point to commit that may fix this
authorJoey Hess <joeyh@joeyh.name>
Mon, 14 Sep 2020 22:30:41 +0000 (18:30 -0400)
committerJoey Hess <joeyh@joeyh.name>
Mon, 14 Sep 2020 22:30:41 +0000 (18:30 -0400)
doc/bugs/concurrent_git-annex-copy_to_s3_special_remote_fails/comment_4_af2e6683949d6168ff4ca606085cf870._comment [new file with mode: 0644]

diff --git a/doc/bugs/concurrent_git-annex-copy_to_s3_special_remote_fails/comment_4_af2e6683949d6168ff4ca606085cf870._comment b/doc/bugs/concurrent_git-annex-copy_to_s3_special_remote_fails/comment_4_af2e6683949d6168ff4ca606085cf870._comment
new file mode 100644 (file)
index 0000000..ebb25ad
--- /dev/null
@@ -0,0 +1,13 @@
+[[!comment format=mdwn
+ username="joey"
+ subject="""comment 4"""
+ date="2020-09-14T21:32:05Z"
+ content="""
+Thinking about this, one thing in git-annex that might somehow involve
+unsafe memory access would be if Remote.S3's use of runResourceT returns
+a value that depends on a resource that then gets freed.
+
+[[!commit ddf963d0194acf9f6f059fa37f3e89e59d682de9]] should avoid all ways that might possibly have happened,
+so if anyone has a way to reproduce this reliably, it would be good to know
+if it still happens after that fix.
+"""]]