket paths for each of these categories in the config.

Updates on System Contracts:

  • Eric Passmore worked on updates to the system contract, specifically the removal of deferred transactions.

  • He mentioned working on this and then plans to run the new release on Jungle testnet before arranging a time for putting it into production.

  • Deferred transactions were created for system resource refunds and in the EOS name bidding process. If the bid was lost, the deferred transaction would expire, and a refund would be automatic. Now, explicit refunds are required if the bid is lost.

  • It was noted that Jungle, with its older/different system contracts, still generates some deferred transactions, “but this shouldn't change anything” and will soon be cleaned up.

No other topics were raised during the meeting, leading to its conclusion. It was acknowledged that numerous bugs require attention, ensuring there is no shortage of tasks to tackle. The participants expressed their anticipation of reconvening at the roundtable next week.

Sources & References

Did this answer your question?