New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Testnet stability before GoS launch #235

Open
adrianbrink opened this Issue Dec 12, 2018 · 20 comments

Comments

Projects
None yet
@adrianbrink
Copy link

adrianbrink commented Dec 12, 2018

We, as the validators, have been involved in testnet deployments for a long time, some of us since December 2017. Even though we really want to see Game of Stakes launching as soon as possible, we also want to have a reliable and fair game. The current process of announcing a starting date/time and then postponing is not only extremely time-consuming but also very stressful.

We don’t have a stable version of the software and the only reason why we caught the last bug (liveness bug in Tendermint) is that a bunch of us have started to set up their own community testnets, irrespective of Tendermint. Given the special status of Game of Stakes for many people in the community (both in terms of signalling and monetary rewards), we would rather see a fair game that doesn’t immediately break due to some simple bug. Instead, we would like to have a testnet that actually test a validator’s preparedness.

We would like to propose that we only launch GoS once we have a testnet that runs for at least a week without any major bugs. A later launch of GoS is better than a botched one. This will enforce the practice that validators should be conservative in the software that we are running.

@zmanian

This comment has been minimized.

Copy link
Member

zmanian commented Dec 12, 2018

Please thumbs up/thumbs down / comment on your point of view.

@fkbenjamin

This comment has been minimized.

Copy link
Contributor

fkbenjamin commented Dec 12, 2018

I was talking to @adrianbrink recently and 100% agree with the proposal above. I think this will in the end benefit the whole experience of the Game of Stakes.

@mdyring

This comment has been minimized.

Copy link
Contributor

mdyring commented Dec 12, 2018

I am all for delaying, until we've had a stable testnet for a meaningful period of time.

It would be a big disappointment for everybody, validators, Tendermint/Cosmos team and "passive" fundraise participants alike, if GoS died due to lack of prior testing.

We have to keep an eye on the long term goal here, which is launch of a stable and production ready network.

@kwunyeung

This comment has been minimized.

Copy link
Contributor

kwunyeung commented Dec 12, 2018

I agree to delay the GoS. It will be better for validators to understand well and test the latest software. It also gives enough time for Tendermint team to solve some issues we found out in the last couple community testnets. The stop of genki-3000 did meant there are bugs which would make GoS failure.

@iammelea

This comment has been minimized.

Copy link
Contributor

iammelea commented Dec 12, 2018

from my point of view @adrianbrink have totally my support in his words. I agree to delay the GoS because imo is the best choice for a successful GoS.

@mnugumanova

This comment has been minimized.

Copy link
Contributor

mnugumanova commented Dec 12, 2018

A delayed but stable release would allow validators to showcase their readiness and resilience in the mainnet like environment

@NodeATeam

This comment has been minimized.

Copy link

NodeATeam commented Dec 12, 2018

Yes, we agree with this issue.

To proceed a fair a stable Game of Stakes environment, I think it is best that we get time to test out fixes and changes to minimize risks before the game starts.

@aaronhenshaw

This comment has been minimized.

Copy link
Contributor

aaronhenshaw commented Dec 12, 2018

I agree with this 100%. We are super excited for Game of Stakes and would like to have more conviction that the software is reliable. A week of a community or cosmos team led testnet without new bugs or software releases feels like enough time to me.

I would just want to reiterate that any software releases should really reset the clock and a new test net should be spun up :)

Thanks for putting this together @adrianbrink and everyone else involved!

@chris-remus

This comment has been minimized.

Copy link
Contributor

chris-remus commented Dec 12, 2018

As one of the validators in the group since day 1, I agree! This approach will move the process in a a beneficial and healthy direction for all involved. Thanks to @adrianbrink for the clear and concise proposal.

@dlguddus

This comment has been minimized.

Copy link

dlguddus commented Dec 12, 2018

b-harvest also agree 100% with launching GoS with enoghly tested stable version of sdk. In my opinion, now cosmos sdk is the time to stop running and catch its breathe, stabilize and firmly stand to be utilized by the great experiment, GoS.

@derfredy

This comment has been minimized.

Copy link
Contributor

derfredy commented Dec 12, 2018

I agree in general... but if we want to give a rest to the community I would include at least a date in the final announcement. Like this: " GoS will start when..bla bla... and NOT EARLIER than this "< DATE >"
Thus, some of us could have a rest until that day.

@nodebreaker-carl

This comment has been minimized.

Copy link
Contributor

nodebreaker-carl commented Dec 12, 2018

I agree! This should have an understanding of the future of Cosmos.
I think stable tests and the gos based on them are very important.

@wimel

This comment has been minimized.

Copy link
Contributor

wimel commented Dec 12, 2018

I agree with delaying GoS.
A lot of time spent learning how it works and how to configure the nodes to continue delaying and delaying just hours before, it's time to continue learning.
As well commented above, it would be good to announce with time to be prepared, I understand that setting specific dates is very difficult, but it would be good to plan in advance the details of GoS.
Thank you for the sincerity and clarification.
The project is good and big, we will continue to support the project and contribute in whatever way we can.

@gin

This comment has been minimized.

Copy link
Contributor

gin commented Dec 12, 2018

👍
🐞📉
📈

@clawmvp

This comment has been minimized.

Copy link
Contributor

clawmvp commented Dec 12, 2018

i also agree.

@timogilvie

This comment has been minimized.

Copy link
Contributor

timogilvie commented Dec 12, 2018

100% agree

@phonikg

This comment has been minimized.

Copy link

phonikg commented Dec 12, 2018

I don't think anyone wins here by rushing things.

@kidinamoto01

This comment has been minimized.

Copy link
Contributor

kidinamoto01 commented Dec 13, 2018

I recalled the original plan was to have a stable testnet for a week before the start of GoS, so postponing now makes sense.

@bunghi

This comment has been minimized.

Copy link
Contributor

bunghi commented Dec 13, 2018

+1
I also agree. ;)

@plzcallmehan

This comment has been minimized.

Copy link
Contributor

plzcallmehan commented Dec 14, 2018

I agree! aye!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment