The Chandra X-ray Observatory has detected relatively hard X-ray emission from the central stars of several planetary nebulae (PNe). A subset has no known late-type companions, making it very difficult to isolate which of several competing mechanisms may be producing the X-ray emission. The central star of NGC 2392 is one of the most vexing members, with substantial indirect evidence for a hot white dwarf (WD) companion. Here we report on the results of a radial velocity (RV) monitoring campaign of its central star with the HERMES échelle spectrograph of the Flemish 1.2 m Mercator telescope. We discover a single-lined spectroscopic binary with an orbital period of 1.902208 ± 0.000013 d and an RV semi-amplitude of 9.96 ± 0.13 km s−1. The high degree of nebula ionisation requires a WD companion (M ≳ 0.6M⊙), which the mass-function supports at orbital inclinations ≲ 7°, in agreement with the nebula orientation of 9°. The hard component of the X-ray spectrum may be explained by the companion accreting mass from the wind of the Roche lobe filling primary, while the softer component may be due to colliding winds. A companion with a stronger wind than the primary could produce the latter and would be consistent with models of the observed diffuse X-ray emission detected in the nebula. The diffuse X-rays may also be powered by the jets of up to 180 km s−1, and active accretion would imply that they may be the first active jets of a post-common-envelope PN, potentially making NGC 2392 an invaluable laboratory to study jet formation physics. The 1.9 d orbital period rules out a double-degenerate merger leading to a Type Ia supernova, and the weak wind of the primary likely also precludes a single-degenerate scenario. We suggest that a hard X-ray spectrum, in the absence of a late-type companion, could be a powerful tool to identify accreting WD companions.