-
Notifications
You must be signed in to change notification settings - Fork 1.2k
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
Audit how lotus picks the initial sync target #1546
Comments
@Schwartz10 Thanks for reporting! I've seen a number of people report this, we are looking into it |
Interopnet doesn't exist anymore, but we should at least audit how we pick our initial sync target before closing this issue...it's a little concerning |
@arajasek should this be in Stabilization milestone? |
@jennijuju Nah, I don't think it's very prevalent. I think we can close this after convincing ourselves our target-picking logic is sound. |
We think this is no longer presists in the miannet. |
Describe the bug
after a clean install of lotus, running
lotus daemon
started syncing to the wrong chain.The "wrong" chain is an assumption from running
lotus sync status
. The right chain (presumably) has a height diff close to chainhead from a fully synced node:Healthy node syncing:
Unhealthy node syncing?
To Reproduce
Not exactly sure how we reproduced this, but it happened once on a local MacOS daemon, and once on a remote AWS instance, in a kubernetes cluster. It seems to happen randomly? Maybe we connected to bad bootstrap peers?
An observation:
When comparing
lotus net peers
on a daemon syncing to the correct chain (judging by its height diff) and a daemon syncing to (presumably) an unhealthy chain, some peers seem to overlap. In both cases (even when syncing to the presumably unhealthy chain) the peer set is large (> 20 peers).Expected behavior
We should sync to the canonical chain.
Version (run
lotus --version
):lotus version 0.3.0+gitfc3b42df
The text was updated successfully, but these errors were encountered: