Hi Otto,

I have tweaked a little on our ISIS links to/from our routers in ASIA. Hopefully it will help in future cases. I have a little problem understanding what other problems you have been observing and or what you would be expecting to see in case of errors in network and time for network to converge.. could you please help me understand your microdep graphs etc.

Thanks..

-- Carsten Pettersson Senior IP Network Engineer carsten@nordu.net | mobile: +45 3118 7401

On 15 Feb 2022, at 16.22, Otto J Wittner otto.wittner@sikt.no wrote:

Hi, Nordunet NOC!

While browsing yesterdays (2022-02-14) events discovered by our https://microdep.uninett.no end-to-end monitoring tool, we observe a collection of "routechange-events" with associated connection downtimes ("Lost") in Nordunet:

## trondheim-mp to adelaide-mp https://microdep.uninett.no/tracetree.html?topo=/dragonlab/mp/trondheim-mp/2...

Day Time Lost(s) Queue(ms) 14 14:01 0.1 0.2 14 14:22 26.9 0.3 14 14:25 0.1 0.2

## trondheim-mp to auckland-mp https://microdep.uninett.no/tracetree.html?topo=/dragonlab/mp/trondheim-mp/2...

Day Time Lost(s) Queue(ms) 14 10:23 2.4 0.2

(Also observable on trondheim-mp to saopaulo-mp: https://microdep.uninett.no/tracetree.html?topo=/dragonlab/mp/trondheim-mp/2...)

## trondheim-mp to zurich-mp https://microdep.uninett.no/tracetree.html?topo=/dragonlab/mp/trondheim-mp/2...

Day Time Lost(s) Queue(ms) 14 17:27 0.6 0.1

26.9, 2.4 and 0.6 second outages... can something perhaps be further optimized?

BR O2

-- Otto J Wittner tel +47 99550566 Sikt http://sikt.no