© 2018 American Physical Society. The GRAPES-3 muon telescope in Ooty, India had claimed detection of a 2 hour (h) high-energy (∼20 GeV) burst of galactic cosmic-rays (GCRs) through a >50σ surge in GeV muons, was caused by reconnection of the interplanetary magnetic field (IMF) in the magnetosphere that led to transient weakening of Earth's magnetic shield. This burst had occurred during a G4-class geomagnetic storm (storm) with a delay of 12h relative to the coronal mass ejection (CME) of 22 June 2015 [P. K. Mohanty et al., Phys. Rev. Lett. 117, 171101 (2016)PRLTAO0031-900710.1103/PhysRevLett.117.171101]. However, recently a group interpreted the occurrence of the same burst in a subset of 31 neutron monitors (NMs) to have been the result of an anisotropy in interplanetary space [P. Evenson et al., Proc. Sci., ICRC2017 (2017) 133] in contrast to the claim in P. K. Mohanty et al., [Phys. Rev. Lett. 117, 171101 (2016)PRLTAO0031-900710.1103/PhysRevLett.117.171101]. A new analysis of the GRAPES-3 data with a fine 10.6° angular segmentation shows the speculation of interplanetary anisotropy to be incorrect, and offers a possible explanation of the NM observations. The observed 28 minutes (min) delay of the burst relative to the CME can be explained by the movement of the reconnection front from the bow shock to the surface of Earth at an average speed of 35 km/s, much lower than the CME speed of 700 km/s. This measurement may provide a more accurate estimate of the start of the storm.