metaverse-vm icon indicating copy to clipboard operation
metaverse-vm copied to clipboard

502 Error when syncing new graph

Open Corfucinas opened this issue 3 years ago • 5 comments

Currently, the mainnet endpoint has an availability issue when we deploy a new graph, it takes a long amount of time to sync since getting new blocks often results in ~# 13 attempts per block range due to 502 error.

The endpoint used is: https://vm.mvs.org/mainnet_rpc/

Corfucinas avatar Mar 31 '21 03:03 Corfucinas

Related on EVM calls. Pls describe the details here @inodelisia

betachen avatar Mar 31 '21 07:03 betachen

@inodelisia @betachen

Here's an old log that I took a while ago, the error is similar on the mainnet

image

Corfucinas avatar Mar 31 '21 07:03 Corfucinas

@Corfucinas Please send me a list of RPC eth_ calls that are used most often on the graph node.

inodelisia avatar Apr 01 '21 08:04 inodelisia

@betachen @inodelisia

While doing some debugging I found out the graph node crashed about 7k blocks ago, I've restarted it but got the following log messages

image

Corfucinas avatar Apr 13 '21 11:04 Corfucinas

Let me know exactly what you need and the node/graph commands to provide you a more detailed explanation if required.

Corfucinas avatar Apr 13 '21 11:04 Corfucinas