System Downtime & Outages

This is a simple page which is designed to record outages.

Sunday 10th May 2009

We were affected by an outage which affected our hosting provider generally. This is described at the Bytemark forums:

Once the Bytemark network came back our guests were back - there were no reboots involved, just a lack of external connectivity.

Friday 5th September 2008

This outage was announced in advance, but is listed here for completeness. (Mostly because the outage was announced only a couple of hours in advance.)

The host machine was brought down so that it could have the memory increased from 4Gb to 8Gb.

Friday 16th November 2007

One outage of approximately 30 minutes due to ISP network outage.

Saturday 13th October 2007

Two outages - one planned and one not.

The server was initially taken offline for approximately 10 minutes to perform a kernel upgrade. This was planned, and announced a week in advance.

Later the server appeared to lose all network connectivity. This coincided with the following output from dmesg:

xen_net: Memory squeeze in netback driver.
printk: 4 messages suppressed.
xen_net: Memory squeeze in netback driver.
printk: 4 messages suppressed.
xen_net: Memory squeeze in netback driver.

(Solution: Login to the xen shell and restart one guest domain, via 'reboot'. All the rest became pingable...)

I recieved a text message from one of our users alerting me of the outage and restored service approximately ten minutes later. Downtime unkown.

Sunday 7th October 2007

The server was taken offline for approximately 10 minutes to perform an upgrade of the xen-utils package.

This outage was anticipated and an announcement was sent a week in advance.

Total downtime upon this host in the order of 30 minutes so far in 2007.