...continued.

June 19th, 2003 10:55 PM CDT
At 5:57 PM CDT today, we had a power failure. The UPS equipment did its job until the power was restored at 7:45 PM. There was no interruption of any services.

June 1st, 2003 1:46 AM CDT
At 1:20 AM CDT, our mail server (mailhost.icompute.com) crashed during system maintenance. It was rebooted and restored to service at 1:30 AM CDT.

May 29th, 2003 7:44 PM CDT
At 7:12 PM CDT, our upstream link stopped working due to some Qwest maintenance activity. We were back on line at 7:24 PM CDT.

May 3rd, 2003 9:00 AM CDT
This morning between 8 AM and 8:45 AM, there were several brief inerruptions of our upstream connectivity, each lasting less than about a minute as we switched over to our new upstream connection. The new connection is close to twice the speed of our old internet link, providing a full megabit of bandwidth. We also took the opportunity to install some improved packet filtering on the main router that will improve our security and stability. (Please call if you have questions.)

April 25th, 2003 7:43 PM CDT
A new upstream connection is being installed at iCompute on or after May 1st. The switch to this new upstream connection will involve only a few seconds of downtime, and should be almost completely transparent to our users. After the switch, our available internet bandwidth (speed) will be twice what it is now. No changes are required of our users to take advantage of this change.

Maintenance Announcement:

4/22/03 - The maintenance previously scheduled for April 24th has been cancelled. The plan was to upgrade our upstream data connection from CAP to DMT (CAP and DMT are high-speed data line technologies commonly called xDSL.)

The inability to obtain sufficient comfort with Qwest's scheduling of the upgrade caused us to reconsider our approach. As a result, a new plan is being followed that will result in a downtime of only a few seconds some time in early May. At that time, our upstream connection is expected to be upgraded to 1 Megabit/second.

(cancelled) 4/24/03 - 8:00 AM - NOON CDT - On Thursday April 24th, between 8 AM and 12 noon Central daylight time, we will be doing maintenance on our upstream connection to the internet.

The work to be performed will be to convert our current "CAP" DSL circuit to a "DMT" circuit, which is a more modern technology that is better supported by our circuit provider (Qwest).

Unfortunately, we were not able to more precisely schedule the time that our upstream link will be converted. We expect that the conversion will take 5-10 minutes, and we will lose our upstream connectivity during the conversion. While the upstream connection is unavailable, none of our services will be accessible. This includes our web servers, ftp, e-mail, and shell access.

If this is a hardship for any of you, or you have questions, please contact us. We apologize in advance for the interruption in service.

4/13/03 - 6:00 AM - CDT - The mail and DNS server was shut down for planned maintenance. We upgraded the mail server hardware to a much more powerful machine, with 80 GBytes of disk space, and the mail server software to EIMS 3.2. The server was back on line by 6:34 a.m. No mail was lost during this procedure. All other services were unaffected.

4/10/03 - 12:39 AM CDT - Our email server crashed at 12:39 a.m. It was restarted at 8:20 a.m. The monitoring system did not catch this, due to an administrative error, hence the delay in restarting. No mail has been lost during this down time. We apologize for the inconvenience this downtime may have caused.

Maintenance Announcement:

on 4/13/03 - 6:00 AM - 7:00 AM CDT - We will be taking system time to upgrade the mail server. The mail server will be unreliable or unavailable during this time. It is expected that all other services will be unaffected. It is possible that we will take the opportunity to do maintenance on other portions of the network as well. If so, it will be posted here as far in advance of the 13th as possible.

3/15/03 - 7:00 PM CST - An internet "attack" that had been growing in intensity for about a month was suppressed by the addition of filters on the main upstream router. The filters disallow the use of the broadcast addresses on our network for certain kinds of traffic. While adding and debugging these filters, there were several brief outages of upstream connectivity, none lasting more than about 2 minutes between 7 PM and midnight on March 15th. We will be adding additional filters to preclude other similar attacks.

2/23/03 - 6:00 AM CST - All of the equipment at iCompute.com was powered down to install a new UPS. Shutdown began at 6:00 a.m. All services were restored by 6:20 a.m.

Maintenance Announcement:

on 2/23/03 - 6:00 AM CST - All of the equipment at iCompute.com will be powered down to replace the UPS that failed on February 12th. We expect this to take up to 45 minutes. Please contact us if this down time will present a problem for you, our customers.

2/13/03 - 1:25 PM CST Fallout from the power failure, plus a failing modem has caused all of our modems to be unresponsive this morning. The problem should now be corrected, although we are now running with fewer dial-in lines until the failed modem can be replaced.

2/12/03 - 10:05 PM CST Our UPS (Uninterruptible Power Supply - sometimes called a battery backup) failed. Most of the systems at iCompute crashed, including the web and mail servers ("mercy" and "hope"). For approximately 20 minutes, no services were available. Everything was back up by about 10:25 PM, but we will be operating without a UPS until a replacement can be installed. An additional interruption of service will be required to install the replacement UPS. We apologize for the interruption of service.

2/12/03 - 11:00 AM CST One of our modems developed a malfunction yesterday afternoon around 3:00 pm CST. The modem was taken offline this morning at 10:15. During the period when the modem was malfunctioning, our dialup users may have experienced intermittent difficulties in establishing a connection.

2/1/03 - 5:30 AM CST At approximately 5:30 AM this morning we experienced a brief interruption of our upstream connectivity due to some planned maintenance at our upstream provider. We believe the downtime lasted a couple of minutes.

Click here for older news (October 2001 to December 2002)

Policies:

Any significant downtimes will be posted on this page.

If the downtime is unplanned, it will be posted here as soon as possible after the fact.

If the downtime is planned, it will be posted here at least one week in advance. Downtimes will be planned on Sunday morning between 6 and 8 AM whenver possible. Urgent, significant planned downtimes that cannot be scheduled a week in advance will still be posted here. If we deem it necessary, we will also send out e-mail to our customers with notification for the downtime as far in advance as possible.

-sysadmin