Merge branch 'master' of ssh://git-annex.branchable.com

14 files changed, 163 insertions, 0 deletions

diff --git a/doc/assistant/remote_sharing_walkthrough/comment_4_978fab3cd165b4ca245e32fc48cf0970._comment b/doc/assistant/remote_sharing_walkthrough/comment_4_978fab3cd165b4ca245e32fc48cf0970._comment
new file mode 100644
index 000000000..ab0b19160
--- /dev/null
+++ b/doc/assistant/remote_sharing_walkthrough/comment_4_978fab3cd165b4ca245e32fc48cf0970._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.246"
+ subject="comment 4"
+ date="2013-11-12T18:18:16Z"
+ content="""
+You can easily use a removable drive as a transfer repository to sync two computers that have no network connection. Just use the webapp to add the drive on one computer. The drive will be set up as a transfer repository by default. The webapp will automatically start copying all your files to it. Then you can disconnect the drive, bring it to the other computer, and repeat the process. Everything from the first computer will then sync over from the drive to the second computer. And repeat moving the drive back and forth to keep things in sync.
+"""]]
diff --git a/doc/assistant/remote_sharing_walkthrough/comment_4_d7e879f7b098964040df2e27a18eda72._comment b/doc/assistant/remote_sharing_walkthrough/comment_4_d7e879f7b098964040df2e27a18eda72._comment
new file mode 100644
index 000000000..71f57e9b3
--- /dev/null
+++ b/doc/assistant/remote_sharing_walkthrough/comment_4_d7e879f7b098964040df2e27a18eda72._comment
@@ -0,0 +1,18 @@
+[[!comment format=mdwn
+ username="https://www.google.com/accounts/o8/id?id=AItOawmkXtBdMgE1d9nCz2iBc4f85xh4izZ_auU"
+ nickname="Ulrich"
+ subject="Using a portable drive as another transfer device?"
+ date="2013-11-12T16:52:12Z"
+ content="""
+I try to understand how to setup git-annex for the following use case:
+
+Two computers, that are paired via remote sharing, using some cloud repository for transfer, and a local NAS for backups.
+
+These two computers are sometimes in the same network, sometimes in different networks, and sometimes even without network at all. From what I read, it should be possible to bypass the cloud when these two machines are on the same network, which sounds great.
+
+Would it be possible to use a portable drive as \"another link\" between these two computers that can be used to sync them even if there is no network between them?
+
+And as you write, if the pairing has been set up manually, then everything is fine - so could it be that it is really easy and only necessary to setup the git-annex on the local drive as an additional remote on both (or only one?) machine?
+
+thanks for any insight!
+"""]]
diff --git a/doc/bugs/git-annex_broken_on_Android_4.3/comment_29_7b3fbe7e38f637fcea511441ac243d93._comment b/doc/bugs/git-annex_broken_on_Android_4.3/comment_29_7b3fbe7e38f637fcea511441ac243d93._comment
new file mode 100644
index 000000000..1cd4c356f
--- /dev/null
+++ b/doc/bugs/git-annex_broken_on_Android_4.3/comment_29_7b3fbe7e38f637fcea511441ac243d93._comment
@@ -0,0 +1,10 @@
+[[!comment format=mdwn
+ username="https://www.google.com/accounts/o8/id?id=AItOawmqz6wCn-Q1vzrsHGvEJHOt_T5ZESilxhc"
+ nickname="Sören"
+ subject="APK now works too"
+ date="2013-11-12T09:10:49Z"
+ content="""
+I got the same results as Michael. The output of hello-nocopy is a longer list of environment variables.
+Even better, the apk from the autobuilder now seems to work fine as well, including the webapp.
+Great work! Thanks for digging into this.
+"""]]
diff --git a/doc/bugs/git-annex_broken_on_Android_4.3/comment_30_26c04584c3c6dacf59e1b6c82042c97c._comment b/doc/bugs/git-annex_broken_on_Android_4.3/comment_30_26c04584c3c6dacf59e1b6c82042c97c._comment
new file mode 100644
index 000000000..3409b9c4a
--- /dev/null
+++ b/doc/bugs/git-annex_broken_on_Android_4.3/comment_30_26c04584c3c6dacf59e1b6c82042c97c._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="https://www.google.com/accounts/o8/id?id=AItOawnegApaT6kTI0Gxy9SNmI9Og-k_fC_aS7Y"
+ nickname="Michael Alan"
+ subject="Yep, the env was printed"
+ date="2013-11-12T10:18:56Z"
+ content="""
+I would recognize the output of a Show instance anywhere.
+"""]]
diff --git a/doc/devblog/day_54__android_bisection_minions/comment_1_bea8fbe2b87d4a4865b92fa796298fa0._comment b/doc/devblog/day_54__android_bisection_minions/comment_1_bea8fbe2b87d4a4865b92fa796298fa0._comment
new file mode 100644
index 000000000..0f8e58827
--- /dev/null
+++ b/doc/devblog/day_54__android_bisection_minions/comment_1_bea8fbe2b87d4a4865b92fa796298fa0._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="https://www.google.com/accounts/o8/id?id=AItOawmW0kg4uiMIhSHeVuvJFyo2VYMl7Qoej0s"
+ nickname="Chris"
+ subject="Woot!"
