| 1 |
Quoth Christian Bricart (on Wednesday, the 7th of December): |
| 2 |
|
| 3 |
>> lo, |
| 4 |
>> |
| 5 |
>> On Tuesday 06 December 2005 21:54, Justin Krejci wrote: |
| 6 |
>> [snip] |
| 7 |
>>> /bin/touch /tmp/esync.update |
| 8 |
>>> /usr/sbin/esync |
| 9 |
>>> /bin/touch /tmp/esync.update |
| 10 |
>> |
| 11 |
> |
| 12 |
> speaking of "state" - wouldn't be /var/state/ the appropriate directory..? |
| 13 |
> The file (or directory within) should be owned by the portage group, as |
| 14 |
> only members of the portage group are able to sync anyway, right? |
| 15 |
|
| 16 |
I would definitely place it there. |
| 17 |
Files placed */tmp dir should are by definition (FHS) not to be expected |
| 18 |
to remain there between two runs of a program. |
| 19 |
|
| 20 |
But that's not why I am replying. |
| 21 |
|
| 22 |
Wouldn't the following suffice in the case of portage syncing? |
| 23 |
esync && touch /usr/portage/lastsync |
| 24 |
|
| 25 |
This file will carry a timestamp of when the last sync happened, and as |
| 26 |
rsync deletes this file, if it's missing that means that the sync |
| 27 |
encoutered problems and your portage tree is in an undefined state. |
| 28 |
|
| 29 |
Opinions? |
| 30 |
|
| 31 |
Peter |
| 32 |
-- |
| 33 |
"I do not think the way you think I think." |
| 34 |
-- Kai, last of the Brunnen G |
| 35 |
-- |
| 36 |
gentoo-server@g.o mailing list |
Archives