How to Integrate Retool with Backblaze B2

Backblaze B2 application keys are the credential type used for API access. Navigate to the Backblaze web console (secure.backblaze.com) and go to App Keys in the left sidebar. Click Add a New Application Key. Configure the application key settings: - Name: 'retool-dashboard' (or a descriptive name) - Allow access to bucket(s): choose 'All' or select specific buckets if you want to restrict Retool to certain buckets only - Type of access: Read and Write (or Read Only if your dashboard is monitoring-only) - File name prefix: leave blank to allow access to all objects, or set a prefix to restrict to specific paths - Duration: leave blank for a non-expiring key, or set a duration in seconds (recommended for tighter security) Click Create New Key. You will see the applicationKeyId and applicationKey displayed together — this is the ONLY time the applicationKey will be visible. Copy both values to a secure location immediately. Next, identify your S3-compatible endpoint URL. Go to the Backblaze Dashboard → Buckets, click on your bucket name, and look for the Endpoint field. The format is s3.REGION.backblazeb2.com where REGION matches your bucket's region (e.g., us-west-004 gives s3.us-west-004.backblazeb2.com). Note this URL — it forms the base URL for your Retool resource. The applicationKeyId maps to the AWS Access Key ID field, and the applicationKey maps to the AWS Secret Access Key field in Retool's AWS Signature V4 authentication.

Open Retool and navigate to the Resources tab. Click Add Resource and select REST API from the list. Configure the resource: - Name: 'Backblaze B2' - Base URL: https://s3.YOUR_REGION.backblazeb2.com (replace with your actual S3-compatible endpoint URL) Scroll to the Authentication section and select AWS v4 from the dropdown. B2's S3-compatible API uses the same AWS Signature V4 signing process that AWS S3 uses. Fill in the fields: - AWS Region: your B2 bucket region (e.g., us-west-004 — use the region code from your endpoint URL, removing the 's3.' prefix) - AWS Access Key ID: your B2 applicationKeyId (not the application key itself — this is the ID that starts with a numeric prefix) - AWS Secret Access Key: your B2 applicationKey value - AWS Service: s3 Retool will use these credentials to sign requests with AWS Signature V4 — the same algorithm B2 expects for S3-compatible API calls. All requests are signed with the correct Authorization header and x-amz-date header automatically. Add a default header to ensure proper content negotiation: - Header name: Accept - Header value: application/xml B2's S3-compatible API returns XML responses by default (matching S3 behavior). Click Save Changes. Test the connection — a successful test will return a 200 response from B2's S3-compatible endpoint.

Create a query to list all accessible B2 buckets. In your Retool app, go to the Code panel and create a new query. Select the Backblaze B2 resource. Configure: - Method: GET - Path: / (root path — this is the ListBuckets operation in S3-compatible API) - No additional headers or parameters needed B2's S3-compatible API returns an XML response for ListBuckets. Add a JavaScript transformer to parse the bucket list from the XML response and format it for display. Retool's query data will contain the raw XML string in the data property. Alternatively, create a query targeting a specific bucket's contents: - Method: GET - Path: /YOUR_BUCKET_NAME (replace with your actual bucket name) - URL parameter: list-type → 2 (enables the ListObjectsV2 format) - URL parameter: max-keys → 100 - URL parameter: prefix → {{ prefixInput.value || '' }} The ListObjectsV2 response includes object key, size, last modified, and ETag for each object. Add a JavaScript transformer to parse the XML and extract the Contents array into clean table rows. Drag a Table component onto the canvas and bind its data to {{ objectsQuery.data }}. Add a Text Input component for prefix filtering — wire it to the prefix URL parameter in the query. Enable 'Run this query automatically when inputs change' so the table updates as users type a prefix filter.

1// Transformer to parse B2 S3-compatible ListObjectsV2 XML response
2// B2 returns XML; we parse it manually from the raw response string
3// Note: data here is the raw text response from B2
4const parser = new DOMParser();
5const xmlDoc = parser.parseFromString(data, 'application/xml');
6const contents = xmlDoc.getElementsByTagName('Contents');
7const objects = [];
8for (let i = 0; i < contents.length; i++) {
9  const item = contents[i];
10  const key = item.getElementsByTagName('Key')[0]?.textContent || '';
11  const sizeBytes = parseInt(item.getElementsByTagName('Size')[0]?.textContent || '0');
12  const lastModified = item.getElementsByTagName('LastModified')[0]?.textContent || '';
13  const etag = item.getElementsByTagName('ETag')[0]?.textContent || '';
14  const sizeMB = (sizeBytes / (1024 * 1024)).toFixed(2);
15  objects.push({
16    key,
17    size_bytes: sizeBytes,
18    size_display: sizeBytes > 1048576 ? sizeMB + ' MB' : (sizeBytes / 1024).toFixed(1) + ' KB',
19    last_modified: lastModified ? new Date(lastModified).toLocaleString() : '',
20    etag: etag.replace(/"/g, ''),
21    file_type: key.split('.').pop()?.toLowerCase() || ''
22  });
23}
24return objects;

