Question: How complicated a metric can KillingVectors cope with?

I have been computing curvature etc of a four-dimensional metric that invloves three arbitary functions using the DifferentialGeometry package. I am now intereted in whether Maple can compute the Killing vectors. I have used the code of Example 3 from the Maple help page on KillingVectors in the  Tensor subpackage of the package DifferentialGeoemtry. If I set the arbitrary functions to be constants using the' auxiliaryequations' part of the code, then Maple outputs Killing vectors, so all good so far. If I leave the three arbitary functions to be arbitrary, however, specifying only that they be nonzero (as in Example 3) Maple had not produced any output after five hours, though when I moved the cursor over the tool bars at the top of the screen it turned to an hour glass, indicating that Maple was busy. Does anyone have experience with whether I am asking too much of Maple or whether I just need to give it more time to produce answers? If the latter, how long? The Example 3 indicates Maple will consider several possibilties including the completely arbitrary option, though that example only involves one arbitary function, so having multiple arbitary functions obviously increases the complexity.

Please Wait...