Update simulated geth client wrapper & LogPoller tests

[reductionista]

In this PR:

1. Update SimulatedBackendClient to wrap simulated.Backend & simulated.Client ( instead of deprecated backends.SimulatedBackendClient )

2. Add SetActiveClient() to SimulatedBackendClient to simulate switching from one rpc server to another while running.

3. Add RegisterHeadByNumberCallback to SimulatedBackendCilent, to register a custom callback function to be called each timeHeadByNumber() is called.

4. Add to SimulateBackendClient ability to run in "optimism mode", which behaves slightly different from geth: returns success rather than "block not found" if FilterLogs() is called with an invalid or unknown block hash.

5. Update BackupPoller test to use the previous 3 features to simulate failing over.to an out-of-sync optimism rpc server during a race condition between pulling a particular block and its corresponding logs. (This is exactly the case which motivated the creation of Backup LogPoller.) This avoids the need to use rawdb, which is no longer accessible in geth.

6. Replace markBlockAsFinalized() and markBlockAsFinalizedByHash() in LogPoller tests with finalizeThroughBlock().

7. Refactor logic in LogPoller tests previously using markBlockAsFinalized(ByHash) to make use of finalizedThroughBlock() instead.

8. Update some tests for new block timestamp behavior in Simulated Geth: automatic time increment between simulated blocks is 1ns now instead of 1s, and Client.AdjustTime() now automatically commits the current block after adjusting its timestamp

9. Update rest of tests in LogPoller and other packages to use new data types associated with (1.), instead of deprecated types

[github-actions]

I see you updated files related to core. Please run pnpm changeset in the root directory to add a changeset as well as in the text include at least one of the following tags:

• #added For any new functionality added.
• #breaking_change For any functionality that requires manual action for the node to boot.
• #bugfix For bug fixes.
• #changed For any change to the existing functionality.
• #db_update For any feature that introduces updates to database schema.
• #deprecation_notice For any upcoming deprecation functionality.
• #internal For changesets that need to be excluded from the final changelog.
• #nops For any feature that is NOP facing and needs to be in the official Release Notes for the release.
• #removed For any functionality/config that is removed.
• #updated For any functionality that is updated.
• #wip For any change that is not ready yet and external communication about it should be held off till it is feature complete.

[reductionista]

Taking this out of draft status, because the most complex part of it--getting Backup LogPoller test to work--is done and passing. Most of the tests that broke due to not being able to mark blocks finalized any more have also been updated and are passing. Just have a few more to go through, as well as updating the types in some tests. Seems like a good time to start getting some eyes on it.

[jmank88]

Let's group these with the other geth lines

[jmank88]

Did you mean to update these functions files? Looks like stray make generate effects that can be discarded

[reductionista]

They can probably be discarded, I just figured it's better to have the latest make generate checked in than something out of date.

One reason I re-ran make generate was because GETH_VERSION was still set to the old version in some of them:

-GETH_VERSION: 1.13.8
-forwarder: ../../../contracts/solc/v0.8.19/KeystoneForwarder/KeystoneForwarder.abi ../../../contracts/solc/v0.8.19/KeystoneForwarder/KeystoneForwarder.bin ed9164cfe4619dff824b11df46b66f4c6834b2ca072923f10d9ebc57ce508ed8
-keystone_capability_registry: ../../../contracts/solc/v0.8.19/CapabilityRegistry/CapabilityRegistry.abi ../../../contracts/solc/v0.8.19/CapabilityRegistry/CapabilityRegistry.bin d4e0661491c2adc7f0d7553287c938fb32664b9f07770f6d57ae6511ce9884cd
-ocr3_capability: ../../../contracts/solc/v0.8.19/OCR3Capability/OCR3Capability.abi ../../../contracts/solc/v0.8.19/OCR3Capability/OCR3Capability.bin 9dcbdf55bd5729ba266148da3f17733eb592c871c2108ccca546618628fd9ad2
diff --git a/core/gethwrappers/operatorforwarder/generation/generated-wrapper-dependency-versions-do-not-edit.txt b/core/gethwrappers/operatorforwarder/generation/generated-wrapper-dependency-versions-do-not-edit.txt
index 1569801b3f..a30823b737 100644
--- a/core/gethwrappers/operatorforwarder/generation/generated-wrapper-dependency-versions-do-not-edit.txt
+++ b/core/gethwrappers/operatorforwarder/generation/generated-wrapper-dependency-versions-do-not-edit.txt
@@ -1,4 +1,4 @@
-GETH_VERSION: 1.13.8
+GETH_VERSION: 1.13.14

Should the above changes get backed out?

In the case of KeystoneForwarder, it was not only using the old version of geth, but the abi itself has changed. I can't remember for sure whether that was the main motivation--it might have been causing some test failures but not sure.

Looks like the only difference in functions_allow_list, functions_coordinator, and llo-feeds were the compiled binary itself, so I assume those can be ignored.

[jmank88]

CI already enforces that the latest generate version is used. If you are able to edit generated files without failing a check, then you found something that is not tracked correctly. There is a gap that old generated files which are not being regenerated are not noticed because they don't produce a diff.

[jmank88]

see: #13138

As a follow-up to prevent this from happening: Our CI process does not see these as problematic, since they are committed and do not produce a diff. We could add a step before make generate that removes all the generated mockery files to ensure that one which do not get re-generated produce a diff that fails CI.

[jmank88]

scratch file?

[reductionista]

Present status:

• All tests in LogPoller package passing

• All deprecated types have been updated (in all packages), Lint is now passing

• Seeing many errors in the Core CI tests still, but nearly all of them are one of 3 things, which seem very related and possibly have the same root cause:
  no contract code at given address
Execution reverted
  `Execution reverted: Target is not a contract"

• One weird error in a core test, that happens in CI but passes locally:
  --- FAIL: TestEVMForwarderPresenter_RenderTable (0.00s)
  forwarders_commands_test.go:48:
  Error Trace: /home/runner/work/chainlink/chainlink/core/cmd/forwarders_commands_test.go:48
  Error: "\n------------------------------------------------------\nID: \nAddress: 0x4Fc0FeeD7E3F8d9bb6758d3084aa0ff6cFB5C486\nChain ID: 4\nCreated At: 2024-05-29T04:33:50Z\n------------------------------------------------------\n" does not contain "1"

• For some reason, this error causes the Clean Go Tidy & Generate CI check to fail:
  cp: cannot create directory '../../operator_ui/..//core/web/assets': No such file or directory

I think the last one was what prompted me initially to run make generate locally, but I guess that didn't fix it.

I'm not sure how many of these errors were already in the branch I'm trying to merge into, but seems like it might be about time to merge it and rebase that branch?

[cl-sonarqube-production]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.4% Duplication on New Code

See analysis details on SonarQube