Upgrade oracle full nodes to pull chaindata from daily checkpoint instead of syncing #27
Labels
Comments
mergify bot
pushed a commit
to celo-org/celo-monorepo
that referenced
this issue
Apr 27, 2021
### Description * Upgrades alfajores and mainnet fullnodes to 1.2.5 * Upgrades baklava fullnodes to the commit that is 1.3.0-beta.2 + the txpool fix * Adds `--set geth.use_gstorage_data=false` for AKS fullnodes to prevent them from trying to pull chaindata without having the ability to via a service account (the work to make this be possible is tracked with celo-org/celo-oracle#27) ### Other changes n/a ### Tested Deployed oracle fullnodes ### Related issues n/a ### Backwards compatibility Yes ### Documentation n/a
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Expected Behavior
There were a couple PRs (celo-org/celo-monorepo#5957 and celo-org/celo-monorepo#6298) that change our full node statefulset to download chaindata less than a day old in the event that a full node is starting up and is very out of sync. This cuts the sync time dramatically, and will help us in the event that our full nodes start having issues (eg corrupted chain data).
The work to pull the chain data from outside of GCP has been done in this PR that is currently unmerged celo-org/celo-monorepo#6690.
This work would include:
celotooljs deploy upgrade fullnodes -e <alfajores | baklava | rc1> --context <each context>and dealing with any potential issues that ariseCurrent Behavior
Full nodes do not download checkpointed chain data at startup if they are out of sync
The text was updated successfully, but these errors were encountered: