#Namadillo hosting quality control
1 messages Β· Page 1 of 1 (latest)
thanks for tagging! we could use this thread to chat here. or maybe we should start a forum thread for this (ecosystem category)
tagging @chrome cairn @civic latch
@light marten has deployed on Laconic in public, but it seems not fully configured #π π₯-housefire-canary-net message
@worthy needle How about a load-balancer for indexers? https://github.com/tududes/rpc-load-balancer
Contribute to tududes/rpc-load-balancer development by creating an account on GitHub.
I would love to put that together.
That will go along with the plans I have to do some vetting for Namada service liveness to go along with the registry and Namada.community
The results of the liveness check could for sure be a load balancer, as well as a set of maintained Laconic registry entries (Namada class data objects in their registry)
we have a plan to put that together too.
btw, we are a service provider on the Laconic network as well. If someone wants to deploy your namada interface on my service, Iβm very willing to support you in doing so.
@light marten I had a chat with @grand sundial yesterday about using the entries in https://github.com/Luminara-Hub/namada-ecosystem/tree/main/user-and-dev-tools/mainnet and script a liveness check that could feed Prometheus and populate a community Grafana instance. With that data, perhaps we can work with you to translate for Laconic services?
I'll discuss with my team about this idea
Ive an idea. Why not have a clean entry point that redirects users to the self-hosted web fonts without changing their browser URLs. That way, we maintain a decentralized (as far as possible) but make everything look more professional and trustworthy. The ownership should stay with Helix or Anoma, as with namada.net.
Then, we could use Cloudflare Tunnel. I am not 100% familiar with it; it is a suggestion. Each host creates a secure connection, so we donβt need to expose actual backend URLs. The system automatically pulls the latest version from GitHub and checks if an instance is up to date. Only compliant ones get traffic. Users are directed to available and updated instances, either in a round-robin fashion or based on some logic we decide. No matter which instance a user ends up on, the entry point URL stays the same.
Helix or Anoma must maintain the entrance point (via Cloudflare) and the required version (via GitHub).
How can I check what version of self-hosted Namadillos are running?
I can retrieve the latest released version from GitHub: https://raw.githubusercontent.com/anoma/namada-interface/refs/heads/main/apps/namadillo/package.json
Now, Iβd like to compare that version with the Namadillo instance Iβm connecting to.
hit the gear
How I envisioned it:
- There would be one single Namadillo app, mirrored/sharded across hundreds of hosts. (like Bittorrent Maelstrom, IPFS)
- The app would would be controlled by the community, not a single party.
- Whenever there is a new release on Github, a proposal is triggered. If the proposal passes, the release will be installed automatically (a bit like Continuous Integration).
- The infrastructure would be load-balanced with health and integrity checks making sure it is always up-to-date.
So, is this the correct way to query it?
/assets/index-Bv-xPXk2.js
wallets={namada,keplr:keplr$3},version$1="1.10.2"
this Bv-xPXk2 might be rando/obfuscated
tagging @sharp valve as well, in case interested
the problem is that loadbalacing the indexer across multiple entities is easy said but i can see a lot of edge cases
We've just upgrade a new version for both housefire and mainnet.
https://namadillo.pwa.mekonglabs.xyz
https://namadillo-housefire.pwa.mekonglabs.xyz
Namadillo
Namadillo
Hello @light marten
Great job on the upgrade !
Please fell free to submit your interface by making a PR on the namada-ecosystem repo
About the mainnet Namadillo instances:
@unique elbow @astral flax @onyx kelp @lunar thicket @dreamy prawn @sage field
Your Namadillo interface is not up-to-date.
Could you please update it and let me know when it's done?
@hollow nexus @shadow sage
If the URL we have here is still correct, your interface can't be reached.
Are you still running it?
Me wrong, its https://app.namada.cc
Namadillo
Ok, great !
Could you please send a PR with the new url?
Ours is this:
https://namadillo.namada.validatus.com/
And we keep it updated always.
Let me do the PR, it's not correct in the repo.
Namadillo
Updated to v1.11.1
https://namadillo.murphynode.net/
Namadillo
Updated
we've opened a PR: https://github.com/Luminara-Hub/namada-ecosystem/pull/126
Hello ! Nobita,
PR merged. Thank you for your submission.
tk so much bro.
You're welcome. Feel free to include every service your team is running. Wishing you a great Sunday βοΈ
yeah, we will. Have a good day too bro.
reports @here that Namadillo instances aren't updating to unlock Phase 3 functionality
Should Namadillo switch phases manually or automatically?
cc @oblique ridge @fallow lantern
need to update the chain repo
do maintainers know?
https://namadillo.denodes.xyz/ is on Phase 3 v1.11
Namadillo
https://namadillo.5elementsnodes.com/ is updated to 1.12 and on Phase 3
Namadillo
sorry, we'll roll it back
we'll need other hosts to update Namadillo as well
we can follow this on the chain registry commits https://github.com/anoma/namada-chain-registry/commits/main/
@civic latch
outdated Namadillo instances can now be switched to "coming soon" here https://namada.community/infra/interfaces
h/t @grand sundial
cc @civic latch
Everyone can now signal which mainnet interfaces are upgraded or not by editing the "Available" key in the JSON file:
π GitHub - Interfaces JSON
Interfaces with the "no" value now appear at the bottom of this page under the "Coming Soon" section:
π Namada Community Interfaces
I'll check and update all the Namadillo instances in a few hours, giving you some time to update.
