-
Notifications
You must be signed in to change notification settings - Fork 302
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
Minor fixes to testnet 78 migration #4596
Conversation
…ix bug with validator migration
Nice, thanks for following up (refs #4567). I'll dig into migration testing for 78 today! |
Looks like the revised migration applies fine, but the chain cannot restart after the migration is run, because the halt bit is still set to true. A common mistake: I did the exact same thing when writing the Testnet77 migration (#4503) and had to follow up with a narrow fix in #4508. Unsurprisingly I see that the competing Testnet77 migration in #4564 does not make this same mistake, hat-tip to @erwanor on that. But let's fix it here, so we can proceed with migration testing, and sort out conflicts in #4654 separately. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Needs to reset the halt bit: I'll append a commit and re-review after testing.
Oof, good catch. I had refactored the migration code to de-duplicate all the common operations in migrations but then reverted it and forgot to add those steps back to the testnet 78 migration. |
Thanks for tackling this, even if it didn't land. I remain in favor of cleaning up the migration tooling (#4294), but don't mind writing more migrations as practice to inform what the shape of that cleanup should be. |
let next_proposal_id: u64 = delta.next_proposal_id().await?; | ||
|
||
// Range each proposal and truncate the fields. | ||
for proposal_id in 0..next_proposal_id { | ||
tracing::info!("truncating proposal: {}", proposal_id); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@zbuc doesn't this have the same issue of pulling data using stronger rules that existed when it was stored?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, will follow up in a new PR
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
… in migration (#4606) ## Describe your changes Don't deserialize proto types to avoid applying new domain validations to old types in migration 78 ## Issue ticket number and link #4596 (comment) ## Checklist before requesting a review - [x ] If this code contains consensus-breaking changes, I have added the "consensus-breaking" label. Otherwise, I declare my belief that there are not consensus-breaking changes, for the following reason: > Migration change
Describe your changes
Configure migrations to run testnet 78 migration, add more logging, fix bug with validator migration
Checklist before requesting a review
If this code contains consensus-breaking changes, I have added the "consensus-breaking" label. Otherwise, I declare my belief that there are not consensus-breaking changes, for the following reason: