rssLink RSS for all categories
 
icon_red
icon_green
icon_red
icon_red
icon_blue
icon_green
icon_green
icon_red
icon_red
icon_red
icon_orange
icon_green
icon_green
icon_green
icon_green
icon_blue
icon_green
icon_orange
icon_red
icon_green
icon_red
icon_red
icon_green
icon_red
icon_red
icon_red
icon_red
icon_orange
icon_green
 

FS#3331 — FS#7283 — ldn-1-6k - Degraded latency to LINX Extreme LAN

Attached to Project— Network
Incident
Whole Network
CLOSED
100%
We notice a degraded latency to networks routed on ldn-1-6k to lan LINX Extreme. Performance degradation occurs between 13:00 and 10:00 the next day (latency is normal between 10:00 and 13:00 only). We are seeking the origin of the problem.



Date:  Tuesday, 18 September 2012, 00:03AM
Reason for closing:  Done
Comment by OVH - Thursday, 06 September 2012, 19:42PM

The initial problem was identified as "end of line blocking". On a 6k architecture, the fact of having congestion on some exit ports provokes the increasing of the latency on the entry ports. We are in this case since the crash on the cable TAT14 (task #7277) which deprived us of a part of our transatlantic capacity.
Due to modifications at the level of the routing to some asiatic networks, we had to discharge the remaining ports to the US on this router. It allowed us to improve the latency.
We notice now problems balancing problems on the different links.
The max latency to BT and Virgin is now under 30 ms (and stable) which is not the initial level but a lot better. We are still working on this problem in order to have back the maximum quality.