<div dir="ltr">Hi,<div><br></div><div>I am running the latest release of the Einstein Toolkit with </div><div>Cactus::presync_mode = &quot;mixed-error&quot;. When I add the thorn<br></div><div>CarpetMask, the system returns an error: CarpetMask wants</div><div>sf_info from SphericalSurface, which is not available.</div><div><br></div><div>Looking at the scheduler, I can see why this is happening</div><div>(relevant section):</div><div><br></div><div><font face="monospace">        GROUP MaskBase_SetupMask: Set up the weight function<br>          GROUP MaskBase_SetupMaskAll: Set up the weight function<br>            CarpetReduce::MaskBase_AllocateMask: [global] Allocate the weight function<br>            CarpetReduce::MaskBase_InitMask: [global] [loop-local] Initialise the weight function<br>            GROUP SetupIMaskInternal: Set up the integer weight function (schedule other routines in here)<br>              CarpetReduce::CoordBase_SetupMask: [global] [loop-local] Set up the outer boundaries of the weight function<br>              CarpetReduce::CarpetMaskSetup: [global] [loop-singlemap] Set up the weight function for the restriction regions<br>            GROUP SetupIMask: Set up the integer weight function (schedule other routines in here)<br>              CarpetMask::CarpetExcludedSetup: [global] [loop-local] Set up the weight function for the excluded regions<br>              CarpetMask::CarpetSurfaceSetup: [global] [loop-local] Set up the weight function for the excluded spherical surfaces<br>            CarpetReduce::MaskBase_SetMask: [global] [loop-local] Set the weight function<br>            GROUP SetupMask: Set up the real weight function (schedule other routines in here)<br>            CarpetReduce::MaskBase_TestMask: [global] Test the weight function<br>        SphericalSurface::SphericalSurface_Set: [global] Set surface radii to be used for initial setup in other thorns<br>        GROUP SphericalSurface_HasBeenSet: Set the spherical surfaces before this group, and use it afterwards<br>          SphericalSurface::SphericalSurface_CheckState: [global] Test the state of the spherical surfaces<br></font></div><div><br></div><div>As you see, CarpetMask is scheduled before SphericalSurface_Set,</div><div>which is what sets sf_info. This is in the BASEGRID bin. </div><div><br></div><div>Since I haven&#39;t touched neither CarpetMask nor SphericalSurface, I </div><div>would think this is a bug, but I maybe I am doing something wrong</div><div>(I&#39;ve never used presync_mode before).</div><div><br></div><div>I also take the chance to point out two tickets Pedro Espino and I</div><div>opened but received no attention:</div><div><a href="https://bitbucket.org/einsteintoolkit/tickets/issues/2487/_evolution_method-leanbssnmol-in-lean-does">https://bitbucket.org/einsteintoolkit/tickets/issues/2487/_evolution_method-leanbssnmol-in-lean-does</a><br></div><div><a href="https://bitbucket.org/einsteintoolkit/tickets/issues/2484/lack-of-convergence-of-neutron-star">https://bitbucket.org/einsteintoolkit/tickets/issues/2484/lack-of-convergence-of-neutron-star</a><br></div><div><br></div><div>Best,<br>Gabriele</div></div>