since moving to Inflow: >99.99%
||The building that houses the Sunguard Data Center where we keep our servers experienced a fire today in an electrical transformer. The data center's emergency generator immediately took over as planned, but the Nashville Fire Department felt that even though the generator was completely isolated from the other building systems, there might be a "Life Safety" issue. As a result, they required the data center to shut down the generator and all Sunguard Nashville systems lost power until the City allowed power to be restored. ICG Link systems were down for three hours and twenty one minutes as a result of this failure.
||One of our mail servers stripped content from the body of messages for 2 hours. A disk problem was finally tracked down and corrected. Mail was delevered but content was lost and clients were asked to request a resent message from those senders whos messages were effected. The system has been altered such that this problem will not happen again.
outage of 50 minutes at 11:45pm due to faulty restart after
outage of 1 hr 16 minutes at 6:44am due to failed memory in
primary file server.
Sunday outage of 3 minutes at 5:32am to upgrade code.
Sunday outage of 14 minutes at 5:28am to upgrade file server
server hardware failure caused intermittent outages this week
as temporary solutions and permanent fully redundant mail servers
Saturday outage of 17 minutes at 5:22am to upgrade file server
Saturday outages (2) of 2 minutes each at 5:58am to upgrade
was unavailable for 60 minutes starting at 4:19pm due to an
upstream DNS server failure causing reverse lookups to fail.
No ICG Link equipment was off line.
worm attack on MS SQL caused spotty outages between 1:38am and
6:30am due to flooded firewall/routers. No ICG Link equipment
was off line.
Saturday outage of 1 hour 5 minutes at 5:24am to upgrade file
outage of 1 minute occured at 2:34pm for reconfiguring the web
with UUNET reliability and a BellSouth fiber cut in Texas degraded
general Internet health across the Southeast US. No ICG Link
equipment was off line.
outage of 7 minutes occurred starting at 10:24pm for restarting
outage of 1 minute occurred starting at 5:30pm for reconfiguring
the web servers.
outage of 13 minutes starting at 11:32PM to install a network
card to upgrade the primary file server.
outage of 39 minutes starting at 5:31PM to install new high-availability,
fully redundant firewall routers.
2:43PM, ATT router problems caused degradation across the Southeast.
It was resolved by Inflow shutting down their ATT link. No ICG
Link equipment was off line.
upgrade of our mail and DNS servers caused loss of DNS services
between 2:15AM and 10:15AM. Most clients would not have experienced
an outage due to cached DNS.
and Sprint had a problem causing sporadic outages across the
Southeast US that was resolved around noon. No ICG Link equipment
was off line.
NetMerchant was inoperative between the hours of 3:17PM and
4:59PM as a result of the IE5 / Mac upgrade issue.
upgrade to NetMerchant to fix problems for users with IE5 on
a MAC was installed causing users on WinME using NS 4.7 to get
a cookie related error.
Cybercash carts refused orders and other carts reported blank
information in log files and e-mail between 3:30PM and 10:55PM
due to an upgrade.
outage of 7 minutes, restored at 5:57AM as a result of
an unscheduled system restart.
planned, all service was down for 8 hours, restored at 8:00AM
as a result of the move of our servers to Inflow. During the
weekend, intermittent outages were reported as DNS changes propagated
across the Internet.