On Mon, Sep 22, 2008 at 8:23 PM, David Roundy <[EMAIL PROTECTED]> wrote: > On Mon, Sep 22, 2008 at 8:22 AM, Dmitry Kurochkin > <[EMAIL PROTECTED]> wrote: >> On Thu, Sep 18, 2008 at 11:11 PM, David Roundy <[EMAIL PROTECTED]> wrote: >>> On Thu, Sep 18, 2008 at 06:04:04PM +0400, Dmitry Kurochkin wrote: >>>> Amended version of the patch. Now there is no conflict in pull and >>>> wibble case added. >>> >>> Actually, it turns out that this test fails (at least sometimes) on an >>> nfs file system with a locked-file error: >>> >>> Running issue27.sh ... FAILED! >>> Output from failed issue27.sh: >>> >>> rm -rf temp1 temp2 >>> mkdir temp1 >>> cd temp1 >>> darcs init >>> echo first > a >>> darcs add a >>> darcs record --pipe --all --patch-name=first <<EOF >>> What is the date? Who is the author? What is the log? >>> Finished recording patch 'first' >>> >>> echo first > b >>> darcs add b >>> darcs record --pipe --all --patch-name=first <<EOF >>> What is the date? Who is the author? What is the log? >>> Finished recording patch 'first' >>> >>> darcs get . ../temp2 >>> darcs: >>> ./_darcs/patches/20080918223706-f64cd-343d3ba892f2dc113953a5ef6e792eda07ae1b04.gz: >>> copyFile: resource busy (file is locked) >>> >>> This is an odd error, and may just be a server problem, so I'll try >>> pushing again. If it fails again, then this is a new issue... >> >> David, did you try to push this again? Is the error still there? > > Yes, the issue27.sh test still fails for me. I assume it's either nfs > that makes the difference, or ghc 6.6.
Strange. So get from one local repo to another fails? I wonder why other test cases pass. Can you try changing get to init+pull? I believe this is a new bug. Unfortunately I do not have nfs or ghc6.6 to reproduce this. Regards, Dmitry _______________________________________________ darcs-users mailing list darcs-users@darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users
