sunjam
2020 年3 月 31 日 18:51
1
我在想是否可以通过刷新 Redis 或更新它来解决这个问题；它最近 8 个多月几乎没动过。我本人之前没有处理过 Redis，但我们的 Tests-Pass Discourse 实例是使用 local_discourse 搭建的，而不是本论坛现在推荐的 Discourse Docker 。我遇到的问题与无法向用户发送邮件有关。欢迎提供任何帮助和建议！
Sidekiq 未运行。许多任务（如发送邮件）由 Sidekiq 异步执行。请确保至少有一个 Sidekiq 进程正在运行。
Sidekiq 心跳测试失败，正在重启
config/unicorn.conf.rb:147:in `check_sidekiq_heartbeat
config/unicorn.conf.rb:164:in `master_sleep'
unicorn-5.5.4/lib/unicorn/http_server.rb:296:in `join'
unicorn-5.5.4/bin/unicorn:128:in `<top (required)>'
/var/www/discourse/vendor/bundle/ruby/2.6.0/bin/unicorn:23:in `load'
/var/www/discourse/vendor/bundle/ruby/2.6.0/bin/unicorn:23:in `<main>'
Falco
2020 年3 月 31 日 19:42
3
建议您备份当前安装，并按照 Discourse 官方标准安装 创建一个全新的安装。
sunjam
2020 年4 月 2 日 15:22
5
我使用 launcher 重新构建或重启 Redis 容器会有帮助吗？
用法：launcher 命令 配置 [–skip-prereqs] [–docker-args 字符串]
选项：
pfaffman
2020 年4 月 2 日 15:37
6
您是按照 Discourse 官方标准安装 指南安装的，还是通过其他方式安装的？如果是标准安装，那么运行 ./launcher rebuild app 很可能就能解决您的问题。
sunjam
2020 年4 月 2 日 15:49
7
还没有，但我们正在安排迁移时间。我们的 Discourse 已经运行了将近三年，我只是不想增加任何不必要的停机时间。
pfaffman
2020 年4 月 2 日 16:10
8
如果您最初没有进行标准安装，那么谁能知道什么能解决您当前的问题呢。
我建议先设置一个新的虚拟机，并在其中进行测试安装以确保其正常运行。在进行迁移时，您可以将旧论坛设置为只读模式，进行备份，然后将其恢复到新虚拟机上，并更改 DNS，从而实现几乎零停机时间。（实际上会有一点停机时间，因为更改 DNS 后，您需要重新构建以安装 Let’s Encrypt 证书）
sunjam
2020 年4 月 2 日 17:05
9
谢谢，Let’s Encrypt 实际上应该很容易，因为它已被配置为只需编辑简单的 Nginx 反向代理。听到这个消息很好，但请预期会有一些停机时间。只需将其安排在非高峰时段即可。
sunjam
2020 年4 月 26 日 19:14
10
我即将进行迁移，并将备份恢复到官方镜像中。关于清理已备份的邮件等内容，有什么需要特别注意的吗？以免在 Sidekiq 等组件正常运行后给用户造成垃圾信息骚扰。谢谢！
riking
2020 年4 月 30 日 22:10
11
Sidekiq 作业队列无法迁移，因此大量邮件将被丢弃，而摘要邮件将恢复正常发送。
sunjam
2020 年5 月 11 日 00:31
12
因此，在尝试按照官方 Discourse Docker 镜像的文档进行安装时遇到了问题。我们得出结论，我们缺少之前在 Docker 中拥有的 Redis 容器、邮件接收容器和数据容器。
看来我们之前的安装包含：app.ymldata.ymlmail-receiver.ymlredis.yml
这种多容器设置与基本安装说明不同。我确实有旧版 /var/discourse 的备份可供参考。
FAILED │················································································································
-------------------- │················································································································
Pups::ExecError: cd /var/www/discourse && su discourse -c 'bundle exec rake db:migrate' failed with return #<Process::Status: pid 645 exit 1> │················································································································
Location of failure: /pups/lib/pups/exec_command.rb:112:in `spawn' │················································································································
exec failed with the params {"cd"=>"$home", "hook"=>"db_migrate", "cmd"=>["su discourse -c 'bundle exec rake db:migrate'"]} │················································································································
bbf0e57ac69f1febe2a5f149aa7e6e12541c3c23aaf199188fdf19d507254b58 │················································································································
** FAILED TO BOOTSTRAP ** please scroll up and look for earlier error messages, there may be more than one. │················································································································
./discourse-doctor may help diagnose the problem.
sunjam
2020 年5 月 11 日 00:42
13
看起来我们漏掉了步骤 ./launcher bootstrap data./launcher start redis
[Mon 11 May 2020 12:53:20 AM UTC] 运行重载命令：sv reload nginx │················································································································
fail: nginx: runsv 未运行 │················································································································
[Mon 11 May 2020 12:53:20 AM UTC] 重载错误： │················································································································
[Mon 11 May 2020 12:53:21 AM UTC] 域名未更改。 │················································································································
[Mon 11 May 2020 12:53:21 AM UTC] 跳过，下次续期时间为：Thu 09 Jul 2020 11:33:04 PM UTC │················································································································
[Mon 11 May 2020 12:53:21 AM UTC] 添加 '--force' 以强制续期。 │················································································································
[Mon 11 May 2020 12:53:21 AM UTC] 正在安装密钥至：/shared/ssl/discuss.noisebridge.info_ecc.key │················································································································
[Mon 11 May 2020 12:53:21 AM UTC] 正在安装完整证书链至：/shared/ssl/discuss.noisebridge.info_ecc.cer │················································································································
[Mon 11 May 2020 12:53:21 AM UTC] 运行重载命令：sv reload nginx │················································································································
fail: nginx: runsv 未运行 │················································································································
[Mon 11 May 2020 12:53:21 AM UTC] 重载错误： │················································································································
run-parts: 正在执行 /etc/runit/1.d/remove-old-socket │················································································································
runsvdir 已启动，PID 为 626 │················································································································
ok: run: redis: (pid 636) 0s │················································································································
chgrp: 无效的用户组：'syslog' │················································································································
ok: run: postgres: (pid 639) 0s │················································································································
rsyslogd: imklog: 无法打开内核日志 (/proc/kmsg)：操作不允许。 │················································································································
rsyslogd: imklog 模块激活失败 [v8.1901.0 请查看 https://www.rsyslog.com/e/2145 ] │················································································································
supervisor pid: 640 unicorn pid: 667
sunjam
2020 年5 月 11 日 01:26
14
好的，我们已成功恢复之前的 Discourse 实例！现在显示：
所有出站邮件已被管理员全局禁用。将不会发送任何类型的电子邮件通知。
没错，导入器已进行此设置，以防止您在迁移后一次性发送 50,000 封邮件。请在您的站点设置中谨慎开启电子邮件功能。
我们的实例完全由志愿者维护，因此过去一个月我一直在自学如何维护它。如果快速修复 Redis 会有帮助（该容器已运行一年未动），我很乐意应用它。