| 1 |
On Mon, 18 Jul 2016 22:21:22 -0400 waltdnes@××××××××.org wrote: |
| 2 |
> On Mon, Jul 18, 2016 at 11:27:09PM +0300, Andrew Savchenko wrote |
| 3 |
> > |
| 4 |
> > As I wrote earlier in this thread, ntp server is not a guarantee |
| 5 |
> > that such problems will not happen. If hardware clocked was |
| 6 |
> > significantly offset during boot, it may take several _hours_ for |
| 7 |
> > ntp to fix this via clock skew. Apparantly commit may be made |
| 8 |
> > during these several hours. |
| 9 |
> |
| 10 |
> I'm amazed that "robust linux servers" are deathly afraid of simply |
| 11 |
> setting the time, and being done with it. And while we're at it, if a |
| 12 |
> developer is doing development on a server machine, he may have other |
| 13 |
> problems to worry about. At home I occasionally manually run a script |
| 14 |
> that includes the 2 lines... |
| 15 |
|
| 16 |
I never said anything about "robust linux servers". If you think |
| 17 |
than only servers need a gradual clock slew instead of stepping, |
| 18 |
you are mistaken. |
| 19 |
|
| 20 |
> /usr/bin/sudo /usr/bin/openrdate -n -s ca.pool.ntp.org |
| 21 |
> /usr/bin/sudo /sbin/hwclock --systohc |
| 22 |
|
| 23 |
And if time delta is significant, the system may become broken |
| 24 |
this way. Congratulations. |
| 25 |
|
| 26 |
The gradual NTP time slew was not invented because people were lazy |
| 27 |
to run two simple commands. Actually it is just the opposite: |
| 28 |
setting system time via a single huge step is much easier to |
| 29 |
implement than a proper adjustment via a series of small time slews. |
| 30 |
For servers it is indeed important in many ways, including |
| 31 |
time stamp based cryptography as kerberos or database integrity. |
| 32 |
|
| 33 |
But desktops also do need a proper time adjustment: |
| 34 |
- Filesystems will not operate correctly with time stamps in |
| 35 |
future, in the best keys they will be marked/reported as needed a |
| 36 |
repair procedure. |
| 37 |
- Cron jobs may go broken too as chithanh mentioned already. |
| 38 |
- Video encoding is not happy with time shifts at all. While small |
| 39 |
predictable slews can be tolerated, large jumps will just broke the |
| 40 |
process. |
| 41 |
- System may become *vulnerable* because of time stamp based attack. |
| 42 |
Though it is not easy to use such behaviour, it is still possible. |
| 43 |
- and many more... |
| 44 |
|
| 45 |
Best regards, |
| 46 |
Andrew Savchenko |
Archives