Add the ability to generate download links for selected objects. B2 provides direct download URLs in the format: https://f002.backblazeb2.com/file/YOUR_BUCKET/YOUR_OBJECT_KEY (for public buckets), or signed download URLs (using the B2 native API's b2_get_download_authorization endpoint) for private buckets. For public bucket download links, create a JavaScript query (not a resource query) that constructs the download URL from the selected row: Create a new query, select 'Run JS Code' as the type: ``` const bucket = 'your-bucket-name'; const key = objectsTable.selectedRow?.key; if (!key) return ''; return `https://f002.backblazeb2.com/file/${bucket}/${encodeURIComponent(key)}`; ``` For object metadata, create a HEAD request query: - Method: HEAD - Path: /YOUR_BUCKET_NAME/{{ objectsTable.selectedRow?.key }} The HEAD response headers contain the object's content-type, content-length, last-modified, and any custom metadata stored as x-amz-meta- headers. Display these in a Container component on the right side of the canvas as a metadata panel. Drag a Button component labeled 'Copy Download URL' and wire its onClick event to copy the generated URL to the clipboard using Retool's built-in clipboard utility. Add a Text component that shows the URL as a clickable link when an object is selected.

1// JavaScript query to build B2 download URL for selected object
2// For public buckets — use B2 native download URL format
3const bucket = 'your-bucket-name'; // replace with your actual bucket name
4const region = 'us-west-004'; // replace with your bucket region
5const key = objectsTable.selectedRow?.key;
6if (!key) {
7  return { url: '', display: 'Select an object to generate a download URL' };
8}
9// B2 S3-compatible download URL format
10const encodedKey = key.split('/').map(encodeURIComponent).join('/');
11const downloadUrl = `https://s3.${region}.backblazeb2.com/${bucket}/${encodedKey}`;
12return { url: downloadUrl, display: downloadUrl };

Build a storage summary view by aggregating object counts and sizes across buckets. Since B2's S3-compatible API requires per-bucket queries (there is no cross-bucket aggregate endpoint), create separate queries for each bucket you want to monitor, or use a dynamic approach with a bucket list. Create a JavaScript query that reads from multiple bucket queries and computes totals: - Total objects across all buckets - Total storage in GB - Largest bucket by object count - Estimated monthly cost at B2's $0.006/GB/month pricing For bucket selection, populate a Select component with a static list of your bucket names (or from the ListBuckets query). Create a single objects query parameterized by the selected bucket name: path /{{ bucketSelect.value }}. This one query handles all buckets by switching the path based on the selected bucket. Add a Chart component showing storage distribution: a Bar Chart where each bar represents a bucket and the height shows total storage in GB. Populate this chart from a Transformer that combines data from each per-bucket query. For Retool Cloud deployments: add a text note to the overview panel reminding operators to whitelist Retool's IP ranges in any B2 bucket firewall rules if bucket access restrictions are configured. Self-hosted Retool deployments use their own egress IPs, which may need to be added to B2 IP allow lists as well.

1// Summary transformer — aggregate object counts and sizes from objectsQuery
2// Run after the main objects query, reads from objectsQuery.data
3const objects = objectsQuery.data || [];
4const totalObjects = objects.length;
5const totalBytes = objects.reduce((sum, obj) => sum + (obj.size_bytes || 0), 0);
6const totalGB = totalBytes / (1024 * 1024 * 1024);
7const estimatedMonthlyCost = totalGB * 0.006; // B2 pricing: $0.006/GB/month
8return {
9  total_objects: totalObjects.toLocaleString(),
10  total_size_gb: totalGB.toFixed(3) + ' GB',
11  total_size_tb: (totalGB / 1024).toFixed(6) + ' TB',
12  estimated_monthly_cost: '$' + estimatedMonthlyCost.toFixed(4),
13  largest_file: objects.reduce((max, obj) => obj.size_bytes > (max?.size_bytes || 0) ? obj : max, null)?.key || 'N/A'
14};