tensorflow: An error occurred during the fetch of repository 'llvm-project':

Click to expand!

Issue Type

Build/Install

Source

source

Tensorflow Version

2.4.1

Custom Code

Yes

OS Platform and Distribution

Linux Ubuntu 20.04

Mobile device

No response

Python version

3.8

Bazel version

3.1.0

GCC/Compiler version

9.4.0

CUDA/cuDNN version

11.6/8

GPU model and memory

RTX3080

Current Behaviour?

Bazel-build code fails when trying to retrieve llvm source code. 

Download from primary fails with 404: 
https://storage.googleapis.com/mirror.tensorflow.org/github.com/llvm/llvm-project/archive/f402e682d0ef5598eeffc9a21a691b03e602ff58.tar.gz

Download from secondary fails with 504:
https://github.com/llvm/llvm-project/archive/f402e682d0ef5598eeffc9a21a691b03e602ff58.tar.gz

Standalone code to reproduce the issue

https://github.com/tensorflow/tensorflow/archive/v2.4.1.tar.gz

Relevant log output

ERROR: An error occurred during the fetch of repository 'llvm-project':
   java.io.IOException: Error downloading [https://storage.googleapis.com/mirror.tensorflow.org/github.com/llvm/llvm-project/archive/f402e682d0ef5598eeffc9a21a691b03e602ff58.tar.gz, https://github.com/llvm/llvm-project/archive/f402e682d0ef5598eeffc9a21a691b03e602ff58.tar.gz] to /home/dgu/.cache/bazel/_bazel_dgu/bbb96d26b2548d6cbb28158a8986fc7d/external/llvm-project/f402e682d0ef5598eeffc9a21a691b03e602ff58.tar.gz: Read timed out

About this issue

Original URL
State: closed
Created 2 years ago
Comments: 20 (9 by maintainers)

Commits related to this issue

Switch to zip file for LLVM source archive. Works around file availability problem from github. See https://github.com/tensorflow/tensorflow/issues/56540 — committed to trevor-m/tensorflow by nluehr 2 years ago
Switch to zip file for LLVM source archive. Works around file availability problem from github. See https://github.com/tensorflow/tensorflow/issues/56540 — committed to trevor-m/tensorflow by nluehr 2 years ago
Switch to zip file for LLVM source archive. Works around file availability problem from github. See https://github.com/tensorflow/tensorflow/issues/56540 — committed to kaixih/tensorflow by nluehr 2 years ago

Most upvoted comments

We (Open-CE team) are also facing the same problem since yesterday. I can’t build TF 2.8 and TF 2.9.1 due to this. .zip format is being downloaded but not .tar.gz format. The recent commits of llvm-project are being downloaded in .tar.gz format. I’d raised these two issues -

Patching workspacebzl file as mentioned in one of the comments works. This issue is seen even with tensorflow-text and tensorflow-serving builds. So, we would need to patch all of these packages of previous release and re-tag.

npanpaliya on Jun 24, 2022

Clarifying my earlier comment, here is a patch that can be used to work around this issue:

diff --git a/third_party/llvm/workspace.bzl b/third_party/llvm/workspace.bzl
index 82d0f71ccbd..6a5cd8e61b5 100644
--- a/third_party/llvm/workspace.bzl
+++ b/third_party/llvm/workspace.bzl
@@ -5,15 +5,14 @@ load("//third_party:repo.bzl", "tf_http_archive")
 def repo(name):
     """Imports LLVM."""
     LLVM_COMMIT = "1cb299165c859533e22f2ed05eb2abd5071544df"
-    LLVM_SHA256 = "5a19ab6de4b0089fff456c0bc406b37ba5f95c76026e6bec294ec28dc28e4cb9"
+    LLVM_SHA256 = "10937113be5c7c551cffd7accf426e35b9071cac3251cb2b711076bdda0e98ea"
 
     tf_http_archive(
         name = name,
         sha256 = LLVM_SHA256,
         strip_prefix = "llvm-project-{commit}".format(commit = LLVM_COMMIT),
         urls = [
-            "https://storage.googleapis.com/mirror.tensorflow.org/github.com/llvm/llvm-project/archive/{commit}.tar.gz".format(commit = LLVM_COMMIT),
-            "https://github.com/llvm/llvm-project/archive/{commit}.tar.gz".format(commit = LLVM_COMMIT),
+            "https://github.com/llvm/llvm-project/archive/{commit}.zip".format(commit = LLVM_COMMIT),
         ],
         build_file = "//third_party/llvm:llvm.BUILD",
         patch_file = ["//third_party/llvm:macos_build_fix.patch"],

benbarsdell on Jun 23, 2022

Hi @danielgusland, Clear the cache by calling bazel clean –expunge Command prompt must be Run as Admin for the bazel build of TensorFlow to be successful. Thank you!

gadagashwini on Jun 23, 2022