秋田预测市场 · Discord 审计

API取消订单延迟显著高于开仓延迟的问题

2026-07-01 · Hyperliquid · 4 条相关讨论

; 2026-07-01 01:15:51

Hi! I was checking mempool and noticed that some accounts have pretty much the same delay for opens and cancels (mempool ts - nonce, only keeping accounts where the nonces look like timestamps). Those accounts are around 150–200 ms median for both. But some accounts have much slower cancels — opens are still around 150–200 ms, while cancels are more like 450–500 ms. I'm unfortunately in the 450-500 group. It looks like it started after June 20 update on my side. It also seems like accounts are gradually moving from the slow group to the fast one (attached chart: X = cancel delay on June 22, Y = cancel delay on June 30) I am missing something?

Hi! I was checking mempool and noticed that some accounts have pretty much the same delay for opens and cancels (mempool ts - nonce, only keeping accounts where the nonces look like timestamps). Those accounts are around 150–200 ms median for both. But some accounts have much slower cancels — opens are still around 150–200 ms, while cancels are more like 450–500 ms. I'm unfortunately in the 450-500 group. It looks like it started after June 20 update on my side. It also seems like accounts are gradually moving from the slow group to the fast one (attached chart: X = cancel delay on June 22, Y = cancel delay on June 30) I am missing something?

frogflip 2026-07-01 01:56:15

i would expect that is due to the new "cancel and cancelByCloid actions have a new fast flag, represented by f: true in the action payload" from this api-announcement: https://discord.com/channels/1029781241702129716/1208476333089497189/1516014224730488882 . updated doc: https://hyperliquid.gitbook.io/hyperliquid-docs/for-developers/api/exchange-endpoint#cancel-order-s

i would expect that is due to the new "cancel and cancelByCloid actions have a new fast flag, represented by f: true in the action payload" from this api-announcement: https://discord.com/channels/1029781241702129716/1208476333089497189/1516014224730488882 . updated doc: https://hyperliquid.gitbook.io/hyperliquid-docs/for-developers/api/exchange-endpoint#cancel-order-s

; 2026-07-01 02:01:43

Looks like that's the reason. I'll check it out. Thanks so much!

Looks like that's the reason. I'll check it out. Thanks so much!

Dung 2026-07-01 02:16:11

can you check and tell if that's really the case? It'll be helpful

can you check and tell if that's really the case? It'll be helpful