-
Notifications
You must be signed in to change notification settings - Fork 968
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Bug] Subgraph stuck forever after (falsly?) detecting a reorg #5586
Labels
bug
Something isn't working
Comments
A bit more information. We are using graph-node in Octant. We use self-hosted instance of graph-node on prod. One of automated testing environments deploys set of contracts and a subgraph per test. This is where we observe this behavior most often. |
paulperegud
added a commit
to golemfoundation/octant
that referenced
this issue
Aug 7, 2024
The cause of the error is likely this: graphprotocol/graph-node#5586 Please run the test locally to make sure it passes. Long term we should find a way to have a stable graph-node. Possible avenues: -> fixing the problem in issue above (#5586) -> finding some other way to work around it -> reducing brittleness of graph-node by removing subraphs on test teardown -> ...
paulperegud
added a commit
to golemfoundation/octant
that referenced
this issue
Aug 7, 2024
The cause of the error is likely this: graphprotocol/graph-node#5586 Please run the test locally to make sure it passes. Long term we should find a way to have a stable graph-node. Possible avenues: -> fixing the problem in issue above (#5586) -> finding some other way to work around it -> reducing brittleness of graph-node by removing subraphs on test teardown -> ...
paulperegud
added a commit
to golemfoundation/octant
that referenced
this issue
Aug 7, 2024
The cause of the error is likely this: graphprotocol/graph-node#5586 Please run the test locally to make sure it passes. Long term we should find a way to have a stable graph-node. Possible avenues: -> fixing the problem in issue above (#5586) -> finding some other way to work around it -> reducing brittleness of graph-node by removing subraphs on test teardown -> ...
paulperegud
added a commit
to golemfoundation/octant
that referenced
this issue
Aug 7, 2024
The cause of the error is likely this: graphprotocol/graph-node#5586 Please run the test locally to make sure it passes. Long term we should find a way to have a stable graph-node. Possible avenues: -> fixing the problem in issue above (#5586) -> finding some other way to work around it -> reducing brittleness of graph-node by removing subraphs on test teardown -> ...
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Bug report
This bug report is related to #5101, but involves a at least one important difference - no reorg!
Multiple copies of the same subgraph (against different deployments of the same set of contracts) are deployed to a node running locally. Anvil is providing RPC and it is not instructed to perform (nor it performs to the best of my knowledge) any reorgs. At some point, graph-node detects the reorg. Whole setup is running inside a docker-compose.
This bug is rather repeatable. More subgraphs is deployed, the higher is the chance of seeing it. More load on the machine leads to more frequent occurances.
Log below was redacted to be of reasonable size (original had 2.5mln of lines). I've removed lines mentioning all but two subgraphs, sections where logs were repeated in a loop, messages with following substrings:
Committed write batch
Done processing trigger
Contract call reverted
Call to getCurrentEpoch
I've observed this bug with both
v0.34.1
andv0.35.1
.Relevant log output
IPFS hash
No response
Subgraph name or link to explorer
No response
Some information to help us out
OS information
Linux
The text was updated successfully, but these errors were encountered: