| 1 |
On 26/04/2013 01:42, William Kenworthy wrote: |
| 2 |
> Does anyone know a good guide to using time sync in VM's, for both |
| 3 |
> windows and linux (gentoo) guests using libvirt? Especially for guests |
| 4 |
> that are resumed, or the whole virtualisation system is hibernated? (ntp |
| 5 |
> refuses to resync after guest pause/save/restore/resume (known problem), |
| 6 |
> even with "tinker panic 0" |
| 7 |
|
| 8 |
|
| 9 |
That's not a bug, it's by design. |
| 10 |
|
| 11 |
If ntpd detects the clock is out by more than X seconds [1], it will not |
| 12 |
try to correct the difference, concluding that something is wrong and a |
| 13 |
human must decide. It can't easily tell the difference between a resumed |
| 14 |
guest (or even that it was resumed at all) and a severe problem. |
| 15 |
|
| 16 |
We fixed this by taking the easy route of least resistance; |
| 17 |
|
| 18 |
1. run ntpdate on startup/restart once before ntpd starts |
| 19 |
2. start ntpd as normal |
| 20 |
3. a colleague wrote a $MAGIC_HOOK to detect resumed guests that runs |
| 21 |
ntpdate once |
| 22 |
|
| 23 |
True, it's a brutal solution and uses a baseball bat where some finesse |
| 24 |
might be less ugly, but it suits our needs just fine. |
| 25 |
|
| 26 |
[1] I forget what X is and am too lazy to look it up. Is it 30 seconds |
| 27 |
or thereabouts? |
| 28 |
|
| 29 |
|
| 30 |
-- |
| 31 |
Alan McKinnon |
| 32 |
alan.mckinnon@×××××.com |
Archives