shovel: reorg handling trouble

kayleegeorge commented 2 months ago

using a475 shovel version (so should be post-reorg fix in v1.5)
polygon reorg caused shovel to halt, error deleting task_update

wagmiwiz commented 2 months ago

Have same with polygon/ronin on v1.6 (lots of l=error v=582d msg=converge-retry ig=..src=ronin errors in in a loop etc)

ryandotsmith commented 2 months ago

It looks like there are 2 issues:

1) There is a reorg and as shovel attempts to delete data associated with the reorg the SQL statements are timing out. I can't debug this without the query that is timing out. It is likely that one of the integration tables has a bunch of data in it and when shovel is deleting the data the query is timing out.

2) The corrupt chain error occurs here and is typically surface when the RPC is serving incorrect data.

For issue #1 please provide more detail as described in this new issue: https://github.com/indexsupply/shovel/issues/272

I'm not sure what can be done about #2. Feel free to open a separate issue with more detail if this continues to be a problem.

indexsupply / shovel

shovel: reorg handling trouble #271