Bug when checkpoint syncing on prater
I tried a few checkpoint providers so far, and they all give me the same error
Mar 23 00:01:45.531 INFO Logging to file path: "/ethclient/lighthouse/beacon/logs/beacon.log"
Mar 23 00:01:45.532 INFO Lighthouse started version: Lighthouse/v3.4.0-38514c0
Mar 23 00:01:45.532 INFO Configured for network name: prater
Mar 23 00:01:45.533 INFO Data directory initialised datadir: /ethclient/lighthouse
Mar 23 00:01:45.533 WARN Running HTTP server on port 5052
Mar 23 00:01:45.534 INFO Deposit contract address: 0xff50ed3d0ec03ac01d4c79aad74928bff48a7b2b, deploy_block: 4367322
Mar 23 00:01:45.593 INFO Connected to external block builder builder_profit_threshold: 0, builder_url: "http://mev-boost:18550/", service: exec
Mar 23 00:01:45.593 INFO Starting checkpoint sync remote_url: https://prater.checkpoint.sigp.io/, service: beacon
Mar 23 00:01:46.446 WARN Remote BN does not support EIP-4881 fast deposit sync, service: beacon
Mar 23 00:01:47.470 CRIT Failed to start beacon node reason: Unable to parse SSZ: OffsetSkipsVariableBytes(388). Ensure the checkpoint-sync-url refers to a node for the correct network
Mar 23 00:01:47.470 INFO Internal shutdown received reason: Failed to start beacon node
Mar 23 00:01:47.470 INFO Shutting down.. reason: Failure("Failed to start beacon node")
Failed to start beacon node
Mar 23 00:01:48.067 INFO Logging to file path: "/ethclient/lighthouse/beacon/logs/beacon.log"
Mar 23 00:01:48.068 INFO Lighthouse started version: Lighthouse/v3.4.0-38514c0
Mar 23 00:01:48.068 INFO Configured for network name: prater
Mar 23 00:01:48.068 INFO Data directory initialised datadir: /ethclient/lighthouse
Mar 23 00:01:48.068 WARN Running HTTP server on port 5052
Mar 23 00:01:48.068 INFO Deposit contract address: 0xff50ed3d0ec03ac01d4c79aad74928bff48a7b2b, deploy_block: 4367322
Mar 23 00:01:48.120 INFO Connected to external block builder builder_profit_threshold: 0, builder_url: "http://mev-boost:18550/", service: exec
Mar 23 00:01:48.121 INFO Starting checkpoint sync remote_url: https://prater.checkpoint.sigp.io/, service: beacon
Mar 23 00:01:48.966 WARN Remote BN does not support EIP-4881 fast deposit sync, service: beacon
Mar 23 00:01:50.217 CRIT Failed to start beacon node reason: Unable to parse SSZ: OffsetSkipsVariableBytes(388). Ensure the checkpoint-sync-url refers to a node for the correct network
Mar 23 00:01:50.217 INFO Internal shutdown received reason: Failed to start beacon node
Mar 23 00:01:50.217 INFO Shutting down.. reason: Failure("Failed to start beacon node")
Failed to start beacon node
Seems like this is a bug with smartnode, but hard to tell... here's my lighthouse config.

Getting this error regardless of whether I add any additional beacon client flags.
Have you tried a different checkpointz provider? https://goerli.beaconstate.ethstaker.cc/ has worked well for me in the past.
I'm 99% sure this is not a smartnode issue.
My first sentence says I’ve tried a few (3 if I remember correctly)
Ah. Well, the Mr. Frundles release of LH from Nov 2022 added a feature that isn't supported by checkpointz.
If the checkpoint sync provider does not support the feature you'll see a pair of warning logs:
WARN Remote BN does not support EIP-4881 fast deposit sync
WARN Failed to finalize deposit cache
But as you can see, the fallback behavior isn't working for you either:
Mar 23 00:01:47.470 CRIT Failed to start beacon node reason: Unable to parse SSZ: OffsetSkipsVariableBytes(388). Ensure the checkpoint-sync-url refers to a node for the correct network
This is because you're probably using smartnode 1.7.5 instead of 1.9.0-rc2, which is the post-shanghai, post-capella beta for smartnode, and is required on Prater test net.
I recommend joining the discord so you get notified about required updates, at least while running a testnet node.
Anyway, run rocketpool service version to check your version. If it isn't 1.9.0-rc2, you can install it with wget https://github.com/rocket-pool/smartnode-install/releases/download/v1.9.0-rc2/rocketpool-cli-linux-amd64 -O ~/bin/rocketpool (assuming x86_64), and then rocketpool service install -d; rocketpool service start
Since you're using LH you will want to upgrade the container tag in rocketpool service config to sigp/lighthouse:v4.0.1-rc.0 because LH had a bug with their 4.0.0 release, which is what 1.9.0 will attempt to use.
See the 'releases' channel on discord for more info
That’s very likely exactly what was happening, thank you for letting me know. I followed the instructions on the website though, so there’s probably a command that needs to be updated somewhere.
The docs on the website fall out of sync every hard fork unfortunately. Probably room for improvement there, you're right.