Major network instability

Published on February 16, 2009 at 12:28 pm by iWeb-Op in: Status

We have experienced some major network instability with the BGP sessions between 11:24 and 12:07. There were two ASN (39625 and 47868) who were injecting very long AS paths (over 255) in the global Internet. This caused our Cisco routers to reset the BGP sessions intermittently when they received the bogus routes. We have filtered out those AS numbers and the network is stable now. We are monitoring the situation. We are sorry for the inconvenience caused. Every thing should be back to normal now.

Comments

  1. iWeb-SLee
    February 16th, 2009 7:30 pm

    See: http://www.datacenterknowledge.com/archives/2009/02/16/routing-snafu-causes-downtime-for-web-hosts/

  2. Alex
    February 16th, 2009 10:10 pm

    That only said what you using outdated IOS and misconstrued bgp.
    I bielive you already read about bgp maxas-limit.

  3. iWeb-SLee
    February 17th, 2009 10:45 am

    The bug id is older than the IOS version being used which has been running fine since. The bug may have been reintroduced somewhere down the line. We have set the maxas-limit to contain this.

  4. User
    February 18th, 2009 2:30 pm

    Was there any outages the day before? (15th feb)

  5. iWeb-SLee
    February 18th, 2009 2:49 pm

    No, there was not. All outage is reported on our blog.

  6. iWeb-Heri
    February 19th, 2009 11:38 am

    Here’s an article at Ars Technica for those who want more details about the incident (where it comes from)

    http://arstechnica.com/security/news/2009/02/czech-router-creativity-causes-net-traffic-chaos.ars

About the iWeb blog

The iWeb Blog covers web hosting, web development, web design, Internet marketing and just about every announcement concerning iWeb.

System Status

Maintenance: Control Center [Resolved]
12:49 February 9, 2012
Incident: Network infrastructure (Smart Servers) [Resolved]
11:18 February 8, 2012