Is there any word on the secret ISP Guru Guild mailing list about this
problem? Is it being worked on? Is anyone projecting a date for when this
problem will be fixed?
Thanks,
Matt
Just FYI, if you have a virtual domain then they don't come through iStar
at all anyway, it would come through our BC Tel link. But that may not
make a difference since both iStar and BC Tel connect to MCI sooner or
later anyway. Also note that any traceroute they attempt from their end
will halt at our BC Tel router, but it can still be useful to locate bad
spots prior to hitting BC Tel.
If there is a bottleneck in California then they'll likely be having the
same trouble accessing a LOT of sites, not just ones we host. Probably
any iStar or BC Tel sites, many PNW sites, etc.
: Is there any word on the secret ISP Guru Guild mailing list about this
: problem? Is it being worked on? Is anyone projecting a date for when this
: problem will be fixed?
Only the same old non-official stuff about MCI and Sprint overselling
their bandwidth and always being in a state of improvement. I haven't
heard of any recent significant changes (bad or good).
Mark
>I've had a few surfers in Texas and California complain that their
>access to IslandNet is extremely slow. Doing a traceroute to their
>ISPs, it appears that there is a nasty bottleneck on the MCI and AT&T
>networks in California. That is, data is moving through ISTAR and on
>through Washington State real quick but it grinds to a crawl when it hits
>the MCI and AT&T machines in California.
>
>Is there any word on the secret ISP Guru Guild mailing list about this
>problem? Is it being worked on? Is anyone projecting a date for when this
>problem will be fixed?
>
>Thanks,
>Matt
{Excerpt from message from mi...@newsguy.com]
Over the course of the past several weeks, members routing through the MCI
backbone to reach the service have reported quite a few instances of
up-and-down connection performance. Up until a few days back, it appeared these
issues had subsided, however, since the beginning of this week, tests results
have shown many instances of heavy packet loss within MCI. Particularly, this
has existed at MCI's connection to the PB-NAP where MCI would peer with our CRL
backbone to transfer data.
Following is a message sent out by MCI's NOC Department, explaining the
situation. With it, users routing through...
bordercore3-atm1-0.SanFrancisco.mci.net
...on their return path, will be impacted by this situation.
>"We are currently workin(g) w/ pacbell on provisioning more bandwith to
>this NAP. We hope to complete this soon."
>
>Regards,
>Steve/iNOC
We hope this information is helpful to folks that utilize the MCI backbone, and
we're just as eager to see things resolved as soon as possible. We will
continue
to post any update information we receive on the situation, and if any members
should receive updates as well, please feel free to forward to us, and we'll
notify others.
Best Regards,
[End]
--
Ray Powell
Victoria, BC
Canada