aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGravatar Joey Hess <joey@kitenet.net>2013-11-02 15:10:53 -0400
committerGravatar Joey Hess <joey@kitenet.net>2013-11-02 15:10:53 -0400
commit95e946a21e1960c9dcf63ad579c58434746295b2 (patch)
tree0e314b1eeadd660cc7d70510b0ac89a74912676e
parent6c84eaf5a01046a0fa175d75e7027f686e00a6ee (diff)
parentaa57be4c5cfb269f2b1afa541186f41f4e324d67 (diff)
Merge branch 'master' of ssh://git-annex.branchable.com
-rw-r--r--doc/bugs/Local_pairing_fails:_received_PairMsg_loop/comment_3_6345b174d04b6613c2c55a6ec9e50c21._comment16
-rw-r--r--doc/forum/lost_in_walkthrough....txt78
-rw-r--r--doc/install/cabal/comment_28_15951dd070a675300420eea137a28ef9._comment10
3 files changed, 104 insertions, 0 deletions
diff --git a/doc/bugs/Local_pairing_fails:_received_PairMsg_loop/comment_3_6345b174d04b6613c2c55a6ec9e50c21._comment b/doc/bugs/Local_pairing_fails:_received_PairMsg_loop/comment_3_6345b174d04b6613c2c55a6ec9e50c21._comment
new file mode 100644
index 000000000..a7eff6744
--- /dev/null
+++ b/doc/bugs/Local_pairing_fails:_received_PairMsg_loop/comment_3_6345b174d04b6613c2c55a6ec9e50c21._comment
@@ -0,0 +1,16 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.47"
+ subject="comment 3"
+ date="2013-11-02T19:02:55Z"
+ content="""
+There is something a little strange in the logs. It shows both Onyx and kubbie are sending PairReqs. Probably one of those is the local computer, logging its own broadcast messages which loop back to it. But normally in pairing, one side starts the pairing process with a PairReq, and the other side pops up an alert and continues the process with a PairAck. It's not normal for both sides to request that pairing start.
+
+Is it possible that you started pairing on both computers separately?
+
+I tried doing that with gnu and darkstar. First I started pairing on gnu. darkstar saw the pair request, but I ignored the alert message about that, and went and started a separate pairing process on darkstar. gnu never showed an alert message for that; it ignored darkstar's PairReqs since it was sending its own.
+
+It's unlikely you'd ignore the alert, but the same thing could happen if the two computers were not able to communicate over the network initially, and pairing were started on both separately. So neither gets a chance to see the other's PairReq and show the alert. Even when they came into communication, they'd each ignore the other's PairReq.
+
+So, that seems like a bug..
+"""]]
diff --git a/doc/forum/lost_in_walkthrough....txt b/doc/forum/lost_in_walkthrough....txt
new file mode 100644
index 000000000..89f4af026
--- /dev/null
+++ b/doc/forum/lost_in_walkthrough....txt
@@ -0,0 +1,78 @@
+I'm trying to follow the steps of the "walkthrough" but I'm experiencing the following issue: when
+I sync one repository and do "git annex get ." I don't get the files from the other repository.
+Here is the transcript of the steps I followed - I've put them in a script (ga.sh) so I can replay
+them and show them on the shell while executing.
+Basically I have two repositories, "/tmp/a/annex" and "/tmp/b/annex", the second cloned from
+the first. All the other steps are the same as in the walkthrough.
+-----------------------------------
+> bash -x ga.sh
++ cd /tmp
++ mkdir a
++ mkdir b
++ cd a
++ mkdir annex
++ cd annex
++ git init
+Initialized empty Git repository in /tmp/a/annex/.git/
++ git annex init a
+init a ok
+(Recording state in git...)
++ cd /tmp/b
++ git clone /tmp/a/annex
+Cloning into 'annex'...
+done.
+warning: remote HEAD refers to nonexistent ref, unable to checkout.
+
++ cd annex
++ git annex init b
+init b ok
+(Recording state in git...)
++ git remote add a /tmp/a/annex
++ cd /tmp/a/annex
++ git remote add b /tmp/b/annex
++ dd if=/dev/urandom of=first bs=1024 count=1
+1+0 records in
+1+0 records out
+1024 bytes (1.0 kB) copied, 9.9167e-05 s, 10.3 MB/s
++ dd if=/dev/urandom of=second bs=1024 count=1
+1+0 records in
+1+0 records out
+1024 bytes (1.0 kB) copied, 0.000241635 s, 4.2 MB/s
++ git annex add .
+add first (checksum...) ok
+add second (checksum...) ok
+(Recording state in git...)
++ git commit -am added
+[master (root-commit) 5078564] added
+ 2 files changed, 2 insertions(+)
+ create mode 120000 first
+ create mode 120000 second
++ git mv first e
++ mkdir x
++ git mv second x
++ git commit -m moved
+fix x/second ok
+(Recording state in git...)
+[master 422492d] moved
+ 3 files changed, 1 insertion(+), 1 deletion(-)
+ rename first => e (100%)
+ delete mode 120000 second
+ create mode 120000 x/second
++ cd /tmp/b/annex
++ git annex sync a
+(merging origin/git-annex into git-annex...)
+(Recording state in git...)
+commit
+ok
+git-annex: no branch is checked out
++ git annex get .
+-------------------
+
+The last "git annex get ." does not retrieve the files in /tmp/a/annex ... why?
+I guess the issue starts when cloning /tmp/a/annex where no commit was done.
+
+Emanuele
+
+PS: I'm using git v1.7.9.5 (ubuntu 12.04) and the latest git-annex static binary
+downloaded a few minutes ago from the git-annex website.
+
diff --git a/doc/install/cabal/comment_28_15951dd070a675300420eea137a28ef9._comment b/doc/install/cabal/comment_28_15951dd070a675300420eea137a28ef9._comment
new file mode 100644
index 000000000..19a570cae
--- /dev/null
+++ b/doc/install/cabal/comment_28_15951dd070a675300420eea137a28ef9._comment
@@ -0,0 +1,10 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.47"
+ subject="comment 28"
+ date="2013-11-02T17:40:18Z"
+ content="""
+Unfortunate about that. I've fixed it in git.
+
+(It would be best to file bug reports about build failures to avoid cluttering up this page.)
+"""]]