Failure to execute NONMEM from R

> On Dec 20, 2017, at 21:54, HUI, Ka Ho <[email protected]> wrote: > > Dear NMusers, > > Recently I need to execute NONMEM 7.4 from Rstudio terminal. I have tried the > following scripts: > `system2(“execute FIT_01.mod”)` #Through PsN, > using `system2` > `shell(“execute FIT_01.mod”)` #Through PsN, > using `shell` > `write(“execute FIT_01.mod”, “FIT_01.bat”)` > `shell(“FIT_01.bat”)` > #Through PsN, using `shell` to call a batch file > `write(“execute FIT_01.mod”, “FIT_01.bat”)` > `shell.exec(“FIT_01.bat”)` #Through > PsN, using `shell.exec` to call a batch file > The above commands, but call through C:/nm74g64/run/nmfe74.bat directly > > The above were also tested using the test dataset `CONTROL5` > > Results: > gave me the `running command '"execute FIT_01.mod"' had status 127` warning > without any execution > > (2)-(4) gave me the following messages: > Starting 1 NONMEM executions. 1 in parallel. > S:1 .. > All executions started. > Starting NMTRAN > NMtran failed. There is no output for model 1. Contents of FMSG: > Not restarting this model. > F:1 .. > execute done > > (5) similar to (1) but the status/error code became 107 > > None of the above has led to a successful NONMEM run. But the above command > lines work when outside R (i.e. through `cmd` +/- PsN) > Is there anyone experiencing the same issue? > > System/software info: > OS: Windows 10 > NONMEM v7.4 > R3.4.2 > Rstudio v1.1.383 > > Best regards, > Matthew Hui > PhD Student > School of Pharmacy > Faculty of Medicine > The Chinese University of Hong Kong >

On Wed, Dec 20, 2017 at 10:47 PM HUI, Ka Ho <[email protected]> wrote: > Thanks for attending the problem, Bill. > > > > Exactly which path should I be looking for? Am I to check if the desired > path exists in the list or not? > > I tried adding the paths for NONMEM, PsN, R and Rstudio into the list and > executed commands in R again – nothing has changed so far. Any idea? > > > > Thanks, > > Matthew > > > > *From:* Bill Denney [mailto:[email protected]] > *Sent:* Thursday, December 21, 2017 11:11 AM > *To:* HUI, Ka Ho <[email protected]> > *Cc:* [email protected] > *Subject:* Re: [NMusers] Failure to execute NONMEM from R > > > > Hi Matthew, > > > > Since everything is working when run directly from the command line (cmd), > I would guess that it's something like a path issue. > > > > Check your path at the command prompt (type "path") and in RStudio > ("Sys.getenv("PATH")"). > > > > Thanks, > > > > Bill > > > On Dec 20, 2017, at 21:54, HUI, Ka Ho <[email protected]> > wrote: > > Dear NMusers, > > > > Recently I need to execute NONMEM 7.4 from Rstudio terminal. I have tried > the following scripts: > > 1. `system2(“execute FIT_01.mod”)` #Through > PsN, using `system2` > 2. `shell(“execute FIT_01.mod”)` > #Through PsN, using `shell` > 3. `write(“execute FIT_01.mod”, “FIT_01.bat”)` > `shell(“FIT_01.bat”)` > #Through PsN, using `shell` to call a batch file > 4. `write(“execute FIT_01.mod”, “FIT_01.bat”)` > `shell.exec(“FIT_01.bat”)` > #Through PsN, using `shell.exec` to call a batch file > 5. The above commands, but call through C:/nm74g64/run/nmfe74.bat > directly > > > > The above were also tested using the test dataset `CONTROL5` > > > > Results: > > 1. gave me the `running command '"execute FIT_01.mod"' had status 127` > warning without any execution > > > > (2)-(4) gave me the following messages: > > Starting 1 NONMEM executions. 1 in parallel. > S:1 .. > All executions started. > Starting NMTRAN > NMtran failed. There is no output for model 1. Contents of FMSG: > > Not restarting this model. > F:1 .. > execute done > > (5) similar to (1) but the status/error code became 107 > > > > None of the above has led to a successful NONMEM run. But the above > command lines work when outside R (i.e. through `cmd` +/- PsN) > > Is there anyone experiencing the same issue? > > > > System/software info: > > OS: Windows 10 > > NONMEM v7.4 > > R3.4.2 > > Rstudio v1.1.383 > > > > Best regards, > > *Matthew Hui* > > PhD Student > > School of Pharmacy > > *Faculty of Medicine* > > *The Chinese University of Hong Kong* > > > >

