Skip to content

network monitor hardening#2239

Open
SantiagoPittella wants to merge 9 commits into
nextfrom
santiagopittella-network-monitor-hardening
Open

network monitor hardening#2239
SantiagoPittella wants to merge 9 commits into
nextfrom
santiagopittella-network-monitor-hardening

Conversation

@SantiagoPittella

Copy link
Copy Markdown
Collaborator

Summary

Why:

  • Hardens the network monitor based on issues observed while running it against devnet v0.15.0. including: fixes panics on unexpected peer responses, initialize the dashboard immediately, fixes the prover probes that silently died, and prevents the checker stalls during faucet PoW solving, and a UI overflow in the service cards.

How:

  • Malformed remote responses no longer panic the checker tasks.
  • The dashboard is served immediately at startup. NTX account deployment moved to a supervised background task that retries forever with exponential backoff, publishing "deploying monitor accounts" / unhealthy statuses on the cards until it succeeds.
  • The proof-test payload is fetched from the RPC with retries instead of being required at startup
  • Faucet PoW challenges are solved on a blocking thread (with a wall-clock timeout) instead of on the async runtime.
  • CSS fixes: values with a copy button (.metric-value) now wrap inside the card (min-width: 0, overflow-wrap: anywhere), so long URLs without natural break points no longer push the copy button outside the card box.

Changelog

[[entry]]
scope       = "network-monitor"
impact      = "fixed"
description = "The monitor no longer panics on malformed or version-skewed peer responses; the block-producer mempool object in the /status JSON is now nullable."

[[entry]]
scope       = "network-monitor"
impact      = "fixed"
description = "The dashboard is served immediately at startup; NTX account deployment retries in the background instead of delaying or aborting the monitor."

[[entry]]
scope       = "network-monitor"
impact      = "fixed"
description = "Remote-prover proof-test probes are respawned if they die, retry payload acquisition, and report stale results."

[[entry]]
scope       = "network-monitor"
impact      = "fixed"
description = "Faucet PoW challenges are solved off the async runtime so other checkers are not stalled."

[[entry]]
scope       = "network-monitor"
impact      = "fixed"
description = "Long URLs without wrap points no longer overflow dashboard cards and push the copy button outside the box."

@SantiagoPittella SantiagoPittella changed the title Santiagopittella network monitor hardening network monitor hardening Jun 10, 2026

@Mirko-von-Leipzig Mirko-von-Leipzig left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Only a light review, but thank you!

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.

2 participants