IPP Software Navigation Tools IPP Links Communication Pan-STARRS Links

Opened 13 years ago

#1511 new defect

psphotStack assertion failure no psf

Reported by: bills Owned by: eugene
Priority: normal Milestone:
Component: psphot Version: current
Severity: normal Keywords: staticsky
Cc:

Description

staticskyRun 241706 and others pop an assertion failure

missing psf?
Backtrace depth: 8
Backtrace 0: p_psAssert
Backtrace 1: psphotResetModelsReadout
Backtrace 2: psphotResetModels
Backtrace 3: (unknown)
Backtrace 4: (unknown)
Backtrace 5: (unknown)
Backtrace 6: libc_start_main
Backtrace 7: (unknown)

elements.
-> pmGrowthCurveGenerate (pmGrowthCurveGenerate.c:121): unknown psLib error

failure to measure stats

-> psphotMakeGrowthCurve (psphotMakeGrowthCurve.c:50): Problem with aperture photometry

Fitting aperture corrections

-> psphotChoosePSFReadout (psphotChoosePSF.c:361): Problem in PSF

Unable to construct flux scale for PSF

-> psphotChoosePSF (psphotChoosePSF.c:25): Problem in configure files

failed to choose a psf model for PSPHOT.STACK.OUTPUT.IMAGE entry 1

The root cause of the assert is that psphotStackReadout does not insure that psphotChoosePSF succeeds before proceeding in the radial apertures loop.

psphotResetModels expects the psf to exist on readout->analysis and pops the assertion.

The reason that psphotChoosePSF is failing is that none of the apMag values in pmGrowthCurveGenerate measured are finite.

I'll update the code to check before calling psphotVectorStats and print a better error message.

The error happens when working on the g band input convolved stack from stack 1068530 after convolving to 6pixel fwhm. The input image looks pretty bad the further convolved image looks worse.

There is a lot of ringing from deconvolution.

This error is seen in 8 of LAP staticsky runs out of the 60000 run so far

Change History (0)

Note: See TracTickets for help on using tickets.