On Thu, Dec 21, 2017 at 4:41 AM, HUI, Ka Ho <[email protected]> wrote: > Thanks for attending the problem, Bill. > > > > Exactly which path should I be looking for? Am I to check if the desired > path exists in the list or not? > > I tried adding the paths for NONMEM, PsN, R and Rstudio into the list and > executed commands in R again – nothing has changed so far. Any idea? > > > > Thanks, > > Matthew > > > > From: Bill Denney [mailto:[email protected]] > Sent: Thursday, December 21, 2017 11:11 AM > To: HUI, Ka Ho <[email protected]> > Cc: [email protected] > Subject: Re: [NMusers] Failure to execute NONMEM from R > > > > Hi Matthew, > > > > Since everything is working when run directly from the command line (cmd), I > would guess that it's something like a path issue. > > > > Check your path at the command prompt (type "path") and in RStudio > ("Sys.getenv("PATH")"). > > > > Thanks, > > > > Bill > > > On Dec 20, 2017, at 21:54, HUI, Ka Ho <[email protected]> wrote: > > Dear NMusers, > > > > Recently I need to execute NONMEM 7.4 from Rstudio terminal. I have tried > the following scripts: > > `system2(“execute FIT_01.mod”)` #Through PsN, > using `system2` > `shell(“execute FIT_01.mod”)` #Through PsN, > using `shell` > `write(“execute FIT_01.mod”, “FIT_01.bat”)` > `shell(“FIT_01.bat”)` > #Through PsN, using `shell` to call a batch file > `write(“execute FIT_01.mod”, “FIT_01.bat”)` > `shell.exec(“FIT_01.bat”)` #Through > PsN, using `shell.exec` to call a batch file > The above commands, but call through C:/nm74g64/run/nmfe74.bat directly > > > > The above were also tested using the test dataset `CONTROL5` > > > > Results: > > gave me the `running command '"execute FIT_01.mod"' had status 127` warning > without any execution > > > > (2)-(4) gave me the following messages: > > Starting 1 NONMEM executions. 1 in parallel. > S:1 .. > All executions started. > Starting NMTRAN > NMtran failed. There is no output for model 1. Contents of FMSG: > > Not restarting this model. > F:1 .. > execute done > > (5) similar to (1) but the status/error code became 107 > > > > None of the above has led to a successful NONMEM run. But the above command > lines work when outside R (i.e. through `cmd` +/- PsN) > > Is there anyone experiencing the same issue? > > > > System/software info: > > OS: Windows 10 > > NONMEM v7.4 > > R3.4.2 > > Rstudio v1.1.383 > > > > Best regards, > > Matthew Hui > > PhD Student > > School of Pharmacy > > Faculty of Medicine > > The Chinese University of Hong Kong > >

