我已经将 Gitorious 从 2.x 更新到 3。经过一些麻烦，我可以拉，推等。但活动流不会更新，也不会创建仓库（"正在创建新仓库"仍然存在）。
> ./script/diagnose
Both MergeRequest and its :status machine have defined a different default for "status". Use only one or the other for defining defaults to avoid unexpected behaviors.
everything healthy? false
git user ok? true
gitorius front web page is up? false
rails process present & owned by git user? true
atleast one gitorious account present? true
repo base dir present, owned by git user? true
tarball dirs present, owned by git user? true
git user has ~/.ssh/authorized_keys file? true
ssh deamon is up? true
git daemon is up? true
poller is up? false
mysql is up? true
ultrasphinx is up? true
queue service is up? true
memcached is up? true
enough disk free? true
enough RAM free? true
healthy cpu load average? true
uptime:
13:40:41 up 81 days, 17:17, 1 user, load average: 1.45, 0.76, 0.37
free:
total used free shared buffers cached
Mem: 2048 1353 694 0 0 289
-/+ buffers/cache: 1063 984
Swap: 0 0 0
vmstat:
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
0 0 0 711344 0 296864 0 0 4 7 0 5 14 3 83 0 0
df:
Filesystem Size Used Avail Use% Mounted on
/dev/vzfs 25G 15G 11G 59% /
none 1.0G 4.0K 1.0G 1% /dev
并且轮询器脚本不存在：
> RAILS_ENV=production script/poller start
zsh: datei oder Verzeichnis nicht gefunden: script/poller
我的文件：
> ls script bin
bin:
bundle create-user git-proxy micro-tests rake restart search_engine snapshot update-ui varnish
console gitorious gts-whats-new migrate repo_housekeeping restore setup.rb unicorn upgrade-gitorious3-config
script:
create_admin destroy_invalid_events git-daemon git-proxy regenerate_assets server suspend_user upgrade_to
create_user diagnose gitorious rails regenerate_ssh_keys shard_git_repositories_by_hash test_ldap_connection
create_wikis fixup_hooks gitorious-config rake-wrapper restart_unicorn snapshot undo_last_upgrade
轮询器脚本在哪里以及如何在 v3 中启动它？这是问题的原因吗？
根据我的Google群组帖子中的答案，您可以（或不）通过以下方式启动轮询器
sudo service resque-worker start
如果您通过 CE 安装程序安装了 Gitorious。就我而言，设置与ce安装程序和自定义修复工作混合在一起。所以这对我不起作用（服务未知）。
但是resque-worker导致/etc/init/resque-worker.conf.
如果你看看里面/etc/init/resque-worker.conf：
description "Run a Resque worker on all queues"
author "Marius Mårnes Mathiesen <marius@gitorious.com>"
start on started rc RUNLEVEL=[35]
stop on runlevel [06]
env PATH=/bin:/usr/bin:/usr/local/bin
env QUEUE=*
env PIDFILE=/var/www/gitorious/app/tmp/pids/resque-worker1.pid
exec /usr/local/bin/chruby-exec ruby-1.9.3-p448 -- /var/www/gitorious/app/bin/rake resque:work
respawn
现在我已经临时执行了在当前终端内启动的QUEUE=* /var/www/gitorious/app/bin/rake resque:work。
根据./script/diagnose轮询器现在正在运行，我收到通知邮件并且活动流已更新。
最后一步（希望）是修复脚本，将其作为服务运行。我会尽快更新这篇文章。
在ce-installer/upgrade.sh中，第 71 行（函数启动-gitorious）我找到了启动resque-worker的解决方案。只需输入：
start resque-worker
（不是！sudo service resque-worker start）