+ date="2013-11-12T15:52:48Z"
+ content="""
+Thanks for getting this fixed.  I'm about to go on a long plane ride, and this will make it a lot easier for me to put movies on my tablet.
+"""]]
diff --git a/doc/forum/Git_annex___39__corrupting__39___itself/comment_1_bcf50a215e2f8771e098aadfff4c300c._comment b/doc/forum/Git_annex___39__corrupting__39___itself/comment_1_bcf50a215e2f8771e098aadfff4c300c._comment
new file mode 100644
index 000000000..2fe95ab33
--- /dev/null
+++ b/doc/forum/Git_annex___39__corrupting__39___itself/comment_1_bcf50a215e2f8771e098aadfff4c300c._comment
@@ -0,0 +1,43 @@
+[[!comment format=mdwn
+ username="https://www.google.com/accounts/o8/id?id=AItOawlEXAlu8qKv_FHYOv-uEohdGgMY_Lw-Ges"
+ nickname="Jonas"
+ subject="Just happened for me as well."
+ date="2013-11-12T08:10:56Z"
+ content="""
+This was the first search result when I googled the error message, I'm not sure what has happened (or if it is a problem since I don't change my files that often).
+
+git-annex version: 4.20131101
+
+Also running git annex with one ssh remote ( laptop1 & laptop2 synching through a ssh remote server ).
+
+Got a bunch of 
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/git-annex does not point to a valid object!
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/master does not point to a valid object!
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/git-annex does not point to a valid object!
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/master does not point to a valid object!
+when I tried a manual git annex sync, and noticed that I had a bunch of those in the log in the webapp as well... 
+
+The gui doesn't show that anything is wrong, and looking at the log it seems that things are atleast uploading
+
+    [2013-11-12 09:05:46 CET] Committer: Committing changes to git
+    [2013-11-12 09:05:46 CET] Pusher: Syncing with born 
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/git-annex does not point to a valid object!
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/master does not point to a valid object!
+    Everything up-to-date
+
+    FILE
+
+       32768  42%    0.00kB/s    0:00:00  
+       76972 100%   42.16MB/s    0:00:00 (xfer#1, to-check=0/1)
+
+    sent 77056 bytes  received 31 bytes  51391.33 bytes/sec
+    total size is 76972  speedup is 1.00
+    [2013-11-12 09:05:48 CET] Transferrer: Uploaded FILE
+    [2013-11-12 09:05:48 CET] Pusher: Syncing with born 
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/git-annex does not point to a valid object!
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/master does not point to a valid object!
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/git-annex does not point to a valid object!
+    error: refs/synced/dbb70cea-4b83-416f-93af-e59082c4d633/master does not point to a valid object!
+    To ssh://me@born/annex
+      54e04cf..333cf10  git-annex -> synced/git-annex
+"""]]
diff --git a/doc/forum/Git_annex___39__corrupting__39___itself/comment_3_ab062b1df3b55fd49852a6220c98249e._comment b/doc/forum/Git_annex___39__corrupting__39___itself/comment_3_ab062b1df3b55fd49852a6220c98249e._comment
new file mode 100644
index 000000000..76e9d00de
--- /dev/null
+++ b/doc/forum/Git_annex___39__corrupting__39___itself/comment_3_ab062b1df3b55fd49852a6220c98249e._comment
@@ -0,0 +1,10 @@
+[[!comment format=mdwn
+ username="http://id.clacke.se/"
+ nickname="clacke"
+ subject="comment 3"
+ date="2013-11-12T17:19:32Z"
+ content="""
+I have been getting this too. Here is what I have done to fix it:
+
+[[tips/recovering_from_a_corrupt_git_repository]]
+"""]]
diff --git a/doc/forum/assistant__44___direct_mode__44___large_files__44___command_line_-_how_do_these_play_together__63__/comment_2_58b1af497cab132acb28cb5f9283ec2a._comment b/doc/forum/assistant__44___direct_mode__44___large_files__44___command_line_-_how_do_these_play_together__63__/comment_2_58b1af497cab132acb28cb5f9283ec2a._comment
new file mode 100644
index 000000000..45cec4db5
--- /dev/null
+++ b/doc/forum/assistant__44___direct_mode__44___large_files__44___command_line_-_how_do_these_play_together__63__/comment_2_58b1af497cab132acb28cb5f9283ec2a._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="https://www.google.com/accounts/o8/id?id=AItOawmkXtBdMgE1d9nCz2iBc4f85xh4izZ_auU"
+ nickname="Ulrich"
+ subject="Thanks"
+ date="2013-11-12T08:57:36Z"
+ content="""
+Ok, thanks for the explanation. I still have to figure out what happens when I annex drop files on a direct repository, and how well all this plays with the Finder on Mac…
+"""]]
diff --git a/doc/tips/recovering_from_a_corrupt_git_repository/comment_2_e98df7275bb032308bb87e3607bdde32._comment b/doc/tips/recovering_from_a_corrupt_git_repository/comment_2_e98df7275bb032308bb87e3607bdde32._comment
new file mode 100644
index 000000000..c018c4ca5
--- /dev/null
+++ b/doc/tips/recovering_from_a_corrupt_git_repository/comment_2_e98df7275bb032308bb87e3607bdde32._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="http://id.clacke.se/"
+ nickname="clacke"
+ subject="comment 2"
+ date="2013-11-12T16:14:50Z"
+ content="""
+I am using git-annex 20131106 and I tried `git annex repair` first. It seems git is making assumptions that if I have object A then I must have object B that A depends on. Or maybe it freaks out because the object is not missing, just full of zeroes. I haven't done any analysis on exactly what situation causes this. When I have time, I will.
