Measurement point: Stealth NYC, 32 Avenue of the Americas.
Published investigation. Updates run automatically.
01 / Network path
Network path.
Endpoint measurements / ICMP IPv4
Latency over time.
Successful endpoint RTT recordings from up to one hour before and after the event.
Enter time range
Times use the selected timezone; choices include recorded samples and intervals across the available chart.
Successful RTT recording Event interval
07 Nov 2026 01:05 – 07 Nov 2026 03:49 UTC. Hover or tap to inspect; drag across the chart to zoom. On touchscreens, first choose Select time range. Escape or Reset zoom restores the full range. Missing recordings are gaps; this is not an application availability or packet-loss chart.
Select a recording to explore nearby path observations.
View recorded values
Successful ICMP IPv4 RTT recordings.
Recorded at (UTC)
RTT
1.196 ms
1.359 ms
1.274 ms
6.917 ms
1.366 ms
1.346 ms
1.358 ms
1.376 ms
1.343 ms
1.272 ms
1.307 ms
1.371 ms
1.229 ms
5.251 ms
1.406 ms
1.241 ms
1.208 ms
1.268 ms
1.395 ms
1.182 ms
1.350 ms
1.210 ms
1.411 ms
1.199 ms
1.210 ms
1.429 ms
1.183 ms
1.237 ms
1.378 ms
1.335 ms
1.203 ms
1.205 ms
1.183 ms
1.352 ms
1.303 ms
1.303 ms
1.214 ms
1.238 ms
1.204 ms
1.354 ms
1.433 ms
1.197 ms
1.279 ms
1.355 ms
1.342 ms
1.325 ms
1.239 ms
1.343 ms
1.177 ms
1.339 ms
1.235 ms
1.362 ms
1.172 ms
1.363 ms
1.217 ms
1.591 ms
1.200 ms
1.340 ms
1.330 ms
1.224 ms
1.315 ms
1.390 ms
1.208 ms
1.333 ms
1.343 ms
1.191 ms
1.387 ms
1.319 ms
1.230 ms
2.823 ms
1.340 ms
1.345 ms
1.354 ms
1.310 ms
1.323 ms
7.174 ms
1.340 ms
1.571 ms
1.333 ms
1.353 ms
1.346 ms
1.192 ms
1.423 ms
1.366 ms
1.335 ms
1.168 ms
1.256 ms
1.341 ms
1.310 ms
1.372 ms
1.342 ms
1.331 ms
1.327 ms
1.320 ms
1.354 ms
1.196 ms
1.399 ms
1.346 ms
1.320 ms
1.225 ms
1.391 ms
1.195 ms
1.376 ms
1.359 ms
1.210 ms
1.217 ms
1.249 ms
1.340 ms
1.350 ms
1.309 ms
1.297 ms
1.179 ms
1.405 ms
1.596 ms
1.231 ms
1.375 ms
1.358 ms
1.346 ms
1.298 ms
1.348 ms
1.189 ms
1.434 ms
1.262 ms
1.220 ms
1.252 ms
1.228 ms
5.022 ms
1.375 ms
1.292 ms
1.353 ms
19.608 ms
1.311 ms
1.198 ms
1.269 ms
1.214 ms
1.444 ms
19.190 ms
1.314 ms
1.345 ms
1.227 ms
1.206 ms
1.342 ms
1.200 ms
1.338 ms
1.495 ms
1.237 ms
1.190 ms
13.531 ms
1.351 ms
1.562 ms
1.369 ms
1.336 ms
1.353 ms
1.196 ms
1.193 ms
1.327 ms
1.321 ms
1.330 ms
1.216 ms
1.194 ms
1.329 ms
1.293 ms
1.368 ms
1.383 ms
1.248 ms
1.230 ms
1.381 ms
1.210 ms
1.315 ms
1.238 ms
1.326 ms
1.377 ms
1.379 ms
1.314 ms
1.332 ms
1.228 ms
5.702 ms
1.327 ms
1.259 ms
1.466 ms
1.371 ms
1.339 ms
1.333 ms
1.204 ms
1.347 ms
1.314 ms
1.386 ms
1.232 ms
1.309 ms
1.264 ms
1.334 ms
1.190 ms
1.208 ms
1.332 ms
1.282 ms
1.249 ms
1.335 ms
1.203 ms
1.390 ms
1.318 ms
1.220 ms
1.198 ms
1.398 ms
1.207 ms
1.339 ms
1.227 ms
1.325 ms
1.346 ms
1.217 ms
1.862 ms
1.318 ms
1.390 ms
1.331 ms
1.359 ms
1.350 ms
1.207 ms
1.351 ms
1.308 ms
1.303 ms
1.279 ms
1.185 ms
1.356 ms
1.202 ms
1.191 ms
1.222 ms
1.366 ms
1.361 ms
1.215 ms
1.186 ms
1.343 ms
1.209 ms
1.395 ms
1.345 ms
1.352 ms
1.228 ms
1.324 ms
1.203 ms
1.425 ms
1.388 ms
1.217 ms
1.328 ms
1.210 ms
1.186 ms
1.228 ms
1.342 ms
1.312 ms
1.217 ms
1.225 ms
1.188 ms
1.368 ms
1.167 ms
1.189 ms
1.307 ms
1.318 ms
1.391 ms
1.317 ms
1.194 ms
1.186 ms
1.360 ms
1.325 ms
1.323 ms
1.207 ms
1.358 ms
1.245 ms
1.402 ms
1.256 ms
1.355 ms
1.375 ms
1.192 ms
1.192 ms
1.328 ms
1.207 ms
1.320 ms
1.217 ms
1.228 ms
1.451 ms
1.177 ms
1.199 ms
1.206 ms
1.193 ms
1.341 ms
1.186 ms
1.320 ms
1.266 ms
1.340 ms
1.193 ms
1.374 ms
1.194 ms
1.393 ms
1.184 ms
1.218 ms
1.326 ms
1.273 ms
1.370 ms
1.352 ms
1.276 ms
1.373 ms
1.337 ms
1.226 ms
1.251 ms
1.377 ms
1.187 ms
1.197 ms
1.240 ms
1.214 ms
1.207 ms
1.316 ms
1.323 ms
1.356 ms
1.198 ms
1.381 ms
1.251 ms
1.197 ms
1.234 ms
1.239 ms
1.292 ms
1.231 ms
1.398 ms
1.322 ms
1.362 ms
1.340 ms
1.251 ms
1.192 ms
1.347 ms
1.196 ms
1.744 ms
1.216 ms
1.249 ms
1.175 ms
Network path / recorded observations
Network path observations.
