summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGravatar Joey Hess <joeyh@joeyh.name>2015-10-02 12:19:02 -0400
committerGravatar Joey Hess <joeyh@joeyh.name>2015-10-02 12:19:02 -0400
commit1fa36ce902db246943e5ecce9666920478e25b34 (patch)
treeae529e746ecf4ba016e21dbd67f4a0081c57ebd3
parent837e6e829478d7a6516609ab22f562c8ef410104 (diff)
response
-rw-r--r--doc/forum/external_special_remote_protocol_and_threads/comment_1_f29c2b0f0f5e349e36eb47176e33498f._comment20
1 files changed, 20 insertions, 0 deletions
diff --git a/doc/forum/external_special_remote_protocol_and_threads/comment_1_f29c2b0f0f5e349e36eb47176e33498f._comment b/doc/forum/external_special_remote_protocol_and_threads/comment_1_f29c2b0f0f5e349e36eb47176e33498f._comment
new file mode 100644
index 000000000..8448052da
--- /dev/null
+++ b/doc/forum/external_special_remote_protocol_and_threads/comment_1_f29c2b0f0f5e349e36eb47176e33498f._comment
@@ -0,0 +1,20 @@
+[[!comment format=mdwn
+ username="joey"
+ subject="""comment 1"""
+ date="2015-10-02T16:14:20Z"
+ content="""
+Well, I can confirm that only one instance of any
+external special remote will be run per git-annex process currently.
+
+Actually, it's a bit more than that -- If multiple threads try to use the
+same external special remote at the same time, everything is necessarily
+serialized, and so using -J won't let multiple downloads happen at once
+from a single external special remote, although it may still usefuly
+parallelize amoung several remotes.
+
+Since that seems like something worth dealing with, perhaps by having a
+pool of external special remote processes, I don't feel comfortable making
+any promises about the behavior. And as you note, it's easy to get multiple
+procesess by running multiple git-annex commands, so that is something an
+external special remote needs to deal with anyway.
+"""]]