|Ticket Number:||20081205_1||Ticket State:||CLOSED|
|Ticket Opened:||2008-12-05 21:40||Ticket Closed:||2008-12-09 13:24|
|Ticket Description:||Router swiLS2 unreachable|
Problem Description:The router LS2 was unreachable from Friday (05 Dec 2008) 19:44 until Monday (08 Dec 2008) 12:14.
The router crashed due to a cache memory failure and did not reboot automatically.
Our engineer exchanged the supervisor board and now the system is back online.
During the whole outage traffic to University of Lausanne was re-routed over swiEL2 and EPFL lost its backup path over LS2.
Other sites also had reduced redundancies such as EHL, CERN, UNIFR, UNIBE, IMD.
Affected:From 2008-12-05 19:44 until undefined
Impact: no more redundancy
Sites/Services: UNIL, EPFL, EHL, IMD, UNIBE, UNIFR
The router is back online with the new supervisor board.
From the crash info it appears that there might be a problem with bad memory cache modules.
Although the router seems to be running fine at the moment, we have decided to replace the the supervisor board anyway. We will thus power off the router once again. Interruption is expected to be another 20 minutes.
LS2 is up again in the old configuration.
We're inspecting the routers crash info to see what went wrong.
Engineer arrived on scene, reports green lights from router. A power cycle is made to boot up the router as is, to watch how far the hardware will come up.
In the same rack as swiLS2, there is a "lab" router called swiLST76, with very similar hardware configuration. We have prepared that test router's supervisor engine so that it can be used as a drop-in replacement for swiLS2. We have also prepared the logistics for performing this replacement on Monday morning.
Added new impacts on redundancy after checking network map.
Arriving at noc and opening ticket
Received Call from Connectis Netwatch on cell phone
For all questions about this ticket, please send mail to firstname.lastname@example.org
or call +41 44 268 15 30.