Hi,
Setting up a build box here, lots of CPUs and RAM for Android
tinderbox and general development and I keep running into the
strangest errors.. following the instructions to the letter I get to
repo sync (with --trace to show what's going wrong) and at random,
several parts of the repo do not sync properly, as follows
: export GIT_DIR=/home/me/android/.repo/projects/bionic.git
: git fetch ~/android/.repo/projects/bionic.git/clone.bundle +refs/
heads/*:refs/remotes/aosp/* refs/tags/*:refs/tags/*
fatal: pack has bad object at offset 150360: inflate returned -5
error: index-pack died
: git fetch aosp --tags +refs/heads/*:refs/remotes/aosp/*
: export GIT_DIR=/home/me/android/.repo/projects/bootable/recovery.git
: git fetch /home/me/android/.repo/projects/bootable/recovery.git/
clone.bundle +refs/heads/*:refs/remotes/aosp/* refs/tags/*:refs/tags/*
error: inflate: data stream error (incorrect data check)
fatal: pack has bad object at offset 147211: inflate returned -3
error: index-pack died
: git fetch aosp --tags +refs/heads/*:refs/remotes/aosp/*
........ This is littered with the occasional Python backtrace:
Exception in thread Thread-150:
Traceback (most recent call last):
File "/usr/lib/python2.7/threading.py", line 552, in
__bootstrap_inner
self.run()
File "/usr/lib/python2.7/threading.py", line 505, in run
self.__target(*self.__args, **self.__kwargs)
File "/home/me/android/.repo/repo/subcmds/sync.py", line 186, in
_FetchHelper
current_branch_only=opt.current_branch_only)
........ And the occasional, slightly rarer in that it only happened
about 10 times today instead of 10,000, a simple hang of the repo sync
usually always after trace says it's doing or has done
: export GIT_DIR=/home/me/android/.repo/projects/docs/
source.android.com.git
: git pack-refs --all --prune 1>| 2>|
: load refs /home/me/android/.repo/projects/docs/
source.android.com.git
: git symbolic-ref -m manifest set to refs/tags/android-4.0.1_r1.2
refs/remotes/m/android-4.0.1_r1.2 refs/tags/android-4.0.1_r1.2 1>| 2>|
(and the terminal is just sitting there, it's responsive, it prints ^C
when I ctrl-C it and other SSH sessions are running just fine)
I'm at a loss. I've done some Googling and found many references to
the same kind of error, mostly about a year old or very very recently
(last two weeks) suggesting it is a common failure somehow, but the
suggestions to fix it I have tried so far simply haven't made any
difference (bumping http.postBuffer to something higher, disabling hw
checksumming on the ethernet card). The system is a 8-core Xeon X2250
with 32GB RAM and a 4TB disk array, works fine for everything else and
used to run virtual machines through Xen but it's now being dedicated.
The hardware checks out, nothing is broken, the array is fine and what
is going over the network seems fine.
I can reproduce the errors almost identically on a VMWare Workstation
instance of the same OS (Ubuntu 11.04 Natty Narwhal) at the same patch
level, and almost identical configuration (the only differences being
a couple of tools I have on the new box to help me out on a remote
terminal, like htop, screeny utilities and so on).
This is really putting a downer on the Android building. Has anyone
got any ideas?
--
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to android-developers@googlegroups.com
To unsubscribe from this group, send email to
android-developers+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/android-developers?hl=en
