在使用TensorFlow Model Server时，我会遇到此错误：版本`glibc_2.28＆＃x27;未找到
在运行教程，我在下方运行代码
%% bash -bg nohup tensorflow_model_server
-rest_api_port = 8501
-model_name = tfrbert
-model_base_path =“/content/drive/mydrive/app/model14/export/export/best_model_by_loss/”＆gt; server.log 2＆gt; 1＆gt;＆amp; 1
对我有用。但是现在我遇到了这个错误：
tensorflow_model_server：/lib/x86_64-linux-gnu/libc.so.6：版本glibc_2.28'找不到（tensorflow_model_server_model_server） -gnu/libstdc ++。so.6：版本 glibcxx_3.4.26找不到（tensorflow_model_server要求） tensorflow_model_server：/lib/x86_64-linux-gnu/libm.so.6：找不到版本`glibc_2.29'（由tensorflow_model_server要求）
我搜索了很多，但没有帮助。有人知道如何解决问题吗？ 多谢。
while running the tutorial, I run the code below
%%bash --bg
nohup tensorflow_model_server
--rest_api_port=8501
--model_name=tfrbert
--model_base_path="/content/drive/MyDrive/app/model14/export/best_model_by_loss/" >server.log 2>&1
it worked for me befor. but now I encounter with this error:
tensorflow_model_server: /lib/x86_64-linux-gnu/libc.so.6: version GLIBC_2.28' not found (required by tensorflow_model_server) tensorflow_model_server: /usr/lib/x86_64-linux-gnu/libstdc++.so.6: versionGLIBCXX_3.4.26' not found (required by tensorflow_model_server)
tensorflow_model_server: /lib/x86_64-linux-gnu/libm.so.6: version `GLIBC_2.29' not found (required by tensorflow_model_server)
I searched a lot, but nothing helped. does anybody know how to solve the problem?
thanks a lot.
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（3）
错误是指：
tensorflow_model_server二进制已编译并链接在使用Glibc-2.28或更新的系统上，您正在尝试在具有GLIBC-2.27或以上的系统上运行它。您应该找到
tensorflow_model_server适当的系统的构建。另外，您可以使用带有GLIBC-2.28或更新的Docker容器，并在该容器中运行
Tensorflow_model_server。另一个选项是构建
Tensorflow_model_server您自己在系统上 - 生成的二进制文件将使用它。The error means: the
tensorflow_model_serverbinary you have was compiled and linked on a system with GLIBC-2.28 or newer, and you are trying to run it on a system with GLIBC-2.27 or older.You should find a build of
tensorflow_model_serverappropriate for your system.Alternatively, you could use a docker container with GLIBC-2.28 or newer, and run
tensorflow_model_serverin that container.Another option is to build
tensorflow_model_serveryourself on your system -- the resulting binary will work on it.当我尝试在
colab上使用训练有素的对象检测模型时，我遇到了同样的错误。原因是colab是2.27安装的glibc版本，而最新版本的TensorFlow Serving Serving（2.9.1）需要GlibBC的更新版本恰好是2.28。您可以阅读有关glibc 的更多信息。这是您查看Glibc的版本：
ldd -version在我的情况下，我不是使用最新版本的Tensroflow Serving，而是使用了
1.15版本，它的工作非常起在Google Colab上很好。这就是我安装1.15版本的方式：I got the same error while I was trying to serve my trained Object Detection model on
Colab. The reason being the GLIBC version installed onColabis2.27whereas the latest version of Tensorflow Serving(2.9.1)requires the updated version of GLIBBC which happens to be2.28. You can read more about GLIBC here.This is how you check the version of GLIBC :
ldd --versionIn my case, rather than using the latest version of Tensroflow Serving, I used the
1.15version which worked very well on Google Colab. This is how I installed the1.15version :只需安装此功能，一切都可以正常工作。
Just install this and everything will work ok .