How to Fix FileNotFoundError: [Errno 2] No Such File or Directory in Python

What the error tells you

A typical traceback looks like this:

Traceback (most recent call last):
  File "/srv/app/app.py", line 19, in config
    with open(CONFIG_PATH, 'r') as f:
FileNotFoundError: [Errno 2] No such file or directory: '/srv/app/config.json'

Three facts are in that message:

• Errno 2 is the operating system error ENOENT - "no entry". The kernel could not find anything at that path.
• The quoted string at the end is the exact path Python asked for. If it is relative (for example 'config.json'), Python resolved it against the current working directory, not against the script's folder.
• The line number shows which open() call failed.

The error does not say why the file is missing. That is your job.

Step 1: reproduce it and read the log

If the error happens inside a web app, the HTTP response is only a 500. The traceback is in the application log:

curl -i http://localhost:8080/config
tail -n 20 /var/log/app/app.log

Copy the path from the last line of the traceback.

Step 2: compare the path with what is on disk

ls -la /srv/app/

Look for a file with a similar name. Common patterns:

• config.json.bak, config.json.old, config.json~ - someone renamed it before a change and never renamed it back.
• Config.json - Linux file names are case-sensitive, so Config.json and config.json are different files.
• config.json - a trailing space in the name. ls -b shows hidden characters.

If the file is not in that folder at all, search for it:

find / -name 'config.json*' -not -path '/proc/*' 2>/dev/null

Step 3: put the file back

When you find the file, move it back to the path the code expects:

mv /srv/app/config.json.bak /srv/app/config.json

If the app opens the file on every request (as in the traceback above), the next request works with no restart. If the app reads the file only once at startup, restart the process after you restore the file.

Then confirm:

curl -i http://localhost:8080/config     # expect HTTP 200

Relative paths: the other common cause

If the path in the error is relative, the file may exist, but Python is looking in a different directory. This happens when a script is started from cron, systemd, or a different shell folder:

# Breaks when the working directory is not the script's folder
open('config.json')

# Works from any working directory
from pathlib import Path
BASE = Path(__file__).resolve().parent
open(BASE / 'config.json')

Print os.getcwd() next to the failing open() to see which directory Python is really using.

Make the app fail early and clearly

An app that only fails when a user hits an endpoint is hard to debug. Check required files at startup and exit with a clear message:

import sys
from pathlib import Path

CONFIG_PATH = Path('/srv/app/config.json')
if not CONFIG_PATH.is_file():
    sys.exit(f'FATAL: config file missing: {CONFIG_PATH}')

Now a missing file stops the deploy right away, with the path in the log, instead of returning 500 to users hours later.

Debugging habit: what changed outside the code?

When "nothing changed in the code" but the app broke overnight, look at the things around the code: file names, permissions, disk space, environment variables, and DNS. Logs plus ls and find answer most of these questions faster than reading code that did not change.