Oracle tripped up by 'leap second'
Save the Clock Tower
Database giant Oracle has issued fixes after its Cluster Ready Services (CRS) software failed to cope with the so-called “leap second” added by scientists at the end of 2008.
The Earth Orientation Centre is responsible for calculating when a leap second should be added or subtracted because the Earth doesn’t always orbit perfectly.
Scientists added a second on 31 December 2008, meaning the final countdown required 11 rather than 10 seconds to bring us into the New Year to allow for the time to hit 23:59:60.
But Oracle’s CRS couldn’t cope with the added Marty McFly-style second, which meant that many of its servers rebooted on their own just after midnight on New Year’s Day.
The firm issued a fix for the embarrassing glitch on Monday after many sys admins grumbled that their CRS nodes were rebooting.
Oracle said versions 10.1.0.2 to 22.214.171.124 of the Oracle Server Enterprise Edition, running on 64-bit Sun Solaris servers with CRS and Oracle patch sets 10.2.0.1 to 126.96.36.199 were affected by the cockup.
At the start of 2009 Microsoft suffered a similar Fail with its 2006 30GB Zune model because the internal clock driver couldn’t handle leap years, causing the iPod imitators to have a lie down when the clock struck midnight. ®
@ Philip Teale
Finally, someone who knows what they are talking about.
I was having a right old laugh reading some of these.
The reason for the earth slowing
Americans get fat and move south to Florida. Conservation of angular momentum means that the earth's rotation slows as a result.
please get your facts right
This CRS bug is solved in 10.2.0.4 and 11g - the latest version/patchset who is already some time out. Oracle Support subscribers can check "Note 759143.1 NTP leap second event causing CRS node reboot" on Metalink for more information.