From fc5c1494b8a668a3b17a092373d9564c4af828db Mon Sep 17 00:00:00 2001 From: helmut Date: Mon, 2 Dec 2013 15:41:09 +0000 Subject: [PATCH] --- ...ncremental_fsck_should_not_use_sticky_bit.mdwn | 15 +++++++++++++++ 1 file changed, 15 insertions(+) create mode 100644 doc/bugs/incremental_fsck_should_not_use_sticky_bit.mdwn diff --git a/doc/bugs/incremental_fsck_should_not_use_sticky_bit.mdwn b/doc/bugs/incremental_fsck_should_not_use_sticky_bit.mdwn new file mode 100644 index 0000000000..4662703cda --- /dev/null +++ b/doc/bugs/incremental_fsck_should_not_use_sticky_bit.mdwn @@ -0,0 +1,15 @@ +### Please describe the problem. + +There are multiple problems that have the same cause: + + * When using a shared repository (core.sharedrepository = group), the directories that contain the actual objects may be owned by different users. In this case adding or removing the sticky bit is prohibited by the operating system. Thus shared repositories can only be incrementally fscked if all objects are owned by the same user. + + * It is not currently possible to run incremental fscks on the local repository and on a special remote at the same time, because both of them use the same flag space, the sticky bit. + +### What steps will reproduce the problem? + +Create a shared repository (core.sharedrepository = group), let a different user add an object and then try to fsck it. + +### What version of git-annex are you using? On what operating system? + +Debian's 4.20131106~bpo70+1