Hi Harri, I saw that you approved this PR. Is it ready to be merged?
We are looking to deploy Namadillo on the main branch of the Namada interface (currently on our forked repo) on the Laconic network.
https://github.com/anoma/namada-interface/pull/1333
Hi, the code is good, I'll just ask for internal approval from the team as well and we should be good to merge π
yeah tk, have a good day bro.
Namadillo interfaces list updated > https://namada.community/infra/interfaces
@gaunt tartan @dreamy prawn @onyx kelp @pearl roost
Please send a PR to update your availability status when your interface is upgraded.
Done β .
Done @grand sundial
We were updated. Where should we send a PR?
Oh, it's already there since we begun to provide it
Is it possible to find out if the latest version of the interface will be updated for Docker?
Ive checked the Namadillos listed at https://namada.net and https://raw.githubusercontent.com/Luminara-Hub/namada-ecosystem/refs/heads/main/user-and-dev-tools/mainnet/interfaces.json and there are a few running without http2.
According to @void trellis We should do that.
http2 on, otherwise sync is unbearably slow
#π-mainnet-operations message
Here is the list of interfaces.
https://namadillo.shield-crypto.com/ @keen compass
https://valley-of-namadillo.grandvalleys.com/ @charred hill
https://namada-interface.palamar.io @modest dune
https://namadillo.stakeandrelax.net @river hemlock
https://namada-interface.nodes.guru @keen bronze
https://interface.loserboy.xyz @gaunt tartan
We've just upgraded the indexer to the latest version (2.2.0) that it has just been released, but it seems that we were too fast. Indexer 2.2.0 is not compatible with Namadillo 1.12.1.
Roll back in progress
rollback completed. Indexer downgraded to 2.1.0
Thanks, I've enabled gzip, http2 and Improved my ngiinx configuration. .
Fixed
While your Namadillo was displaying a banner about a outdated/incompatible indexer, it would still be fully functional and the latest namada-indexer tagged v2.2.0 contains bug fixes.
A new release of Namadillo will have the bumped version to make that warning go away with v2.2.0
Hello,
I'm not sure if a pre-made Docker image exists...
Maybe someone from the eng team can tell you more about it.
Did you run into this issue at all?
While having the postgres db volume intact, I updated to namada-indexer v2.2.0 which resulted in the following error: [+] Running 17/17 β chain Built 0.0s β governance Built 0.0s β parameters Built...
yup me too!
it happens in v2.1.0 too!
are you running any stable release right now?
yup v2.1.0
im redeploying it π
should be done soon
v2.0.1 is what worked for me.
im still using v2.1.0 and this happend
Same here.
What does it mean? Shall we upgrade the indexer to 2.2.0 and ignore the Namadillo warning ?
namadillo: v1.15.0
Fixed in v1.15.1
JFYI:
In the future, pls only actally upgrade your infra once an announcement has been made on #software-releases
Would not be better to have some sanity check instead and release only those version that can be deployed ?
For operators is much more efficient to implement automated process that deploy a new version as soon as it is released rather than watching and react on the discord channel.
Discord channel is great and we should keep it, but it should not be treated as a tool gate for prod deployments.
I agree. It was JFYI
Hey, folks! Does Proposal #13 show up in your Namadillo 1.15.1?
Namadillo
Hi team, from version v1.14 I've got the error when build interface on Laconic network, you guy can check the build-webapp.sh which I've added before. I think I am missing or using an outdated Rust library. Do you have any suggestions?
@static burrow @civic latch @chrome cairn @worthy needle
It's hard to say exactly why the wasm-dependencies failed to build from that output :/ I had temporarily seen cargo build issues, but they seem to have worked themselves out. It may be good to try to run the build again
Slightly unrelated, we can change on of the lines in the script to speed things up:
Since yarn --cwd ${PKG_DIR} wasm:build || exit 1 is being called, we can change:
yarn --cwd ${PKG_DIR} build || exit 1
To:
yarn --cwd ${PKG_DIR} build:only || exit 1
So as not to invoke wasm:build twice
One more thing: We also recently switched to a multicore build of the wasm for Namadillo hosts. To build it with this support, we use:
yarn --cwd ${PKG_DIR} wasm:build:multicore || exit 1
But this requires that Namadillo is served with a couple headers specified, here is an example for Nginx:
https://github.com/anoma/namada-interface/blob/3d79203fa5942877f6b668d4cc433009e9d6c870/docker/namadillo/nginx.conf#L8-L9
Sorry this wasn't caught in the original PR!
I'm guessing but it looks like the development for that script was done in an env that already had rust (cargo) installed. You'll have to include something like this in the entrypoint perhaps:
apt update
apt install curl -y
curl https://sh.rustup.rs -sSf | sh -s -- --default-toolchain stable -y
well done bro, thank you so much. It work perfect right now
check out at: https://namadillo.pwa.mekonglabs.xyz/
Namadillo
big love bro
Hi Justin, I've opened a PR to decrease building time and update package list before building on Laconic SP to resolve uncommon errors.
https://github.com/anoma/namada-interface/pull/1911
btw we've upgraded to v0.18.0 and it worked well
https://namadillo.pwa.mekonglabs.xyz
synced up π
awesome dude, let's shield π