Crashing is a way for hackers to get a series of users who can log in by collecting information about users and passwords that have been leaked on the Internet, generating a corresponding dictionary table, and trying to log in to other websites in bulk. Many users use the same account and password on different sites, so hackers can try to log in to site B by obtaining the user's account on site A, which can be interpreted as a crash attack.

Crashing can be protected by big data security techniques, such as: using data asset combing to discover sensitive data, using database encryption to protect core data, and using database security operations and maintenance to prevent operations and maintenance personnel from crashing attacks.

Mentioning the crash database, we can't leave out the towing and washing of the database, which is briefly introduced for you.

Drag library refers to the hacker invasion of valuable network sites, the registered user's information database all stolen behavior, because of the harmonic, also often referred to as stripping pants, 360's library with the plan to reward the submission of vulnerabilities of white hats, is also named. After obtaining a large amount of user data, hackers will realize the valuable user data through a series of technical means and the black industry chain, which is often called washing the library. Finally, hackers will get the data to try to log in on other websites, which is called bumping, because many users like to use a unified username and password, and bumping can also make hackers gain a lot of money.

Crashing method is as follows:

1, using n password dictionary to crash m accounts, the appearance of this is that an account in a certain short period of time, there may be many password attempts. So, you can add restrictions at the account level, e.g., an account with more than 5 wrong passwords in a day will be banned from logging in for 1 day.

2, with a few passwords to hit n accounts, the appearance of this is that the frequency of passwords will be very high, so you can count the number of times each password is wrong for a period of time, more than a certain threshold, this password is prohibited from logging in for a period of time.

3, with n groups of one-to-one account passwords to crash again, this case of crash purely from the account, password dimension, there will be no obvious anomalies.