Nitro V6 Conntrack Idle Timeout - Silent Drop and Keepalive
TCP established idle timeout default 432,000s -> 350s on Nitro V6 ENIs: how a pooled connection dies silently and how keepalive prevents it
Guided views
Explore this system
Step through curated paths without changing the source diagram.
Beat
Next
ReadyChapter 01 / 01
Guided chapter
Diagram guideExplore this system
Inspecting compiled semantics
E ExportT ThemeS Style0 Reset+ Zoom in- Zoom outEsc Close
Find a node
⌕/
No matching nodes
Semantic passport
Verified source
Authored reach
Route probeChoose a start node
Pick two semantic nodes on the diagram
Choose the source, then the destination. Direction matters.
Semantic lensCompare system roles
Choose up to two semantic kinds. One reveals its real traffic; two compare only direct authored relationships.
Choose a kind to inspect its nodes and touching relationships.
Semantic radar
Building overview
Click nodeDrag to pan
What Changed
• Nitro V6 (m8i / r8i, June 2025) lowers the SG conntrack TCP established idle default to 350s
• Previous generations keep 432,000s (5 days); P6e-GB200 is the exception on V6
• This is ENI / hypervisor level tracking, not kernel nf_conntrack
Silent Drop
• Expired entries emit no FIN / RST to either side
• Pools reuse dead connections; first requests after long idle hang
• Karpenter / attribute-based ASGs can move workloads to V6 unintentionally
Keepalive Fix
• Start probes before 240s so no layer ever sees the flow as idle
• Per-socket SO_KEEPALIVE is required - sysctl only tunes intervals
• Safe across mixed Nitro generations, LBs, and NAT Gateway
Sequence diagram • Built with Archify • Create yours ↗ • Hover to trace • R route • Click to focus • +/− zoom • M radar • [/] views • P play story • T theme • E export