[Pool] Off by a second leap second failures

Martin Burnicki martin.burnicki at burnicki.net
Sun Jul 12 09:48:23 UTC 2015


John Winters wrote:
> On 12/07/15 09:25, Martin Burnicki wrote:
> [snip]
>> Do you still have those syslog files and can check if this is the case?
> 
> Just restored them from the nightly backups, and I have a lot of egg on
> my face.
> 
> Yes, ntpd did exactly what was expected of it.  The trouble is, this is
> a fairly newly installed Debian Jessie system and something had snuck in
> which I didn't know about (and as far as I can see, isn't even configured).
> 
> systemd-timesyncd
> 
> ntpd and systemd-timesyncd seem to have spent the four hours after
> midnight squabbling over the system time, with ntpd setting it right,
> and then systemd-timesyncd immediately setting it wrong again.
> systemd-timesyncd's configuration file has the time servers commented
> out, and its man page is missing so I'll need to do some research to
> find out what it uses by default.

Please have a look at
https://github.com/systemd/systemd/issues/437

I really don't understand why they use the Google servers by default,
which do Google's leap smearing, and not just the pool servers.

> Time to disable another feature of systemd.

Systemd does everything better than the original tools. I just wait that
it also replaces the kernel. I'm sure it can do the tiny kernel tasks
much better than the kernel can. ;-))

Martin



More information about the pool mailing list