Phoenix NLME Model Warnings

I get the following set of errors pretty frequently with the model I’m trying to run. its a PBPK model I coded and its worked prior, but I’ve made a change recently to estimate a specific new parameter and now getting all sorts of errors…clearly Phoenix does not appreciate my model.

anyway, any help in understanding what these errors mean may help me diagnose my error. thank you!

Elliot

Error in Population:
Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\14988 registry key, error 5, Access is denied.

Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\8932 registry key, error 5, Access is denied.

Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\10704 registry key, error 5, Access is denied.

Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\6268 registry key, error 5, Access is denied.

Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\8532 registry key, error 5, Access is denied.

Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\11536 registry key, error 5, Access is denied.

Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\8580 registry key, error 5, Access is denied.

Unable to open the HKEY_LOCAL_MACHINE\SOFTWARE\MPICH\SMPD\process\4404 registry key, error 5, Access is denied.

iMethod = 6
Eng 6 process 2 of 8 is alive
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc
sigmaLchol= 0.133705
Initial fcntheta1 value 196.909897
corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1
DLSODE- At T (=R1) and step size H (=R2), the
corrector convergence failed repeatedly
or with ABS(H) = HMIN
In above, R1 = 0.2101419299838E+01 R2 = 0.3172448663666E-15
DLSODE- ISTATE (=I1) illegal
In above message, I1 = -5
DLSODE- Run aborted.. apparent infinite loop
iMethod = 6
Eng 6 process 1 of 8 is alive
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc
sigmaLchol= 0.133705
Initial fcntheta1 value 196.909897
corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1
DLSODE- At T (=R1) and step size H (=R2), the
corrector convergence failed repeatedly
or with ABS(H) = HMIN
In above, R1 = 0.2101418427349E+01 R2 = 0.1307332705062E-14
DLSODE- ISTATE (=I1) illegal
In above message, I1 = -5
DLSODE- Run aborted.. apparent infinite loop
unable to read the cmd header on the pmi context, Error = -1
.
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
SuspendThread failed with error 5 for process 1:045B29C7-CA32-4e37-9432-023DDF5D97D6:‘mpiNLME7.exe @nlmeargs.txt /e -1 cols1.txt data1.txt out000001.txt’
unable to suspend process.
unable to read the cmd header on the pmi context, Error = -1
.
iMethod = 6
Eng 6 process 0 of 8 is alive
entering naive pooled engine 6, iteration limit= 1000
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

READFLAGS
environmental flags and variables
0 iflagnp
0 iflagrestart
1 norderAGQ
1 iflagfocehess
1 iflagverbose
1 Allow Gaussian Fit
1 METHODblup
10 NDIGITblup
1 METHODlagl
13 NDIGITlagl
1 iflagIEXP
0.00200000009 tolmodlinz
0 nrep_pcwres
500 Nsample
200 Nburn
0 iflaganagrad
1 iodelevelused
NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc

Initial thetas and bounds
i Thetalo(i) Thetainit(i) Thetahi(i)
1 -0.1000+101 0.1000E-02 0.1000+101
2 0.1000E-02 0.1337E+00 0.1000+101
sigmaLchol= 0.133705

initial packed param vector and typx vector
i packed(i) typx(i)
1 0.001000 0.001000
2 0.133705 0.133705

Initial fcntheta1 value 196.909897

Nsub from initxtheta= 3
Nobstotal = 41

corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
unable to read the cmd header on the pmi context, Error = -1
.
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
received kill command for a pmi context that doesn’t exist: unmatched id = 6
unable to read the cmd header on the pmi context, Error = -1
.
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
iMethod = 6
Eng 6 process 3 of 8 is alive
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc
sigmaLchol= 0.133705
Initial fcntheta1 value 196.909897
corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1
unable to read the cmd header on the pmi context, Error = -1
.
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
iMethod = 6
Eng 6 process 4 of 8 is alive
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc
sigmaLchol= 0.133705
Initial fcntheta1 value 196.909897
corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1
unable to read the cmd header on the pmi context, Error = -1
.
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
iMethod = 6
Eng 6 process 5 of 8 is alive
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc
sigmaLchol= 0.133705
Initial fcntheta1 value 196.909897
corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1
unable to read the cmd header on the pmi context, Error = -1
.
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
iMethod = 6
Eng 6 process 6 of 8 is alive
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc
sigmaLchol= 0.133705
Initial fcntheta1 value 196.909897
corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1
unable to read the cmd header on the pmi context, Error = -1
.
Error posting readv, An existing connection was forcibly closed by the remote host.(10054)
iMethod = 6
Eng 6 process 7 of 8 is alive
Starting Engine 6

Problem dimensions

Nsub = 3
getntheta()= 2
getnsigma()= 0
lenEta = 0
lenSigma = 1

NorderAGQ not applicable to this engine 6
NorderAGQ set to 1 but unchanged in nlmeflags.asc
sigmaLchol= 0.133705
Initial fcntheta1 value 196.909897
corresponding ELSobj with zeroed blups= 318.466834

Starting -loglikelihood = 196.909897

optimizing thetas, iteration= 1

job aborted:
rank: node: exit code[: error message]
0: RMTL2438PK17AF7.celerion.local: 123
1: RMTL2438PK17AF7.celerion.local: 0: process 1 exited without calling finalize
2: RMTL2438PK17AF7.celerion.local: 0: process 2 exited without calling finalize
3: RMTL2438PK17AF7.celerion.local: 123
4: RMTL2438PK17AF7.celerion.local: 123
5: RMTL2438PK17AF7.celerion.local: 123
6: RMTL2438PK17AF7.celerion.local: 123
7: RMTL2438PK17AF7.celerion.local: 123
Error: Exception: Fortran Exception

NlmeErrorExit: close files and exit(0)

Error: Exception: Fortran Exception

NlmeErrorExit: close files and exit(0)

Model execution failed or cancelled

It will be good to revert back to the working code and then do your changes step by step to see where the code breaks.

Elliot thank you for sharing the project. It seems the model runs when Max ODE is matrix exponent but not in stiff.
Indeed the model is stiff and we might not have enough data to support it.

Another avenue is trying to tweak some of the advanced parameters in the engine:

like ODE rel Tol Abs Tol or max step. This really problem dependent.