---
title: 手动下载TF Serving编译依赖包
description: "TF Serving编译过程中，需要下载依赖包，可能会因为网络问题下载失败，报错如图1所示："
url: https://www.hiascend.com/document/detail/zh/TensorFlowCommercial/latest/migration/tfmigr1/atlastfserv_26_0011.html
sourcePath: /source/zh/TensorFlowCommercial/900/migration/tfmigr1/atlastfserv_26_0011.html
indexId: 7068c4270d0283edd8031e432c8a5d34d3d24c984e69a1de3ed521f2b975711d78
---
# 手动下载TF Serving编译依赖包

TF Serving编译过程中，需要下载依赖包，可能会因为网络问题下载失败，报错如图1所示：

图1 TF Serving编译报错

解决方法如下所示：

1. 参考如下链接下载所需依赖包，将依赖包上传至服务器任意路径下（例如：“${HOME}”）。
  tensorflow(https://github.com/tensorflow/tensorflow/archive/590d6eef7e91a6a7392c8ffffb7b58f2e0c8bc6b.tar.gz)、rules_closure(https://github.com/bazelbuild/rules_closure/archive/316e6133888bfc39fb860a4f1a31cfcbae485aef.tar.gz)、bazel-skylib(https://github.com/bazelbuild/bazel-skylib/archive/0.7.0.tar.gz)、rapidjson(https://github.com/Tencent/rapidjson/archive/v1.1.0.zip)、abseil-cpp(https://github.com/abseil/abseil-cpp/archive/36d37ab992038f52276ca66b9da80c1cf0f57dc2.tar.gz)、libevent(https://github.com/libevent/libevent/archive/release-2.1.8-stable.zip)和llvm(https://github.com/llvm-mirror/llvm/archive/7a7e03f906aada0cf4b749b51213fe5784eeff84.tar.gz)。

  下载的依赖包需要重命名才能使用，如下表所示：

**表1 依赖包名称修改**

| 依赖包 | 修改前 | 修改后 |
| --- | --- | --- |
| tensorflow | tensorflow\-590d6eef7e91a6a7392c8ffffb7b58f2e0c8bc6b.tar.gz | 590d6eef7e91a6a7392c8ffffb7b58f2e0c8bc6b.tar.gz |
| rules\_closure | rules\_closure\-316e6133888bfc39fb860a4f1a31cfcbae485aef.tar.gz | 316e6133888bfc39fb860a4f1a31cfcbae485aef.tar.gz |
| bazel\-skylib | bazel\-skylib\-0.7.0.tar.gz | 0.7.0.tar.gz |
| rapidjson | rapidjson\-1.1.0.zip | v1.1.0.zip |
| abseil\-cpp | abseil\-cpp\-36d37ab992038f52276ca66b9da80c1cf0f57dc2.tar.gz | 36d37ab992038f52276ca66b9da80c1cf0f57dc2.tar.gz |
| libevent | libevent\-release\-2.1.8\-stable.zip | release\-2.1.8\-stable.zip |
| llvm | llvm\-7a7e03f906aada0cf4b749b51213fe5784eeff84.tar.gz | 7a7e03f906aada0cf4b749b51213fe5784eeff84.tar.gz |


2. 在编译TF Serving时添加“--distdir”参数，如下所示：

```
bazel --output_user_root=/opt/tf_serving build -c opt --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --distdir=${HOME}/tensorflow_serving/model_servers:tensorflow_model_server
```