如何调试Python Azure函数错误：[ERRNO 28]设备上没有剩余的空间？
我具有通常执行ETL过程的Python Azure函数。我的功能还包括将文件从API下载到临时目录，并将文件上传到容器中。我收到以下错误[ERRNO 28]设备上没有空间，我试图检查每个可能的位置，因为这是一个空间问题，我认为我在存储帐户中有足够的空间我重新启动了我的功能应用以清除我的临时目录。
I have a Python Azure function which normally do the etl process. My function also includes the downloading of files from API to Temp directory and Uploading the files to container. I am getting the following error [Errno 28] No space left on device, I had tried to check every possible place since it is a space issue, I think i have enough space in my storage account and also i had restarted my func-app to clear out my temp directory.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
Microsoft-Hosted代理为您的源提供10 GB的存储空间，并构建输出 - 在这里。
如果您的源代码和构建输出大于10GB，我建议使用自托管代理来完成您的构建。您可以尝试使用以下命令来清空缓存，以查看是否有助于python存储问题。 - ＆gt; PIP安装-NO-CACHE-DIR TENSORFLOW-GPU。
Azure功能的每个会话的最大数据为5GB。即使您的存储帐户允许无限数据，Azure功能也将无法管理大量数据。结果，此错误很可能是由功能引起的。
作为一个临时修复，您可以扩大应用程序服务计划的规模，然后再次扩展到原始内容，并且必须清洁空间并开始工作。
进一步参考 Ref1 ， 有关更多信息。
Microsoft-hosted agents provide 10 GB of storage for your source and build outputs- here.
If your source code and build outputs are larger than 10GB, I recommend using the Self Hosted Agent to complete your build. You can try emptying the cache with the below command once to see if that helps with the python storage issue. --> pip install --no-cache-dir tensorflow-gpu.
The maximum data per session for Azure Functions is 5GB. Even if your storage account allows for unlimited data, the Azure function will be unable to manage large amounts of data. As a result, this error is most likely caused by a function.
As a temporary fix, you can scale up your app service plan and the again scale down to original and it has to clean space and start working.
Further refer ref1,ref2 for more information.
Azure函数的限制为每个会话最大值5GB数据。因此，即使您的存储帐户可以使用无限的数据，Azure功能也不会一次处理如此庞大的数据。因此，很可能此错误来自函数本身。
Azure Functions have a limit of 5GB data per session max. So even if your storage account can take unlimited data, azure function will not be to handle such huge data at a time. So most probably this error comes from function itself.