From 9c041ed475e60a3159e42bca5f40fee2be90cc78 Mon Sep 17 00:00:00 2001 From: Ilya_Shlyakhter Date: Thu, 4 Oct 2018 19:30:07 +0000 Subject: [PATCH] Added a comment --- .../comment_3_cf05eba2bcc10f48092334f9ce889863._comment | 8 ++++++++ 1 file changed, 8 insertions(+) create mode 100644 doc/git-annex-sync/comment_3_cf05eba2bcc10f48092334f9ce889863._comment diff --git a/doc/git-annex-sync/comment_3_cf05eba2bcc10f48092334f9ce889863._comment b/doc/git-annex-sync/comment_3_cf05eba2bcc10f48092334f9ce889863._comment new file mode 100644 index 0000000000..5a075b3ca8 --- /dev/null +++ b/doc/git-annex-sync/comment_3_cf05eba2bcc10f48092334f9ce889863._comment @@ -0,0 +1,8 @@ +[[!comment format=mdwn + username="Ilya_Shlyakhter" + avatar="http://cdn.libravatar.org/avatar/1647044369aa7747829c38b9dcc84df0" + subject="comment 3" + date="2018-10-04T19:30:07Z" + content=""" +Maybe, have a config option for the number of retries or the maximum time to keep trying? An important property of git-annex-sync is that, even though it does merging, it does not normally fail. It'd be good to keep that guarantee even in the face of concurrent access to the repo. Alternately, you could push to a new branch synced/uuid/branchname, and then when git-annex-sync is later done at the target repo locally, merge from each of the synced/*/branchname . +"""]]