summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGravatar http://joeyh.name/ <http://joeyh.name/@web>2014-07-05 20:34:39 +0000
committerGravatar admin <admin@branchable.com>2014-07-05 20:34:39 +0000
commit2889841026403e43bc6e6a0ae11cc3264be5d99b (patch)
tree46d9f58a5c32bf3d788cd6123674b2816bdd0e5a
parentdc7c9a5db72ff05d5eb835f2406f2626a33553df (diff)
Added a comment
-rw-r--r--doc/bugs/Assistant_merge_loop/comment_1_ccf46511b924f86b488dba25060baa06._comment14
1 files changed, 14 insertions, 0 deletions
diff --git a/doc/bugs/Assistant_merge_loop/comment_1_ccf46511b924f86b488dba25060baa06._comment b/doc/bugs/Assistant_merge_loop/comment_1_ccf46511b924f86b488dba25060baa06._comment
new file mode 100644
index 000000000..c5f9c2187
--- /dev/null
+++ b/doc/bugs/Assistant_merge_loop/comment_1_ccf46511b924f86b488dba25060baa06._comment
@@ -0,0 +1,14 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.55"
+ subject="comment 1"
+ date="2014-07-05T20:34:39Z"
+ content="""
+I am seeing some evidence of this in my own family's repo, where one node updated to 5.20140613 and started making series of empty commits with message \"merge refs/heads/synced/master\" and only 1 parent (so not really a merge).
+
+Quite likely [[!commit d6711800ad261fb4c37fc361bc84918d1e296bc4]] is at fault. Probably the fastForwardable check isn't quite right.
+
+This should only affect direct mode repositories. When only one node has the problem, it won't be bad, but if multiple nodes are doing this, their repos never converge and keep growing.
+
+Hmm, I think I have partially reproduced it with 2 direct mode repos, each having the other as a remote. `git annex sync` repeatedly in each does not add unncessary commits, but running the assistant in each does. In this particular case, it manages to converge eventually after several commits.
+"""]]