summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGravatar Joey Hess <joey@kitenet.net>2012-12-14 14:40:45 -0400
committerGravatar Joey Hess <joey@kitenet.net>2012-12-14 14:40:45 -0400
commite79f7fed376fa18ddac89bfc099f468fdef81baa (patch)
tree4054fd82cc34b3c5a1a71ec7d85626c8b3630082
parent04ef2614897c967dcd160e4fe50f31b8a84e5668 (diff)
parentddd0ee543d63577d56da92cd925c736f910afa3f (diff)
Merge branch 'master' of ssh://git-annex.branchable.com
-rw-r--r--doc/bugs/OSX_app_issues/comment_16_0b7cd3d5952c5abf36a89a68a4afc1e7._comment8
-rw-r--r--doc/bugs/OSX_app_issues/comment_17_c2de94a48e7958b9efffd89dda9144ff._comment59
-rw-r--r--doc/design/assistant/blog/day_107__memory_leak.mdwn2
-rw-r--r--doc/forum/Compression_in_special_remotes___40__specifically_S3__41____63__.mdwn1
4 files changed, 69 insertions, 1 deletions
diff --git a/doc/bugs/OSX_app_issues/comment_16_0b7cd3d5952c5abf36a89a68a4afc1e7._comment b/doc/bugs/OSX_app_issues/comment_16_0b7cd3d5952c5abf36a89a68a4afc1e7._comment
new file mode 100644
index 000000000..0b110ce52
--- /dev/null
+++ b/doc/bugs/OSX_app_issues/comment_16_0b7cd3d5952c5abf36a89a68a4afc1e7._comment
@@ -0,0 +1,8 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="4.153.8.117"
+ subject="comment 16"
+ date="2012-12-13T22:40:59Z"
+ content="""
+Today's daily build of the OSX app has a further change that *might* help. I removed the system library and frameworks from the bundle.
+"""]]
diff --git a/doc/bugs/OSX_app_issues/comment_17_c2de94a48e7958b9efffd89dda9144ff._comment b/doc/bugs/OSX_app_issues/comment_17_c2de94a48e7958b9efffd89dda9144ff._comment
new file mode 100644
index 000000000..83c2023db
--- /dev/null
+++ b/doc/bugs/OSX_app_issues/comment_17_c2de94a48e7958b9efffd89dda9144ff._comment
@@ -0,0 +1,59 @@
+[[!comment format=mdwn
+ username="https://www.google.com/accounts/o8/id?id=AItOawkfHTPsiAcHEEN7Xl7WxiZmYq-vX7azxFY"
+ nickname="Vincent"
+ subject="OS/X daily 2012-12-13"
+ date="2012-12-14T09:22:42Z"
+ content="""
+Thanks for the update - I tried again, similar results.
+
+same platform. installed image md5sum 1bb50b3ee5eda3cd7f4b4a70cdae1855
+
+Procedure was the same.
+ - download, bunzip2, mount, drag app to Applications.
+ - chrome had one tab open, iconified
+ - double-click application icon
+
+Chrome opens up and shows the config window. Type in the same path
+after checking that the final element of the path did not already exist.
+
+ ps aux|grep git
+ me 89194 99.0 0.0 2460884 4160 ?? R 8:03pm 0:12.58 git init --quiet /Users/me/me/annex/
+ me 89245 0.2 0.0 2423356 220 s001 R+ 8:07pm 0:00.00 grep git
+ me 89182 0.0 0.3 2668772 44208 ?? S 8:03pm 0:00.30 git-annex webapp -psn_0_55022710
+ me 89177 0.0 0.0 2433432 868 ?? S 8:03pm 0:00.00 /bin/sh /Applications/git-annex.app/Contents/MacOS/git-annex-webapp -psn_0_55022710
+
+I ran dtruss on the two processes of interest, including when I sent them kill -9 in case that showed anything of interest.
