Skip to content
This repository has been archived by the owner on Mar 25, 2022. It is now read-only.

Latest commit

 

History

History
100 lines (63 loc) · 5.1 KB

README.md

File metadata and controls

100 lines (63 loc) · 5.1 KB
Raw

ibm-cos-sanity-check

A tool for calculating SHA3-256 of each file in a given bucket.

Still Working In Progress, but should be functional.

Usage

Deploy the docker image on IBM Cloud Code Engine as Job, set some environment variables, then let the job run. You can set up a IBM Log Analysis to see if anything goes wrong. The program will list all sub directories and calculate SHA3-256 of each file listed. The result will be stored in file per folder, contains all file's checksum in that folder.

For example, I have a folder:

  • some_prefix/
    • subfolder1/
      • files_in_subfolder_1...
    • subfolder2/
      • files_in_subfolder_2
    • files_in_folder...

Each file in current folder (files_in_folder) will be calculate, and the results will be stored at some_prefix/SANITY_CHECK.txt, and files_in_subfolder1's checksum will be stored at some_prefix/subfolder1/SANITY_CHECK.txt, files_in_subfolder2's checksum will be stored at some_prefix/subfolder2/SANITY_CHECK.txt.

The format of SANITY_CHECK.txt is:

# Prefix: /<YOUR_PREFIX>
SHA3-256:<SHA_RESULT>:<TIMESTAMP>:<OBJECT_KEY>
ERROR:<ERROR_MESSAGE>:<TIMESTAMP>:<OBJECT_KEY>
...

The first line start with # and it gives the prefix of current folder.

Start from the second line, each line gives <FORMAT>:<PAYLOAD>:<TIMESTAMP>:<OBJECT_KEY>, <FORMAT> has three options: SHA3-256, ERROR and DEBUG. For SHA3-256, <PAYLOAD> gives the SHA3-256 result of object. ERROR means something goes wrong when calculating the hash, and the <PAYLOAD> gives the error message. For DEBUG, the payload is always SKIP. The <TIMESTAMP> counts in milliseconds, it is the last modified time of the object.

If there are something wrong during the listing file or initialization stage, please check the log, there should be a ERROR message.

Environment variables

IAM_ENDPOINT

Set the IBM IAM endpoint URL, this is not required, the default value is https://iam.cloud.ibm.com/identity/token.

COS_ENDPOINT

Required. This is the S3 endpoint of your IBM COS. For more details, please refer to this document.

Please do notice, when running this docker in Code Engine, please use Direct endpoint for free bandwidth, use public endpoint will cause public endpoint fee.

COS_CREDENTIAL

Required. This is the JSON credential, just copy it from your IBM COS console, it should looks like this:

{
  "apikey": "<SOMETHING>",
  "cos_hmac_keys": {
    "access_key_id": "<SOMETHING>",
    "secret_access_key": "<SOMETHING>"
  },
  "endpoints": "<SOMETHING>",
  "iam_apikey_description": "<SOMETHING>",
  "iam_apikey_name": "<SOMETHING>",
  "iam_role_crn": "<SOMETHING>",
  "iam_serviceid_crn": "<SOMETHING>",
  "resource_instance_id": "<SOMETHING>"
}

COS_BUCKET_LOCATION

Required. If your credential use HMAC, then make sure this location is correct. Otherwise the library won't be able to talk with IBM COS service.

COS_BUCKET_NAME

Required. This is your COS bucket name.

APP_PREFIX

Not Required, but good to have one. This is the prefix. You cannot set it to empty string in IBM Cloud Code Engine console, so when you not set that, default value is empty, aka the root of your bucket. However I don't recommend running this docker with root folder, since I cannot make sure my code will function correctly. So try to limit the scope by using the prefix.

APP_SANITY_NAME

This is the result file name. This is not required, the default value is SANITY_CHECK.txt.

APP_DEBUG

This parameter will put program in debug mode. In debug mode, the calculation will not performed, thus it always gives SKIP as result. Also, in debug mode, the result will not be write into the bucket.

APP_DISABLE_REUSE

Existence of this parameter will disable the reuse of the result. By default the program will try to read the old sanity file and reuse the result to save some vCPU time (if and only if the timestamp in result matched the last modified time of the object). But if you want to override this, just set this env to any value.

TODOs

Currently the program cannot distinguish different storage class. From the documentation, standard class will return null when trying to get the storage class. So only object which storageClass == null will be calculated, rest of them will gives error.

Also, I want to implement some checking when some files' checksum is represented in the sanity check file, but considering some files might in archived mode, I have no idea of how to do this.

Last but not least, this is only my side project, I moved my recently (in past 3 years) unused freezing cold personal data from Google drive to IBM Cloud COS, by using the archived storage class, I can get the price as low as 1.02USD per TB per month, and also free myself from maintaining a bunch of offline hard drives or online NAS in my house. The tape drive is too expensive for me, so I guess this is the best plan.

Feel free to start a discussion, submit an issue, or create a pull request.