attack-pass F6: O-1.14 host re-rented after the fleet listing-cap correction
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
parent
2e942495a7
commit
9380508f65
1 changed files with 4 additions and 1 deletions
|
|
@ -155,7 +155,10 @@ host stood up. Vast accepted and dropped five CPU-class rents within 30 s each (
|
|||
W-2123) under the account's automatic new-account spend limit (support ticket open since 6 October), and RunPod has
|
||||
no 2019-class CPU pod; so O-1.14 reads INCOMPLETE on the box proxies today and stays a precondition of the freeze.
|
||||
Next try: the US laptop when it registers on the relay (the exe is ready), or Vast once the spend limit lifts; the
|
||||
bench script is staged and runs in minutes.
|
||||
bench script is staged and runs in minutes. Correction, 10:4x UK (fleet agent): the provider dropped nothing; every
|
||||
rent stood up and ran, hidden by Vast's instance listing cap of 25 rows on an account holding 38, so the hosts sat
|
||||
idle and were destroyed. The fallback is re-rented under the same word (i7-9700K class, two-hour cap from its start,
|
||||
read by id); the result replaces this line when it lands.
|
||||
|
||||
Result (average, verified on the box): class v4 `mx8+sh256x27` runs 4.90 to 5.06 ms per warp cold on one EPYC
|
||||
9454P core (nice 19, taskset), 8.23 ms on the half-core proxy (both SMT siblings busy). Under 10 ms. Status
|
||||
|
|
|
|||
Loading…
Reference in a new issue