+Mail me if you need that but the gist is git init was doing
+
+ workq_kernreturn(0x1, 0x1019CC000, 0x0) = -1 Err#22
+ workq_kernreturn(0x1, 0x1019CC000, 0x0) = -1 Err#22
+ workq_kernreturn(0x1, 0x1019CC000, 0x0) = -1 Err#22
+ ...
+ workq_kernreturn(0x1, 0x1019CC000, 0x0) = -1 Err#22
+ dtrace: 2006687 drops on CPU 0
+ workq_kernreturn(0x1, 0x1019CC000, 0x0) = -1 Err#22
+
+and the git-annex webapp was doing stuff like
+
+ psynch_cvwait(0x7FD7F1418888, 0x2BE010002BF00, 0x600) = -1 Err#260
+ sigreturn(0x7FFF543DD810, 0x1E, 0x2) = 0 Err#-2
+ __pthread_canceled(0x0, 0x2BE010002BF00, 0x7FFF543DD8C8) = -1 Err#22
+ psynch_cvwait(0x7FD7F1418888, 0x2C8010002C900, 0x600) = -1 Err#260
+ sigreturn(0x7FFF543DD810, 0x1E, 0x2) = 0 Err#-2
+ __pthread_canceled(0x0, 0x2C8010002C900, 0x7FFF543DD8C8) = -1 Err#22
+ psynch_cvwait(0x7FD7F1418888, 0x2D0010002D100, 0x600) = -1 Err#260
+ sigreturn(0x7FFF543DD810, 0x1E, 0x2) = 0 Err#-2
+ __pthread_canceled(0x0, 0x2D0010002D100, 0x7FFF543DD8C8) = -1 Err#22
+ psynch_cvwait(0x7FD7F1418888, 0x2D7010002D800, 0x600) = -1 Err#260
+ sigreturn(0x7FFF543DD810, 0x1E, 0x10DCEDC00) = 0 Err#-2
+ __pthread_canceled(0x0, 0x2D7010002D800, 0x7FFF543DD8C8) = -1 Err#22
+ psynch_cvwait(0x7FD7F14189D8, 0x230100002400, 0x2300) = 0 0
+ read(0x6, \"\377\0\", 0x1000) = 1 0
+ setitimer(0x0, 0x116903E50, 0x0) = 0 0
+ write(0x7, \"\377\0\", 0x1) = 1 0
+ sigreturn(0x7FFF543DD810, 0x1E, 0x0) = 0 Err#-2
+ __pthread_canceled(0x0, 0x2BC010002BD00, 0x7FFF543DD8C8) = -1 Err#22
+
+which may be unrelated browser event loop stuff I guess.
+
+"""]]
diff --git a/doc/design/assistant/blog/day_107__memory_leak.mdwn b/doc/design/assistant/blog/day_107__memory_leak.mdwn
index 9482552df..e33008f97 100644
--- a/doc/design/assistant/blog/day_107__memory_leak.mdwn
+++ b/doc/design/assistant/blog/day_107__memory_leak.mdwn
@@ -1,6 +1,6 @@
More bugfixes today. The assistant now seems to have enough users that
they're turning up interesting bugs, which is good. But does keep me too
-busy to add many more bugs^Wcode.
+busy to add many more bugs\^Wcode.
The fun one today made it bloat to eat all memory when logging out of a
Linux desktop. I tracked that back to a bug in the Haskell DBUS library
diff --git a/doc/forum/Compression_in_special_remotes___40__specifically_S3__41____63__.mdwn b/doc/forum/Compression_in_special_remotes___40__specifically_S3__41____63__.mdwn
new file mode 100644
index 000000000..49f3d75ef
--- /dev/null
+++ b/doc/forum/Compression_in_special_remotes___40__specifically_S3__41____63__.mdwn
@@ -0,0 +1 @@
+I use git-annex mostly to archive my photos. They're generally large raw files and I am encrypting them going to the S3 special remote. Are there any plans to compress objects before encrypting them? It would save on S3 costs, only add some overhead on get and copy, and I don't think there would be any problems looking up data later.