summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGravatar Joey Hess <joey@kitenet.net>2014-03-05 18:01:35 -0400
committerGravatar Joey Hess <joey@kitenet.net>2014-03-05 18:01:35 -0400
commit962f34b501c1724cbfc7b18b680cfaea86469f24 (patch)
treeecd4bd1c2fadb8668e7dc54f8f82b9891867eaab
parent1db1469905cd08da2b4a2f5e310437ff5233f286 (diff)
parentd155a4f8b5103d8e2c8caad027f27a9140f34170 (diff)
Merge branch 'master' of ssh://git-annex.branchable.com
-rw-r--r--doc/bugs/assistant_eats_all_CPU/comment_21_8ce65a701604b9d13941844c62f62f23._comment14
-rw-r--r--doc/bugs/assistant_eats_all_CPU/comment_22_0d8de9a8e4b8e2ef3b9c7d839fbcad0c._comment12
2 files changed, 26 insertions, 0 deletions
diff --git a/doc/bugs/assistant_eats_all_CPU/comment_21_8ce65a701604b9d13941844c62f62f23._comment b/doc/bugs/assistant_eats_all_CPU/comment_21_8ce65a701604b9d13941844c62f62f23._comment
new file mode 100644
index 000000000..2e74010df
--- /dev/null
+++ b/doc/bugs/assistant_eats_all_CPU/comment_21_8ce65a701604b9d13941844c62f62f23._comment
@@ -0,0 +1,14 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.146"
+ subject="strace or it didn't happen"
+ date="2014-03-05T21:02:17Z"
+ content="""
+This bug is about a particular sequence of system calls. It is not about every possible way that git-annex can use CPU! If you think you have reproduced this bug, you need to include a strace demonstrating the problem, or you're just adding noise.
+
+@Thomas your log shows git-annex doing a git pull over XMPP. This will take CPU, as will any git pull. It is not related to this bug.
+
+@myownlittl the assistant has to scan your repository when it's started to find any files you have modified since the last time it was run. That is not related to this bug. However, I will probably be adding a config knob to disable the startup scan.
+
+@Oliver, you said you had seen this same bug on Ubuntu, but did not show a strace, so I don't know if you are really experiencing the same bug.
+"""]]
diff --git a/doc/bugs/assistant_eats_all_CPU/comment_22_0d8de9a8e4b8e2ef3b9c7d839fbcad0c._comment b/doc/bugs/assistant_eats_all_CPU/comment_22_0d8de9a8e4b8e2ef3b9c7d839fbcad0c._comment
new file mode 100644
index 000000000..ccaf726b7
--- /dev/null
+++ b/doc/bugs/assistant_eats_all_CPU/comment_22_0d8de9a8e4b8e2ef3b9c7d839fbcad0c._comment
@@ -0,0 +1,12 @@
+[[!comment format=mdwn
+ username="http://joeyh.name/"
+ ip="209.250.56.146"
+ subject="comment 22"
+ date="2014-03-05T21:08:34Z"
+ content="""
+@anarcat, it seems to me that you need to find a minumal way to reproduce this.
+
+What architecture, what specfic model of CPU?
+
+Can you reproduce it by running the assistant in a new, empty git annex repository? If not, find the difference between that repository and the repository it happens in. If so, it would probably make sense to build git-annex from source with most features disabled to see if one of them causes it, and if not, try to get it down to a minimal test case suitable to being submitted as a GHC bug.
+"""]]