Offline Resources
- After downloading and uploading the offline image file to the corresponding deployment server, it needs to be imported before it can be used.
- For specific import steps, please refer to the examples provided at the bottom of this document.
- Linux amd64
- Linux arm64
| Name | Download Link |
|---|---|
| mingdaoyun-hdp HDP Microservices | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-hdp-linux-amd64-1.2.1.tar.gz |
| mingdaoyun-core Core Microservices | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-core-linux-amd64-7.4.2.tar.gz |
| mingdaoyun-sc Storage components | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-sc-linux-amd64-3.2.0.tar.gz |
| mingdaoyun-flink Data Integration | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-flink-linux-amd64-1.19.720.tar.gz |
| mingdaoyun-file File Storage | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-file-linux-amd64-1.7.0.tar.gz |
| Name | Download Link |
|---|---|
| mingdaoyun-hdp HDP Microservices | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-hdp-linux-arm64-1.2.1.tar.gz |
| mingdaoyun-core Core Microservices | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-core-linux-arm64-7.4.2.tar.gz |
| mingdaoyun-sc Storage components | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-sc-linux-arm64-3.2.0.tar.gz |
| mingdaoyun-flink Data Integration | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-flink-linux-arm64-1.19.720.tar.gz |
| mingdaoyun-file File Storage | https://pdpublic.mingdao.com/private-deployment/offline/mingdaoyun-file-linux-arm64-1.7.0.tar.gz |
-
When performing offline deployment in standalone mode for the first time, the default HDP only needs to download four offline image packages
mingdaoyun-hdp,mingdaoyun-core,mingdaoyun-sc, andmingdaoyun-flink. After uploading to the server, load it through thedocker load -i xxx.tar.gzcommand. -
If you need to download the offline image package of the old version, you only need to change the version number in the download link by default.
-
If the deployment environment limits the upload size of a single file, making it impossible to upload a large offline image package, you can refer to the following method to split the image package and upload it, and restore the original file on the server.
👉 Reference method: Split and restore offline image package
Take the offline image package
mingdaoyun-hdp-linux-amd64-1.2.1.tar.gzas an example:-
Calculate the MD5 value of the image package locally for subsequent verification of file integrity.
md5sum mingdaoyun-hdp-linux-amd64-1.2.1.tar.gz -
Use the
splitcommand in the local environment to split the image package. The following example sets each shard to 800 MB and uses a numeric suffix:split -b 800M -d -a 3 mingdaoyun-hdp-linux-amd64-1.2.1.tar.gz mingdaoyun-hdp-linux-amd64-1.2.1.tar.gz.-b: Specify the size of each fragment. The unit supportsB,K,M, etc.-d: Use numbers as shard suffix.-a: Specify the suffix length, 3 digits in the example.
-
Upload all slices to the same directory on the server.
-
Merge the shards on the server and restore the original image package.
cat mingdaoyun-hdp-linux-amd64-1.2.1.tar.gz.* > mingdaoyun-hdp-linux-amd64-1.2.1.tar.gz -
Calculate the MD5 value again. After confirming that the result is consistent with the result before splitting, you can import the image.
md5sum mingdaoyun-hdp-linux-amd64-1.2.1.tar.gz
-
**After the image package is downloaded and uploaded to the corresponding deployment server, you need to import the image through the following command before it can be used: **
- Docker
- Kubernetes
-
Import the image
docker load -i xxx.tar.gz -
View the image
docker images
-
Unzip the image file
gunzip -d xxx.tar.gz -
Import the image
ctr -n k8s.io image import xxx.tar -
View the image
crictl images