谷歌云计算引擎冻结

Question

今天我发现在Google云中我的vm实例出了问题。我无法通过gcloud compute ssh连接它。唯一的解决办法是关闭虚拟机并重新启动。我猜到了错误发生的正确日志文件位置。/var/log/daemon.log

Nov 23 17:39:01 globo-de systemd[1]: Started Clean php session files.
Nov 23 18:09:01 globo-de systemd[1]: Starting Clean php session files...
Nov 23 18:09:03 globo-de systemd[1]: Started Clean php session files.
Nov 23 18:39:01 globo-de systemd[1]: Starting Clean php session files...
Nov 23 18:39:04 globo-de systemd[1]: Started Clean php session files.
Nov 23 18:56:49 globo-de google-ip-forwarding: WARNING Exception running ['ip', 'route', 'ls', 'table', 'local', 'type', 'local', 'scope', 'host', 'dev', 'eth0', 'proto', '66']. [Errno 12] Cannot allocate memory.
Nov 23 18:57:43 globo-de google-accounts: ERROR Exception calling the response handler. [Errno 12] Cannot allocate memory.#012Traceback (most recent call last):#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/metadata_watcher.py", line 196, in WatchMetadata#012    handler(response)#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/accounts_daemon.py", line 259, in HandleAccounts#012    self.oslogin.UpdateOsLogin(enable=False)#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 79, in UpdateOsLogin#012    status = self._GetStatus()#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 60, in _GetStatus#012    retcode = self._RunOsLoginControl('status')#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 47, in _RunOsLoginControl#012    return subprocess.call([constants.OSLOGIN_CONTROL_SCRIPT, action])#012  File "/usr/lib/python2.7/subprocess.py", line 168, in call#012    return Popen(*popenargs, **kwargs).wait()#012  File "/usr/lib/python2.7/subprocess.py", line 390, in __init__#012    errread, errwrite)#012  File "/usr/lib/python2.7/subprocess.py", line 916, in _execute_child#012    self.pid = os.fork()#012OSError: [Errno 12] Cannot allocate memory
Nov 23 18:58:02 globo-de google-ip-forwarding: WARNING Exception running ['ip', 'route', 'ls', 'table', 'local', 'type', 'local', 'scope', 'host', 'dev', 'eth0', 'proto', '66']. [Errno 12] Cannot allocate memory.
Nov 23 18:59:35 globo-de google-ip-forwarding: WARNING Exception running ['ip', 'route', 'ls', 'table', 'local', 'type', 'local', 'scope', 'host', 'dev', 'eth0', 'proto', '66']. [Errno 12] Cannot allocate memory.
Nov 23 18:59:54 globo-de google-accounts: ERROR Exception calling the response handler. [Errno 12] Cannot allocate memory.#012Traceback (most recent call last):#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/metadata_watcher.py", line 196, in WatchMetadata#012    handler(response)#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/accounts_daemon.py", line 259, in HandleAccounts#012    self.oslogin.UpdateOsLogin(enable=False)#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 79, in UpdateOsLogin#012    status = self._GetStatus()#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 60, in _GetStatus#012    retcode = self._RunOsLoginControl('status')#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 47, in _RunOsLoginControl#012    return subprocess.call([constants.OSLOGIN_CONTROL_SCRIPT, action])#012  File "/usr/lib/python2.7/subprocess.py", line 168, in call#012    return Popen(*popenargs, **kwargs).wait()#012  File "/usr/lib/python2.7/subprocess.py", line 390, in __init__#012    errread, errwrite)#012  File "/usr/lib/python2.7/subprocess.py", line 916, in _execute_child#012    self.pid = os.fork()#012OSError: [Errno 12] Cannot allocate memory
Nov 23 19:03:53 globo-de google-ip-forwarding: WARNING Exception running ['ip', 'route', 'ls', 'table', 'local', 'type', 'local', 'scope', 'host', 'dev', 'eth0', 'proto', '66']. [Errno 12] Cannot allocate memory.
Nov 23 19:07:52 globo-de google-accounts: ERROR Exception calling the response handler. [Errno 12] Cannot allocate memory.#012Traceback (most recent call last):#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/metadata_watcher.py", line 196, in WatchMetadata#012    handler(response)#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/accounts_daemon.py", line 259, in HandleAccounts#012    self.oslogin.UpdateOsLogin(enable=False)#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 79, in UpdateOsLogin#012    status = self._GetStatus()#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 60, in _GetStatus#012    retcode = self._RunOsLoginControl('status')#012  File "/usr/lib/python2.7/dist-packages/google_compute_engine/accounts/oslogin_utils.py", line 47, in _RunOsLoginControl#012    return subprocess.call([constants.OSLOGIN_CONTROL_SCRIPT, action])#012  File "/usr/lib/python2.7/subprocess.py", line 168, in call#012    return Popen(*popenargs, **kwargs).wait()#012  File "/usr/lib/python2.7/subprocess.py", line 390, in __init__#012    errread, errwrite)#012  File "/usr/lib/python2.7/subprocess.py", line 916, in _execute_child#012    self.pid = os.fork()#012OSError: [Errno 12] Cannot allocate memory
Nov 23 19:09:37 globo-de google-ip-forwarding: WARNING Exception running ['ip', 'route', 'ls', 'table', 'local', 'type', 'local', 'scope', 'host', 'dev', 'eth0', 'proto', '66']. [Errno 12] Cannot allocate memory.
Nov 23 19:11:32 globo-de systemd[1]: phpsessionclean.service: Failed to fork: Cannot allocate memory
Nov 23 19:11:56 globo-de systemd[1]: phpsessionclean.service: Failed to run 'start' task: Cannot allocate memory
Nov 23 19:12:24 globo-de systemd[1]: Failed to start Clean php session files.
Nov 23 19:12:50 globo-de systemd[1]: phpsessionclean.service: Unit entered failed state.
Nov 23 19:13:37 globo-de systemd[1]: phpsessionclean.service: Failed with result 'resources'.
Nov 23 19:14:58 globo-de systemd[1]: apt-daily.service: Failed to fork: Cannot allocate memory
Nov 23 19:16:13 globo-de systemd[1]: apt-daily.service: Failed to run 'start' task: Cannot allocate memory

我该怎么做才能避免再次发生这种情况？

- treffner

Answer 1

我猜问题是由于google-cloud-compute引擎虚拟机没有交换文件。因此，我使用在Google上找到的以下命令创建了一个交换文件。

sudo dd if=/dev/zero of=/var/swap bs=2048 count=524288
sudo chmod 600 /var/swap
sudo mkswap /var/swap
sudo swapon /var/swap

然后我将以下一行添加到/etc/fstab中，以使其永久化。

/var/swap none swap sw 0 0

- treffner

---

现在工作正常吗？我也遇到了同样的问题。 - Muthu

是的，我的冻结问题已经解决了。 - treffner

Answer 2

另一种解决方案是避免交换，将vm.swappiness设置为0。 < p > < code >＃sysctl -w vm.swappiness = 0 然而，这更像是操作系统镜像问题，而不是Google云的问题。

- Fabricio Figueiredo