I like Colif's reply. But beyond the "stop shaking!" advice, here's a thought. From time to time a HDD will encounter an error when reading. The normal action when that happens is that it does a small partial reset by stopping its current work, retracting the heads to the zero track to ensure it is positioned in a known place, then re-using the last co-ordinates to re-seek the Sector it was trying to find and read. Most often that succeeds, and it keeps on going. SOMETIMES it fails a second time, and the whole sequence will repeat. This will continue if failures continue, until a limit is reached and the HDD reports to the OS that there is an error of a Bad Sector that cannot be read.
It also is normal that this "Head Seek" process...