[darcs-users] non-atomic pull

[Aggelos Economopoulos]

On Saturday 11 October 2003 15:44, David Roundy wrote:
> On Sat, Oct 11, 2003 at 03:29:53PM +0300, Aggelos Economopoulos wrote:
> > Is this a known problem?
> >
> > aggelos at erwin:~/src/darcs/darcs> darcs pull
> > Pulling from http://abridgegame.org/repos/darcs...
> >
> > Fail: user error
> > Reason: Failed to download URL http://abridgegame.org/repos/darcs/_darcs/
> > patches/
> > refuse_to_setpref_an_invalid_pref_field.-David_Roundy_2droundy9abridgegam
> >e.org7-Sat_Oct_11_07.38.49_EDT_2003 error code: 52
> >
> > aggelos at erwin:~/src/darcs/darcs> darcs pull
> > Pulling from http://abridgegame.org/repos/darcs...
> > [refuse to setpref an invalid pref field.
> > David Roundy <droundy at abridgegame.org>**Sat Oct 11 07:38:49 EDT 2003]
> > Shall I pull this patch? [ynvaswh] y
> > So far so good... the merge succeeded.
> > Finished pulling.
> > aggelos at erwin:~/src/darcs/darcs>
>
> I'm afraid I don't quite follow what you quote above.  It seems that you
> ran pull once and got a network error of some sort, and then ran again, and
> everything worked fine.  Where's the problem?

Well, 52 is CURLE_GOT_NOTHING. You're assuming the http server incorrectly 
returned no data, but I was thinking of the race described below.

> > It seems that apply appends the recorded patch names to _darcs/inventory
> > _before_ writing the patch files in _darcs/patches.
>
> I presume that you mean pull, not apply here.  Pull does write the patches
> prior to appending to the inventory files.

No, not pull. Let's name A the 'darcs apply' process on your server, B your 
http server serving the requests of the 'darcs pull' on my machine. I was 
wondering if something like the following might have happened:

A -> write to inventory
A -> open(O_CREAT) new patch file
[A gets scheduled out]

B -> read inventory
[darcs pull issues request for the new patch]
B -> read from currently empty file

c -> write patch data

Although this is an unlikely race due to network latency, I can't confirm that 
apply does the right thing, so I just thought I'd ask.

> Are you using compression perhaps? When compression is used, the file write
> is done asynchronously, so in theory you could have inventory updated prior
> to the patch files being written.

Why would my inventory being out of sync have anything to do with the curl 
error? Can you explain it further?

Thanks,
Aggelos