Connecting to www.dropbox.com (www.dropbox.com)|128.121.243.106|:443... failed: Connection timed out

[Xuhd08]

when i build docker use Dcokerfile, the CK error unable to resolve

command:
RUN . /opt/vitis_ai/conda/etc/profile.d/conda.sh && conda activate vitis-ai-tensorflow && cd /opt/tvm-vai// && python -m ck pull repo:ck-env && python -m ck install package:imagenet-2012-val-min

failed log:

Installing to /opt/tvm-vai/CK-TOOLS/dataset-imagenet-ilsvrc2012-val-min
72.81
72.81
72.81 Checking whether '/opt/tvm-vai/CK-TOOLS/dataset-imagenet-ilsvrc2012-val-min/ILSVRC2012_img_val.tar' already exists ...
72.81
72.81 Downloading ILSVRC'12 validation dataset from 'https://www.dropbox.com/s/57s11df6pts3z69/ILSVRC2012_img_val_500.tar' ...
72.81 --2024-12-01 18:06:29-- https://www.dropbox.com/s/57s11df6pts3z69/ILSVRC2012_img_val_500.tar
72.81 Resolving www.dropbox.com (www.dropbox.com)... 128.121.243.106, 2a03:2880:f10c:83:face:b00c:0:25de
72.82 Connecting to www.dropbox.com (www.dropbox.com)|128.121.243.106|:443... failed: Connection timed out.
202.5 Connecting to www.dropbox.com (www.dropbox.com)|2a03:2880:f10c:83:face:b00c:0:25de|:443... failed: Network is unreachable.
202.5 Error: Downloading ILSVRC'12 validation set from 'https://www.dropbox.com/s/57s11df6pts3z69/ILSVRC2012_img_val_500.tar' failed!
202.6
202.6 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
202.6 CK detected a PROBLEM in the third-party CK package:
202.6
202.6 CK package: imagenet-2012-val-min
202.6 CK repo: ck-env
202.6 CK repo URL: https://github.com/ctuning/ck-env
202.6 CK package URL: https://github.com/ctuning/ck-env/tree/master/package/imagenet-2012-val-min
202.6 Issues URL: https://github.com/ctuning/ck-env/issues
202.6
202.6 CK stable package URL: https://cKnowledge.io/c/package/imagenet-2012-val-min
202.6
202.6 Please, check that there is no discussion about this issue at https://cKnowledge.io/c/package/imagenet-2012-val-min
202.6
202.6 Please, submit the original comand line and the full log to:
202.6 * https://github.com/mlcommons/ck/issues
202.6 * https://github.com/mlcommons/ck-mlops/issues
202.6 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
202.6
202.6 CK error: [package] package installation failed!

Dockerfile.vitis_ai:100

99 | ENV HOME $TVM_VAI_HOME
100 | >>> RUN . $VAI_ROOT/conda/etc/profile.d/conda.sh &&
101 | >>> conda activate vitis-ai-tensorflow &&
102 | >>> cd $TVM_VAI_HOME/ &&
103 | >>> python -m ck pull repo:ck-env &&
104 | >>> python -m ck install package:imagenet-2012-val-min --path=/opt/tvm-vai/CK-TOOLS
105 |

ERROR: failed to solve: process "/bin/bash -c . $VAI_ROOT/conda/etc/profile.d/conda.sh && conda activate vitis-ai-tensorflow && cd $TVM_VAI_HOME/ && python -m ck pull repo:ck-env && python -m ck install package:imagenet-2012-val-min --path=/opt/tvm-vai/CK-TOOLS" did not complete successfully: exit code: 1
ERROR: docker build failed.

but the www.dropbox.com can be accessed by Chrome. and in terminal use wget command can download the ILSVRC2012_img_val_500.tar. how to solve this problem? anyone can help?

[gfursin]

Thank you very much for reporting. We need to move all such packages to a more reliable repository at MLCommons. Let me sync with MLCommons admins about that!
CC @nathanw-mlc

[gfursin]

By the way, I just realized that you are using a previous version of my automation framework (CK). We can still update it if there are users but I suggest you to try the next generation of CK that I have developed a few years ago (Collective Mind aka CM) and continue improving at MLCommons: https://github.com/mlcommons/ck/tree/master/cm & https://github.com/mlcommons/ck/releases . We have fixes some of those URLs in the new framework but I need to double check.

[nathanw-mlc]

Hey folks, MLCommons has never officially used Dropbox to distribute those files, so I can't help there. I believe the latest version of CM/CK fetches the files from our official Cloudflare R2 buckets.

[gfursin]

Hey Nathan. Yes, that's correct. But I am curious if it makes sense to move such files from DropBox to MLCommons Cloudlare R2 buckets for the deprecated framework (if someone still uses it) or it's better/simpler to migrate to newer automation recipes from CM? I am neutral here.

[nathanw-mlc]

Ah, I see. I'd rather push people onto the newer versions than continue to support deprecated versions. We're running a pretty lean team here, so I don't think we have the bandwidth for that kind of support.

[gfursin]

Sure. No worries - it's understandable!

[Xuhd08]

OK, I try to update it to CM. Thanks!