IssueTracker

This will create a public issue which anybody can view and comment on.

Please provide as much information as possible. At least, this should include a description of your issue and steps to reproduce the problem. If possible please provide a summary of what steps or workarounds you have already tried, and any docs or articles you found (un)helpful.

Problem you have encountered:

Since the 24th-25th of July 2024, the deployment of our App Engine app (standard, using nodejs18) is super long (went from 1-2 minutes to 11+ minutes) and often even never finishing the `npm ci --quiet` step in the build and thus failing the deployment. This happened without us changing anything in our code, at least from what we can see. No new packages were added, the package-lock.json hasn't changed since a few months.

One thing I noticed is that before the 24th-25th of July, the step to install the Node packages was written as executing `npm install --quiet (NODE_ENV=production)`, and now it writes that it executes `npm ci --quiet (NODE_ENV=production)`; I can see this in the corresponding Build details.

When the Build fails with a timeout (takes 1 hour), here is what I see at the end of the Build details:

```
Step #2 - "build": === Node.js - Npm (google.nodejs.npm@1.1.0) ===
Step #2 - "build": ***** CACHE MISS: "npm_modules"
Step #2 - "build": Installing application dependencies.
Step #2 - "build": --------------------------------------------------------------------------------
Step #2 - "build": Running "npm ci --quiet (NODE_ENV=production)"
Step #2 - "build": npm warn deprecated puppeteer@21.11.0: < 22.5.0 is no longer supported
TIMEOUT
ERROR: context deadline exceeded
```

The corresponding `gcloud app deploy` command fails with the following message: "ERROR: (gcloud.app.deploy) Error Response: [13] An internal error occurred.", and here is the exact command we are running (unchanged since a few months at least): `gcloud --quiet --project $GCP_PROJECT_ID app deploy app.yaml -v dev --no-cache`

Lastly, here is the content of app.yaml (also unchanged for a few months/years):

```
runtime: nodejs18
service: document-manager

automatic_scaling:
  max_instances: 2
  max_concurrent_requests: 80

instance_class: F4_1G
```

When I try to execute `npm ci` locally, using the same Node version as the Build (18.20.3 since I see the text "Using runtime version from GOOGLE_RUNTIME_VERSION: 18.20.3" in the Build details), it works in a few seconds.

What you expected to happen:

No change in time to deploy our App Engine app.


Steps to reproduce:


Other information (workarounds you have tried, documentation consulted, etc):

I also tried to remove the `--no-cache` in the `gcloud app deploy` command, but it still took a super long time to finish (10+ minutes).

Thank in advance for your help; this blocks our pipeline and new deployment 70-80% of the time, so it's quite urgent for us to find a workaround or the reason of that issue. Happy to provide more info if needed.