Submind YouTube summaries
Thumbnail for Hyperledger Indy Contributors Meeting - 2026/08/25

Hyperledger Indy Contributors Meeting - 2026/08/25

Watch on YouTube

Video summary

The August 25th, 2026 meeting of the Hyperledger Indy Contributors focused primarily on the ongoing transition from Ubuntu 22.04 and a strategic shift toward a Docker-only implementation for both Indie Plenum and Indie Node. The presenter explained that moving away from Debian packages was necessary because the tooling for creating those packages is falling out of support, which creates significant overhead. By adopting Docker images where Indie Plenum generates a base image and Indie Node builds upon it, the team aims to simplify the release workflow and streamline the overall deployment methodology. This new approach involves running tests on local environments to ensure all components pass before moving into the production phase for Indie Node. A key discussion point centered on how future upgrades will be managed within this new Docker architecture, specifically addressing whether the upgrade process would remain manual or become automated. The presenter proposed a mechanism where an external signal could stop the current Docker container processing, allowing Kubernetes or Docker to automatically fetch and deploy the latest image upon restart based on a configured repository. This method relies on setting up a specific Docker repository where the desired image is labeled as 'latest,' giving network operators control over what gets deployed while leveraging container orchestration tools to handle the update logic without requiring a full server replacement. Despite the potential for automation in fetching new images, the consensus was that the upgrade process would still require a manual component due to the underlying operating system changes from version 20.04 to 22.04. While scripts exist that could theoretically upgrade the OS without reinstalling servers, the team decided not to pursue that path for this specific transition. Consequently, operators will need to coordinate manually across different service servers to ensure data storage paths remain consistent and that the network stays in agreement with the new images. The presenter acknowledged that while this manual coordination is complex and may take several months to complete for entire networks, it is a necessary step to ensure reliability before fully automating the process. The meeting concluded with an acknowledgment of the challenges involved in upgrading legacy systems, noting that previous transitions from Ubuntu 16.04 to 20.04 were also quite complex and required manual intervention despite promises of easier future paths. The team agreed to proceed with the current plan using Docker Compose configurations that point to existing data storage locations to minimize disruption, even though this still necessitates human oversight for each node. With the build processes for Indie Node currently in progress and testing phases upcoming, the contributors expressed readiness to tackle the upgrade timeline, confirming that they will reconvene in two weeks to review further developments.
Read the full video transcript
Greetings. >> Hello. >> How are you today, Lynn? >> Doing all right. >> Give me a sec here to get the the set up here. >> Excellent. >> Okay. >> All right. I'm just making a copy of last meeting's notes and then we can get started here. Um welcome to the August 25th, 2026 Indy Contributors call. Um we are managed by the Linux Foundation anti-trust policy and the Linux Foundation Decentralized Identity Code of Conduct. Um if you have any questions about that, let me paste the link in the channel here. There we go. All right. Um Let me bring up one more link here before I share my screen. Share my screen. Hey um on the agendas today, we just have the Ubuntu 22.04 upgrade. And do you have anything else you want to talk about today, Lynn? >> Nope. >> All right, awesome. So, um as part of the Ubuntu 22.04 upgrade, um what I'm doing is I'm moving forward to a Docker implementation only. Um and what this looks like is um uh Indie Plenum produces Docker images, and then Indie Node will produce Docker images based off of Indie Plenum's images. Uh this simplifies the release workflow. Instead of creating Debian packages, there's there's a lot of overhead getting the Debian packages working correctly and the tooling seems to be falling out of support. Um and so this should simplify our deployment methodology. Uh what I'm showing here is uh in my local environment, I've got the the build working. Uh it creates a base image and then the production image for Indie Plenum, and then it runs through the the tests. And we can see that all the tests are passing. Um uh and then this will move over into um the Indie Node environment. Um I actually don't remember where I left off on this. I think it's still work in progress. I think I'm working on getting the actions triggered right now. Um uh and then hopefully we'll have an Indie Node and Indie Plenum build based off of um Ubuntu in a Docker container, and then we'll move to the testing phase. Any questions, Lynn? >> Yeah, so I think you said last time that the the upgrade from 20.04 to 22.04 is going to have to be uh manual this time. And it sounds like the reason is because we have to switch to Docker containers. >> Exactly. >> And so what about future upgrades? Will we There's a There's a feature in Indie where you can trigger an upgrade from a remote location and it will automatically update itself. Will that be able to happen in the future? Will all upgrades from here on out have to be manual? >> Um I'm trying to uh make it so that uh we can send a signal to the Docker container to stop processing to to exit. And then if your Docker image is configured to be off of the latest image on restart, uh Kubernetes or Docker, uh I'm hoping that it will gather the the latest image um based on the repository that you're configured to access. And that would be the upgrade mechanism for the Docker environment. is to trigger a um a stop and then um rely on Kubernetes or Docker to fetch the latest image on startup. >> Okay, so uh hopefully latest image is a is configurable as well, so you can get whatever image you want, I guess, for so that your whole your whole uh make sure your whole network is in agreement with whatever image >> Yeah, so likely for your network you would want to set up a docker repository and then have whatever image you want deployed labeled as latest in that docker repository. Then that would give you the control to determine what image is being deployed on your in your environment. There still needs to be some testing done on that process to ensure that we have a reliable way to manage that. But that's the direction I'm looking at. >> Okay. Sounds like it's going to be a several month upgrade when we finally get to upgrading the the networks. But that's that's where it is, so all good. I think that's all the questions I have for now. >> I'm hoping that as part of the upgrade process we could do it in place using docker compose. So that we don't have to go through a several months long process. We just point the docker compose configuration to use the same data storage paths as the previous install. But it would still require a manual process on each of the service servers. >> Yeah, that's that's the part that's the hard part is the manual part where we have to coordinate all the different node operators and and >> Yeah. >> help them to get through the the >> I think there would be a manual process. Well, there's going to be a manual process regardless cuz you're upgrading your OS level, too. >> Well, there's the scripts that can run that will that will make it so the OS doesn't require an upgrade, but it we're not doing that anyway, so it doesn't matter. But, there's I think there's scripts that'll upgrade from 20 to 22 without requiring you to manually install a new server, but I could be wrong on that, I guess. When we were going from 16 to 20, there was no path, right? You There was no easy path, anyway. It was quite complex, and so that one had to be manual. But, we were told at the time that the future ones would not be as difficult. I guess that was not quite true, but that's okay. >> Um So, that's the status update. I'm in process of getting the Indy node builds worked out. Okay. Um thanks for >> Anything else? >> Thanks for coming to the Indy contributors call. We'll see everyone in 2 weeks. >> Bye.