.github/workflows/codeql-analysis.yml

@@ -2,6 +2,7 @@ name: "Code scanning - action"
 
 on:
   push:
+  pull_request:
     paths-ignore:
       - '**.md'
   schedule:

README.md

@@ -219,11 +219,15 @@ Artifacts are retained for 90 days by default. You can specify a shorter retenti
 The retention period must be between 1 and 90 inclusive. For more information see [artifact and log retention policies](https://docs.github.com/en/free-pro-team@latest/actions/reference/usage-limits-billing-and-administration#artifact-and-log-retention-policy).
 
 ## Where does the upload go?
-In the top right corner of a workflow run, once the run is over, if you used this action, there will be an `Artifacts` dropdown which you can download items from. Here's a screenshot of what it looks like<br/>
-<img src="https://user-images.githubusercontent.com/16109154/72556687-20235a80-386d-11ea-9e2a-b534faa77083.png" width="375" height="140">
+
+At the bottom of the workflow summary page, there is a dedicated section for artifacts. Here's a screenshot of something you might see:
+
+<img src="https://user-images.githubusercontent.com/16109154/103645952-223c6880-4f59-11eb-8268-8dca6937b5f9.png" width="700" height="300">
 
 There is a trashcan icon that can be used to delete the artifact. This icon will only appear for users who have write permissions to the repository.
 
+The size of the artifact is denoted in bytes. The displayed artifact size denotes the raw uploaded artifact size (the sum of all the individual files uploaded during the workflow run for the artifact), not the compressed size. When you click to download an artifact from the summary page, a compressed zip is created with all the contents of the artifact and the size of the zip that you download may differ significantly from the displayed size. Billing is based on the raw uploaded size and not the size of the zip.
+
 # Limitations
 
 ### Zipped Artifact Downloads
@@ -252,7 +256,11 @@ If file permissions and case sensitivity are required, you can `tar` all of your
       name: my-artifact
       path: my_files.tar    
 ```
+### Too many uploads resulting in 429 responses
+
+A very minute subset of users who upload a very very large amount of artifacts in a short period of time may see their uploads throttled or fail because of `Request was blocked due to exceeding usage of resource 'DBCPU' in namespace` or `Unable to copy file to server StatusCode=TooManyRequests`.
 
+To reduce the chance of this happening, you can reduce the number of HTTP calls made during artifact upload by zipping or archiving the contents of your artifact before an upload starts. As an example, imagine an artifact with 1000 files (each 10 Kb in size). Without any modification, there would be around 1000 HTTP calls made to upload the artifact. If you zip or archive the artifact beforehand, the number of HTTP calls can be dropped to single digit territory. Measures like this will significantly speed up your upload and prevent uploads from being throttled or in some cases fail.
 
 ## Additional Documentation
 

package.json

@@ -29,7 +29,7 @@
   },
   "homepage": "https://github.com/actions/upload-artifact#readme",
   "dependencies": {
-    "@actions/artifact": "^0.5.0",
+    "@actions/artifact": "^0.5.1",
     "@actions/core": "^1.2.6",
     "@actions/glob": "^0.1.0",
     "@actions/io": "^1.0.2"

src/upload-artifact.ts

@@ -37,6 +37,12 @@ async function run(): Promise<void> {
       )
       core.debug(`Root artifact directory is ${searchResult.rootDirectory}`)
 
+      if (searchResult.filesToUpload.length > 10000) {
+        core.warning(
+          `There are over 10,000 files in this artifact, consider create an archive before upload to improve the upload performance.`
+        )
+      }
+
       const artifactClient = create()
       const options: UploadOptions = {
         continueOnError: false