Closed AstaFrode closed 8 months ago
Hey @AstaFrode, do you know what events are in the block when this issue is encountered? Are there any XCM-related events in there?
@cdamian The blocks that went wrong had the following events,no XCM related events.
You can check it out in the CESS browser. The block hash is 0xbbc3aef1f9040e76ef38793ba75a58c0a76dfea76b57ad8aa0cb9518b7188377
Hi, I had the same problem, have you solved it yet?
@ygcool I now rely on docker's restart mechanism to keep the service running. @cdamian Is there any progress on this issue?
We did not find the cause of the memory leak so far. An alternative would be to use the recently added event retriever - https://github.com/centrifuge/go-substrate-rpc-client/blob/master/registry/retriever/event_retriever_live_test.go#L52
I'm experiencing the same with v4.2.1
. But it actually does not fail with OOM in my case, but consumes about 20GB+ RAM and a lot of CPU before returning an error on an attempt to decode events from a runtime where the event definition has one more field than the type in my EventRecords
.
When the program is running normally, this error suddenly appears and directly causes the program to be killed.
golang:
go 1.19
go-substrate-rpc-client:
github.com/centrifuge/go-substrate-rpc-client/v4 v4.0.13
related code:
system:
error message: