Investigate F3 participation lease start before existing lease

[masih]

Reported by a number of folks

2024-12-13T22:15:18.724+0700    INFO    f3      lf3/participation.go:118        Renewing F3 participation
2024-12-13T22:15:18.725+0700    WARN    f3      lf3/participation.go:184        F3 participation ticket starts before the existing lease. Acquiring a new ticket.       {"attempts": 0, "err": "ticket starts before existing lease"}
2024-12-13T22:15:19.225+0700    INFO    f3      lf3/participation.go:118        Renewing F3 participation
2024-12-13T22:15:19.226+0700    WARN    f3      lf3/participation.go:184        F3 participation ticket starts before the existing lease. Acquiring a new ticket.       {"attempts": 0, "err": "ticket starts before existing lease"}

Reports so far:

• https://filecoinproject.slack.com/archives/C0556MSR945/p1734104320110779
• https://filecoinproject.slack.com/archives/CR98WERRN/p1734169589139339
• https://filecoinproject.slack.com/archives/CPFTWMY7N/p1733543526322559

Versions on which issue is observed:

lotus-miner version 1.30.0+mainnet+git.9c1964e8a
lotus version 1.30.0+mainnet+git.9c1964e8a  

otus version 1.30.0+mainnet+git.9c1964e8a.dirty
lotus-miner version 1.30.0+mainnet+git.9c1964e8a.dirty

[masih]

For those affected by this issue, can I ask you to upgrade to 1.31 please?

[masih]

https://filecoinproject.slack.com/archives/CPFTWMY7N/p1734391198252789?thread_ts=1733543526.322559&cid=CPFTWMY7N

I didn't check the logs before rebooting, but I noticed there seems to be an issue in relation to synchronization with the processes performing the chain sync/operations and the miners. If miners are restarted, the problem happens. If the chain processes are restarted without touching the miners, it becomes happy.

[rjan90]

Another report here: https://filecoinproject.slack.com/archives/CPFTWMY7N/p1736483146882879. The user is running v1.30.0, but will upgrade and report back if they are seeing the error happening again.

[rjan90]

Following up on the message above, the user reported that it looks normal again (not seeing the logs), after upgrading to v1.31.0 (link.