25 observations
Choose a measurement type to see the closest of the available sampled observations to the selected chart point. Each probe is separate and may follow a different path.
Before
ICMP traceroute · Recorded address 2
Recorded trace · destination not reached by this trace
RTT is the round trip to each responding hop, not the delay of the preceding link. TTL order may omit routers and does not prove direct links. A missing reply is not proof of packet loss. ASN labels are dated address associations; they do not establish the forwarding path or the operator at the event time. Up to three recorded traces per method and period are selected across the archive. Where no individual trace is available, a comparison may summarize a window; MTR observations aggregate multiple probes. Address aliases are independent across measurement types.
Same method / same recorded address
Compare the traces.
Compare one selected trace from before the event with one during it. Values are individual replies, not a statistical comparison.
ICMP traceroute · Recorded address 2
Before: UTC · During: UTC
Same destination address and probe method. A matching TTL does not establish the same forwarding path.
TTL
Before RTT
During RTT
Responding address
1
0.134 msAS8002 · Stealth
0.131 msAS8002 · Stealth
Same responder
2
0.181 msAS8002 · Stealth
0.189 msAS8002 · Stealth
Same responder
3
1.816 msAS13335 · Cloudflare
1.488 ms
Same responder
TCP traceroute / 443 · Recorded address 2
Before: UTC · During: UTC
Same destination address and probe method. A matching TTL does not establish the same forwarding path.
TTL
Before RTT
During RTT
Responding address
1
0.185 msAS8002 · Stealth
0.145 msAS8002 · Stealth
Same responder
2
0.182 msAS8002 · Stealth
0.179 msAS8002 · Stealth
Same responder
3
10.922 ms
0.878 ms
Same responder
4
No reply
No reply
Not established
5
20.240 msAS13335 · Cloudflare
1.518 msAS13335 · Cloudflare
Different responders
6
No reply
1.239 msAS13335 · Cloudflare
Not established
7
1.281 msAS13335 · Cloudflare
Not recorded
Not established
RTTs include the return path. Different responders, missing replies and independent probe flows limit what a change can localize. MTR snapshots remain separate in the replay.
Event timeline.
Latency anomaly. Times shown are UTC.
Event opened
UTC
Last event observation
UTC
Recovery observed
UTC
Recorded recovery describes this event, not current reachability.
02 / Assessment
Assessment.
Investigation stateRecovered
Path assessment
Findings differ by destination.
Findings differ between destination comparisons. The evidence does not point to one location for the whole event.
What we observed
A sustained increase in measured round trip time was observed.
Stored traces were compared using matching source and destination addresses.
Recording counts describe collected measurements, not application availability or packet loss. Exact packet paths and the return path remain unverified.
Measurements at a glance.
Baseline RTT1.348 msUsual round trip before the event
Last recorded event RTT1.862 msLast successful event recording
Event p95 RTT3.600 ms95th percentile of successful recordings
Change from baseline+0.513 msLast event RTT compared with the baseline
Successful / expected recordings90 / 90Recorded endpoint samples during the event
RTT is round trip time; p95 describes the upper range of successful recordings. Endpoint measurements differ from individual trace replies and are not an uptime percentage.
Record sources and timestamps
Analysis updated
UTC
Public snapshot published
UTC
Public observation window
– UTC
Event and source timestamps are UTC. The latency chart uses its selected timezone. Page refresh times are separate from measurement timestamps.
03 / Method and limits
Method and limits.
How to interpret the sampled path and its limits
Corroborating the path.
Comparisons use matching destinations before and during the event. ICMP traces use echo probes. TCP/443 traces send SYN probes to port 443 only; they do not complete a TCP handshake or perform a TLS or HTTP exchange.
The Stealth gNMI interface verifies the active route-table choice and maps its primary next hop to the Stealth logical interface, configured attachment and, where applicable, the known LAG member set at each snapshot. Stable Stealth replies, together with mapped interface evidence showing no corresponding errors or sustained capacity pressure, reduce the likelihood of a Stealth forwarding issue. An 80% load warning warrants investigation; it does not prove congestion. Lower interval averages can conceal bursts or queue pressure. Optical readings assess the physical link, not congestion.
What remains unseen.
Probe reply handling can differ from packet forwarding. A slow intermediate reply that disappears at later hops does not locate a forwarding problem. A destination that does not respond to a trace is not thereby shown to be unavailable.
Stealth does not use ECMP in this monitored scope. The route snapshot does not identify which physical LAG member carried a particular independent probe, attest line-card forwarding at that packet instant, prove the exact application packet flow or observe the return path. RTT includes the return path, which may differ. No remote reverse traceroute is available. Tunnels and silent routers may hide parts of the path. Address registration alone does not establish the forwarding operator or responsibility.