impact on git of changing linux o/s system-date frequently

Neal Kreitzinger· Jan 21, 2011, 00:01 UTC · lore

What are the implications of linux o/s system-date changes on git?

We plan on having test VM's on which the linux o/s system-date (linux "date" command) is changed several times to test time lapse scenarios, e.g. go forward 30 days in the future. We use git to pull updates from the development machine to mirror repos and cloned repos on the test VM, and to create test areas (local repos) on the test VM. We do not use git-push or git-commit on the test VM at this time.

v/r, Neal

David Brown· Jan 21, 2011, 07:13 UTC · re: Neal Kreitzinger · lore

Re: impact on git of changing linux o/s system-date frequently

On Thu, Jan 20 2011, Neal Kreitzinger wrote:

Show 8 quoted lines

> What are the implications of linux o/s system-date changes on git?
>
> We plan on having test VM's on which the linux o/s system-date (linux "date" 
> command) is changed several times to test time lapse scenarios, e.g. go 
> forward 30 days in the future.  We use git to pull updates from the 
> development machine to mirror repos and cloned repos on the test VM, and to 
> create test areas (local repos) on the test VM.  We do not use git-push or 
> git-commit on the test VM at this time.

Without making commits on the machine, there won't be any history being created with misordered dates, so there shouldn't be any issues there.

The index should also be pretty resiliant to time warping as well, and if you aren't modifying any of the git managed files, I don't think you can provoke anything.

If you're doing edits of files, there's a slight possibility the edited version could end up with the same ctime/mtime as the cache in the index, and git would miss the change. Again, it doesn't sound like you are doing this.

David

-- 
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.