[LEAPSECS] Hetzner mail to customers: 1 megawatt more power due to leap second
Carlos Alberto Lopez Perez
clopez at igalia.com
Thu Jul 5 09:46:07 EDT 2012
On 05/07/12 13:42, mike cook wrote:
> Le 05/07/2012 13:17, Carlos Alberto Lopez Perez a écrit :
>> The reason for this huge surge is the additional switched leap second
>> which can lead to permanent CPU load on Linux servers.
> Somewhat disingenuous. The statement should read :
> .....can lead to permanent CPU load on badly maintained Linux servers.
> as the bug was known and corrections were available for some time.
AFAIK *all* distributions with a kernel >2.6.26 were affected (that is
basically all major distros except RHEL5).
"" Hey guys: the next weekend we are going to issue a leap second, so
you must execute this script or run "date -s `date`" and restart the ntp
daemon when it happens, or as an alternative you can stop the ntp daemon
and start it after the leap second ""
is _not_ a fix nor a "correction"
This problem should be fixed where it lays: at the kernel level .
However this problem is recurring, since it happened something similar
with the previous leap second  and it was supposed to be fixed, and
the odds are that it will happen again (and again).
Computers are not designed to cope with random time drifts, and
certainly is not only Linux, but also other OSes or devices (specially
RT devices) that could have problems with this (now or on the future).
Just think about the havoc this leap second caused on Internet:
thousands of servers offline or misbehaving with the CPU at the 100%.
 I am pretty sure that this issue had a worldwide cost of several
millions of dollars.
Carlos Alberto Lopez Perez http://neutrino.es
Igalia - Free Software Engineering http://www.igalia.com
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 900 bytes
Desc: OpenPGP digital signature
Url : <http://six.pairlist.net/pipermail/leapsecs/attachments/20120705/554a4ca7/attachment.pgp>
More information about the LEAPSECS