/glacier-targz-uploader

command line tools for Amazon Glacier. Can archive a whole directory without creating an intermediate tar.gz file.

Primary LanguageJava

glacier-targz-uploader

(日本語の解説はこのブログ記事をどうぞ: http://blog.livedoor.jp/nabokov7/archives/1517648.html )

What is this ?

Set of command line tools to archive/retrieve your data to/from Amazon Glacier vault.

  • Can upload a large directory as tar.gz without creating an intermediate file.
  • Can resume upload after interrupt / transient failure.

Currently, following 5 tools are available:

  • TarGzUploader: Can compress and upload all contents of the specified directory without creating an intermediate tar.gz file.
  • StreamUploader: Can send whatever comes into stdin to the vault.
  • Downloader: Downloads the specified archive.
  • ListInventory: Outputs a list of inventories in the vault. (This is mostly a copy-paste of the code in the official manual. It just outputs the response JSON as is.)
  • EmptyInventory: As you cannot delete a vault unless it is empty, this command helps you by deleting everything in the specified vault.

Instructions

  • Create your glacier vault in aws management console, set your credentials in AWSCredentials.properties, and then:

Either

  • Run the executable jar files in builds/ directory.

    e.g. java -Xmx1G -Dfile.encoding=UTF-8 -jar listVaultInventory.jar -vault [vault_name]

    (-Dfile.encofing=UTF-8 is sometimes necessary to handle non-ascii file names correctly)

Or...

  • Compile and create executable jars for each of the classes under src/cmdline.

For usage, run the executable jar without args. Usages in more detail are embedded as comments. Remember to set your own AWS credentials in AWSCredentials.properties file.

Prerequisites (to compile the code)