Skip to content

Conversation

@KaloyanTanev
Copy link
Contributor

No description provided.

Copy link
Contributor

@OisinKyne OisinKyne left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

leaving unrelated comments about pruning/history, but the blobcount flags look correct

compose-el.yml Outdated
Comment on lines 36 to 37
--Sync.AncientBodiesBarrier=${NETHERMIND_ANCIENT_BODIES_BARRIER:-0}
--Sync.AncientReceiptsBarrier=${NETHERMIND_ANCIENT_RECEIPTS_BARRIER:-0}
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Screenshot 2025-11-06 at 16 46 42

Setting 0 would mean downloading the full chain. shouldnt we bump this to like 20000000 or higher? (currently at 23.7m on mainnet)

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Digging a little further, does it make sense to prune history with this flag
Screenshot 2025-11-06 at 17 02 50

and maybe we need a pruning trigger too? not clear to me if pruning will happen without one? (doesn't seem right that it wouldn't unless this is set, so maybe i don't understand, maybe its the difference between in-memory pruning vs full pruning, which is what hybrid offers)
Screenshot 2025-11-06 at 17 00 36

compose-el.yml Outdated
--http.addr=0.0.0.0
--http.port=8545
--metrics=0.0.0.0:8008
--txpool.blobpool-max-count=0
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks like reth defaults to no pruning (archive node), we should fix that or we'll eat everyone's disk if they swap to reth
Screenshot 2025-11-06 at 17 07 05

compose-el.yml Outdated
--Sync.AncientReceiptsBarrier=4367322
--Sync.AncientBodiesBarrier=${NETHERMIND_ANCIENT_BODIES_BARRIER:-0}
--Sync.AncientReceiptsBarrier=${NETHERMIND_ANCIENT_RECEIPTS_BARRIER:-0}
--TxPool.BlobsSupport=Disabled
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have a new concern about this. IIUC, an update was made to allow a CL to fetch blobs from a local ELs mempool. analysis showed that 30-70% of all blobs get imported this way depending on the client. Disabling the blobpool may hurt performance more than we gain from not proposing blobs in the rare case of a local block. Maybe nethermind will separate this behaviours into a different flag. we probably shouldn't turn on this feature if my concern is valid, will try and dig up the reference i remember on twitter.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

NethermindEth/nethermind#9686. Nethermind addressed this for us :D --Blocks.BlockProductionBlobLimit option

@DiogoSantoss DiogoSantoss merged commit db3fd54 into main Dec 11, 2025
3 checks passed
@DiogoSantoss DiogoSantoss deleted the kalo/disable-blobs-el branch December 11, 2025 17:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants