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

This commit is contained in:
Joey Hess 2014-04-20 15:47:34 -04:00
commit dbe1f40f94
5 changed files with 83 additions and 0 deletions

View file

@ -0,0 +1,12 @@
[[!comment format=mdwn
username="http://joeyh.name/"
ip="209.250.56.191"
subject="comment 1"
date="2014-04-20T17:16:50Z"
content="""
From the git-annex terminal, try to run:
/data/data/ga.androidterm/lib/lib.start.so
Paste me any output..
"""]]

View file

@ -0,0 +1,22 @@
[[!comment format=mdwn
username="cbaines"
ip="86.166.14.171"
subject="comment 2"
date="2014-04-20T19:40:11Z"
content="""
When I run that, I get:
[[!format sh \"\"\"
/system/bin/sh: /data/data/ga.androidterm/lib/lib.start.so: not found
\"\"\"]]
Which makes some sense, as the file is not there. The following files can be found under /data/data/ga.androidterm:
[[!format sh \"\"\"
/data/data/ga.androidterm/cache/com.android.renderscript.cache/ <- empty directory
/data/data/ga.androidterm/lib/libga-androidterm4.so
/data/data/ga.androidterm/shared_prefs/ga.androidterm_preferences.xml
\"\"\"]]
I tried running libga-androidterm4.so, but I just got Segmentation fault back.
I also tried using logcat to see if I could see anything obvious going wrong when running the app for the first time after installation, but I could not see anything obvious in the logs (there was a lot of noise, so I might of missed something), will anything useful appear with the use of a filter?
"""]]

View file

@ -0,0 +1,25 @@
[[!comment format=mdwn
username="http://joeyh.name/"
ip="209.250.56.191"
subject="comment 1"
date="2014-04-20T17:11:02Z"
content="""
Is there one commit with this long message, or 12 thousand commits each adding another (recovery from race) to the pyramid?
<pre>
- Also safely handles a race that can occur if a change is being pushed
- into the branch at the same time. When the race happens, the commit will
- be made on top of the newly pushed change, but without the index file
- being updated to include it. The result is that the newly pushed
- change is reverted. This race is detected and another commit made
- to fix it.
</pre>
If there is only one message, then it must have tried 12k times to commit to the git-annex branch and each time something else pushed or commited to the git-annex branch and overwrote its commit. This seems statistically unlikely. (Also there's locking to prevent multiple local git-annex processes from committing at the same time.)
There have been a few other unexplained reports of this race detection code repeatedly triggering.
> \"shutdown the assistant in yet another panic\"
This implies some hasty, perhaps unusual shutdown method, and some unusual situation. I think you could tell me more about what was going on.
"""]]

View file

@ -0,0 +1,13 @@
[[!comment format=mdwn
username="http://joeyh.name/"
ip="209.250.56.191"
subject="comment 5"
date="2014-04-20T18:35:39Z"
content="""
You might want to try upgrading to msysgit 1.9.0, which is the one I am using for windows development.
So, your screenshot shows that git checkout -B annex/direct/master worked ok when you ran it, but then when git-annex init failed, it was running
the same command, only with a -q also. The -q is only quiet mode, shouldn't matter.
(The .t folder is created when running git-annex test, so not relevant.)
"""]]

View file

@ -0,0 +1,11 @@
[[!comment format=mdwn
username="http://joeyh.name/"
ip="209.250.56.191"
subject="comment 1"
date="2014-04-20T16:52:15Z"
content="""
Last time rsync seemed to fail like this it was really the remote git-annex-shell failing on an encoding problem:
[[!commit 0b12db64d834979d49ed378235b0c19b34e4a4d6]]
It seems I would need to see the transfer info files you moved out of the way to say more. Or you could copy the back, reproduce the problem, find the git-annex-shell command that is being run (using --debug), and see if you can run it on the remote system and reproduce the problem there without rsync in the picture, in order to get the actual error message.
"""]]