+"""]]
diff --git a/doc/tips/recovering_from_a_corrupt_git_repository/comment_3_11bece6dfac090edbcd783b266c482a3._comment b/doc/tips/recovering_from_a_corrupt_git_repository/comment_3_11bece6dfac090edbcd783b266c482a3._comment
new file mode 100644
index 000000000..63f443a31
--- /dev/null
+++ b/doc/tips/recovering_from_a_corrupt_git_repository/comment_3_11bece6dfac090edbcd783b266c482a3._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="http://id.clacke.se/"
+ nickname="clacke"
+ subject="comment 3"
+ date="2013-11-12T16:17:42Z"
+ content="""
+Luckily, so far the objects getting corrupted have been in past commits (maybe only in packfiles?), not the latest ones, so I have been able to recover even though HEAD has been unique to the local repo.
+"""]]
diff --git a/doc/tips/recovering_from_a_corrupt_git_repository/comment_4_86e99017f7585ac2f76753051214637e._comment b/doc/tips/recovering_from_a_corrupt_git_repository/comment_4_86e99017f7585ac2f76753051214637e._comment
new file mode 100644
index 000000000..8d454fc3e
--- /dev/null
+++ b/doc/tips/recovering_from_a_corrupt_git_repository/comment_4_86e99017f7585ac2f76753051214637e._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="http://id.clacke.se/"
+ nickname="clacke"
+ subject="comment 4"
+ date="2013-11-12T17:10:06Z"
+ content="""
+Hm. This is bad. My hunch that btrfs was the culprit seems to have been wrong. After having switched things around, along with lots of `git annex sync` in various places, I now have a corrupt repo on ext4. It must be git or git-annex that does something wrong.
+"""]]
diff --git a/doc/tips/recovering_from_a_corrupt_git_repository/comment_6_c8953732ce353cdf0c4fb81ddc98c04a._comment b/doc/tips/recovering_from_a_corrupt_git_repository/comment_6_c8953732ce353cdf0c4fb81ddc98c04a._comment
new file mode 100644
index 000000000..eeb3ceef2
--- /dev/null
+++ b/doc/tips/recovering_from_a_corrupt_git_repository/comment_6_c8953732ce353cdf0c4fb81ddc98c04a._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.246"
+ subject="comment 6"
+ date="2013-11-12T18:10:21Z"
+ content="""
+If you can reliably corrupt a git repository, it's highly likely your hardware (disk or memory) is broken.
+"""]]
diff --git a/doc/tips/recovering_from_a_corrupt_git_repository/comment_6_d0da84df0241dc6ccf0aa0c7598917df._comment b/doc/tips/recovering_from_a_corrupt_git_repository/comment_6_d0da84df0241dc6ccf0aa0c7598917df._comment
new file mode 100644
index 000000000..16b2329b9
--- /dev/null
+++ b/doc/tips/recovering_from_a_corrupt_git_repository/comment_6_d0da84df0241dc6ccf0aa0c7598917df._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="http://id.clacke.se/"
+ nickname="clacke"
+ subject="comment 6"
+ date="2013-11-12T17:22:01Z"
+ content="""
+This is my symptom too: [[forum/Git annex 'corrupting' itself]]
+"""]]
diff --git a/doc/tips/recovering_from_a_corrupt_git_repository/comment_7_addf49556e4c33d55a41c393f519d0a4._comment b/doc/tips/recovering_from_a_corrupt_git_repository/comment_7_addf49556e4c33d55a41c393f519d0a4._comment
new file mode 100644
index 000000000..5047bcca4
--- /dev/null
+++ b/doc/tips/recovering_from_a_corrupt_git_repository/comment_7_addf49556e4c33d55a41c393f519d0a4._comment
@@ -0,0 +1,10 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.246"
+ subject="comment 7"
+ date="2013-11-12T18:12:49Z"
+ content="""
+> I tried git annex repair first. It seems git is making assumptions that if I have object A then I must have object B that A depends on. Or maybe it freaks out because the object is not missing
+
+`git annex repair` is supposed to deal with these situations. If it fails to fix such a broken repository, please file a detailed bug report, ideally with a link to a copy of the repository.
+"""]]