Slingshot Recovery

The Slingshot Recovery effort includes the Restore and Repair programs. You can learn more about each of these here:

• Slingshot Restore
• Slingshot Repair

A dashboard displaying progress for both Slingshot Restore and Slingshot Repair is available at slingshot.filecoin.io/recovery.

Restore

If you're a client participating in the Slingshot Restore Program, this directory has the metadata used to track your deals on chain. Please feel free to create pull requests for the following scenarios.

Tracking CIDs being restored

If you need to update the list of payload CIDs being tracked for your dataset onboarding, PR the list of (each of) your dataset's CIDs in this repository, at the path recovery/cids/restore/.

• CID lists must be .csv
• The file name should match your assigned dataset name, in kebab-case (also referred to as the dataset-slug)
  • check your dataset-slug in the Client Info spreadsheet or the recovery dataset list
• Your file's first row should say payload_cid,data_size,filenames where payload_cid is the CID as defined here, data_size is the corresponding size in bytes of the data stored in that CID, and filenames is an array of names of files in that CID (to be used for indexing and retrieval in the future)
• All CIDs must be base32 CID v1's only (i.e., ba...)
• Submit payload CIDs only (these will most likely begin with bafyk... or bafyb...)
  • Not piece CIDs
  • Not deal proposal CIDs (CIDs beginning with baga... or bafyr... are incorrect)
  • If your payload CIDs are in Qm... format, they must be converted to base 32 CIDv1's; if you've installed cid-tool, you can run cid base32 Qm... to convert
• If you need to add more CIDs, simply create another PR and append to your existing file
• data_size should always be in bytes
• The filenames array for each CID must follow this format: "['file_1.ext','file_2.ext']". Even with just a single file in a CID, this needs to be listed as "['file.ext']".

Example

File name: chosen-dataset.csv

File path: recovery/cids/restore/chosen-dataset.csv

File content:

payload_cid,data_size,filenames
bafybeicelmss5jrbnze37rhyiuhat,17180000000,"['file_1.txt']"
bafybeicelmss7s2zmqekui7ysukid,17180000000,"['file_2.txt']"
bafybeicelmssa6xkdvebnzu2xmaol,17180000000,"['file_3.txt']"
bafybeicelmssdtuhui4ea32ohda6z,17180000000,"['file_4.txt','file_5.txt']"
bafybeicelmssfokmkaugllwx255ss,34359738368,"['file_6.txt','file_7.txt','file_8.txt']"
bafybeicelmsskocznkpqc4mj23got,34359738368,"['file_9.txt']"

Updating address for client deal making

If you need to update the client address being used to track your deals on chain, update the Client Info spreadsheet and PR recovery/files/client-list.json. The Slingshot admin team will periodically be updating this file but if you'd like to see updates on the dashboard sooner, feel free to submit a PR any time.

• each JSON object in client-list.json is unique
• client addresses are tracked per dataset, so if you use the same address for different datasets (note: not recommended, but possible) you must add a new object for each dataset with the same address

Example

Adding a new address f1th1s1smyn3waddr3ssfr0mwh1chd3alsaremad3 for client "Restorer" who is assigned to Landsat 7 and 8, and who's Slack handle is "OnboardingLegend".

PR client-list.json and add the following to the list:

  {
    "client_address": "f1th1s1smyn3waddr3ssfr0mwh1chd3alsaremad3",
    "dataset_slug": "landsat-7",
    "client_name": "Restorer",
    "client_slack_handle": "OnboardingLegend"
  },
  {
    "client_address": "f1th1s1smyn3waddr3ssfr0mwh1chd3alsaremad3",
    "dataset_slug": "landsat-9",
    "client_name": "Restorer",
    "client_slack_handle": "OnboardingLegend"
  },

Repair

If you are interested in participating in the Repair effort, sign up to be a Storage Provider on Estuary and participate in the conversation in #slingshot + #slingshot-announcements in Filecoin Slack.