The diode's operating characteristics have changed in one or both: (A) forward resistance, (B) reverse volt rating.
The 1N4007 is rated at 1000V. Your schematic exposes each diode to 325V. So item B should not have changed, unless the PN junction has been stressed.
If A increases greatly, then that would explain the results you got by testing the diode individually.
If rev V rating goes down, it explains the results when you used the entire bridge.
Did any diode heat up, or smoke? That would be expected when something goes wrong at 230 VAC.
Each diode must be checked individually. In both directions. It is easy to test it with forward voltage. However to test it for high voltage you must subject it to high voltage. This is risky as you know.
The capacitor value is not so high that we would expect a great deal of surge current on power-up. However if it was exposed to the wrong polarity at any time, then it can be damaged.
As a test, you can try creating a half-wave supply with just one diode and the capacitor. The capacitor should acquire 325 VDC.
When you disconnect the diode, the capacitor should hold that charge, and only lose it slowly.
Discharge the capacitor each time before you test the next diode. Attach a safe load such as a small bulb rated for that volt level.
These are all hazardous things to do when you work with high voltage. If it were me I would hold a clear plastic shield between me and the components.