After one halted attempt at docking with the ISS, the CRS-10 Dragon capsule successfully completed the maneuver. In the time between the two attempts, Dragon must have maintained some kind of orbit with some relationship to ISS's orbit, to make a time for the next attempt available and yet to remain a safe distance away from the ISS without need of active control.
What orbit did CRS-10 maintain during the wait for the 2nd attempt, and how did it move relative to the ISS during that time?
In the reference frame of the ISS, the course looks approximately like this:
(Note: distances not to scale.)
In an earth-bound reference frame, this is simply a pair of Hohmann transfers between two orbits below and above the ISS:
As the spacecraft was below the ISS, its orbital period was slightly shorter than ISS' orbit, so it eventually came out in a safe distance ahead (and still below) the ISS.
A slight prograde burn raised the apogee of its orbit above the ISS. In the apogee, a slight prograde burn circularized the orbit.
Now being in an orbit above the ISS with an orbital period was slightly longer than ISS' orbit, it eventually came out behind (and still above) the ISS.
Now another pair of burns, this times retrograde, lowered the orbit back below the ISS, still behind but chasing up, ready for another rendezvous attempt.
Note well: I'm using the terms "chaser" and "target" in this answer. These are technical terms with a long heritage (relatively to the space era). The chaser vehicle acts to change its orbit. The target vehicle keeps its orbit.
Suppose the chaser and target are orbiting an object with no atmosphere, and that the chaser is in the same orbital plane as is the target, and that the chaser has the same semi-major axis as the target. Tweak the chaser's other orbital parameters just right and the chaser will follow a more or less oval shape about the target. Think NASCAR, where drivers only turn left. (I prefer non-oval races. Then again, F1 is boring.)
Tweak it just right and, ignoring atmospheric drag, the chaser will "orbit" the target in an oval centered on but always missing the target. Atmospheric drag is not something to ignore. The vehicles that bring crew and resupplies to the ISS are bullets. In comparison, the ISS is a huge butterfly. Fortunately, the ISS is a cooperative target. (Other targets: Not so much.)
Given that the Internet is chock full of mentions about racetracks but never mentions the details, I gather that those details are not to be released. But the summary description as an oval where the vehicle only turns in one direction thanks to gravitation: that's just Kepler's laws.
Note well: If the DoD had it's druthers, $F=ma$ and $F=GMm/r^2$ would be classified as TS NOFORN, as would be Kepler's laws. That's one of those details.
The answer is "essentially the same orbit". When you're that close to another orbiting object, slight differences in distance from the body being orbited have only a tiny impact on orbital velocity and thus rate of drift away from the rendezvous target.
You enter a lower or higher orbit to catch the object you're trying to rendezvous with, then "approximately the same orbit" when you're preparing for docking.
