In the last run, there were two problems at the RPC layer returned
from light nodes' RPC end points. I think exercising the light client
proxy RPC system is something that can/should be done via unit
testing, and that likely these errors are (in production) transient
and (in CI) very likely to fail for test environment issues.
These are mostly the timeouts that I think we're still hitting in CI.
At this point, the tests (on master) pass on my local machine (which is quite beefy) so I think this is just the first in (perhaps?) a sequence of changes that attempt to change timeouts and load patterns so that the tests pass in CI more reliably.
In the transaction load generator, the e2e test harness previously distributed load randomly to hosts, which was a source of test non-determinism. This change distributes the load generation to the different nodes in the set in a round robin fashion, to produce more reliable results, but does not otherwise change the behavior of the test harness.
This PR make some tweaks to backfill after running e2e tests:
- Separates sync and backfill as two distinct processes that the node calls. The reason is because if sync fails then the node should fail but if backfill fails it is still possible to proceed.
- Removes peers who don't have the block at a height from the local peer list. As the process goes backwards if a node doesn't have a block at a height they're likely pruning blocks and thus they won't have any prior ones either.
- Sleep when we've run out of peers, then try again.