-----Original Message----- From: Tarj Sahota [mailto:[email protected]] Sent: Thursday, December 21, 2017 10:18 PM To: Devin Pastoor <[email protected]> Cc: HUI, Ka Ho <[email protected]>; Bill Denney <[email protected]>; [email protected] Subject: Re: [NMusers] Failure to execute NONMEM from R I get the same errors as Matthew (with NM 7.4.1+gfortran on windows) - I forgot to mention that in order to get it working on my machine I also need to use shell() rather than system()/system2()/... along with Sys.unsetenv("GFORTRAN_STDOUT_UNIT"). Not sure if that helps you, but it fixes the issue with my windows machine. Tarj. On Thu, Dec 21, 2017 at 11:49 AM, Devin Pastoor <[email protected]> wrote: > I believe the problem is actually that is cross referenced > http://discuss.go-isop.org/t/psn-execute-failed-to-start-when-called-f > rom-rstudio/1119 > It is not a path issue > > Mike and I have had a small amount of offline exploration and can > confirm it a pervasive issue with NM74 being called from Rstudio. This > is not just PsN, I also use a separate language (go) to manage the > nmfe commands and invoking that from Rstudio via system2/processx/sys > all fails in the same fashion. At this point, I believe it is likely a > permissions issue, but I am unsure of if the (primary) culprit is nonmem or > Rstudio. > > We will need to get Bob's input whether there was a change in the > mechanics of how nonmem is invoked from 73 --> 74 to hopefully narrow > the surface area of what could be causing the problem. > > On Wed, Dec 20, 2017 at 10:47 PM HUI, Ka Ho > <[email protected]> > wrote: >> >> Thanks for attending the problem, Bill. >> >> >> >> Exactly which path should I be looking for? Am I to check if the >> desired path exists in the list or not? >> >> I tried adding the paths for NONMEM, PsN, R and Rstudio into the list >> and executed commands in R again – nothing has changed so far. Any idea? >> >> >> >> Thanks, >> >> Matthew >> >> >> >> From: Bill Denney [mailto:[email protected]] >> Sent: Thursday, December 21, 2017 11:11 AM >> To: HUI, Ka Ho <[email protected]> >> Cc: [email protected] >> Subject: Re: [NMusers] Failure to execute NONMEM from R >> >> >> >> Hi Matthew, >> >> >> >> Since everything is working when run directly from the command line >> (cmd), I would guess that it's something like a path issue. >> >> >> >> Check your path at the command prompt (type "path") and in RStudio >> ("Sys.getenv("PATH")"). >> >> >> >> Thanks, >> >> >> >> Bill >> >> >> On Dec 20, 2017, at 21:54, HUI, Ka Ho <[email protected]> >> wrote: >> >> Dear NMusers, >> >> >> >> Recently I need to execute NONMEM 7.4 from Rstudio terminal. I have >> tried the following scripts: >> >> `system2(“execute FIT_01.mod”)` #Through PsN, >> using `system2` >> `shell(“execute FIT_01.mod”)` #Through >> PsN, using `shell` >> `write(“execute FIT_01.mod”, “FIT_01.bat”)` `shell(“FIT_01.bat”)` >> #Through PsN, using `shell` to call a batch file `write(“execute >> FIT_01.mod”, “FIT_01.bat”)` `shell.exec(“FIT_01.bat”)` #Through PsN, >> using `shell.exec` to call a batch file The above commands, but call >> through C:/nm74g64/run/nmfe74.bat directly >> >> >> >> The above were also tested using the test dataset `CONTROL5` >> >> >> >> Results: >> >> gave me the `running command '"execute FIT_01.mod"' had status 127` >> warning without any execution >> >> >> >> (2)-(4) gave me the following messages: >> >> Starting 1 NONMEM executions. 1 in parallel. >> S:1 .. >> All executions started. >> Starting NMTRAN >> NMtran failed. There is no output for model 1. Contents of FMSG: >> >> Not restarting this model. >> F:1 .. >> execute done >> >> (5) similar to (1) but the status/error code became 107 >> >> >> >> None of the above has led to a successful NONMEM run. But the above >> command lines work when outside R (i.e. through `cmd` +/- PsN) >> >> Is there anyone experiencing the same issue? >> >> >> >> System/software info: >> >> OS: Windows 10 >> >> NONMEM v7.4 >> >> R3.4.2 >> >> Rstudio v1.1.383 >> >> >> >> Best regards, >> >> Matthew Hui >> >> PhD Student >> >> School of Pharmacy >> >> Faculty of Medicine >> >> The Chinese University of Hong Kong >> >>

-----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of HUI, Ka Ho Sent: 22 December 2017 02:27 To: Tarj Sahota; Devin Pastoor Cc: Bill Denney; [email protected] Subject: [EXTERNAL] RE: [NMusers] Failure to execute NONMEM from R Thanks Tarj for your advice and it works! sys.unsetenv("GFORTRAN_STDOUT_UNIT") is the line that is needed. Matthew -----Original Message----- From: Tarj Sahota [mailto:[email protected]] Sent: Thursday, December 21, 2017 10:18 PM To: Devin Pastoor <[email protected]> Cc: HUI, Ka Ho <[email protected]>; Bill Denney <[email protected]>; [email protected] Subject: Re: [NMusers] Failure to execute NONMEM from R I get the same errors as Matthew (with NM 7.4.1+gfortran on windows) - I forgot to mention that in order to get it working on my machine I also need to use shell() rather than system()/system2()/... along with Sys.unsetenv("GFORTRAN_STDOUT_UNIT"). Not sure if that helps you, but it fixes the issue with my windows machine. Tarj. On Thu, Dec 21, 2017 at 11:49 AM, Devin Pastoor <[email protected]> wrote: > I believe the problem is actually that is cross referenced > http://discuss.go-isop.org/t/psn-execute-failed-to-start-when-called-f > rom-rstudio/1119 > It is not a path issue > > Mike and I have had a small amount of offline exploration and can > confirm it a pervasive issue with NM74 being called from Rstudio. This > is not just PsN, I also use a separate language (go) to manage the > nmfe commands and invoking that from Rstudio via system2/processx/sys > all fails in the same fashion. At this point, I believe it is likely a > permissions issue, but I am unsure of if the (primary) culprit is nonmem or > Rstudio. > > We will need to get Bob's input whether there was a change in the > mechanics of how nonmem is invoked from 73 --> 74 to hopefully narrow > the surface area of what could be causing the problem. > > On Wed, Dec 20, 2017 at 10:47 PM HUI, Ka Ho > <[email protected]> > wrote: >> >> Thanks for attending the problem, Bill. >> >> >> >> Exactly which path should I be looking for? Am I to check if the >> desired path exists in the list or not? >> >> I tried adding the paths for NONMEM, PsN, R and Rstudio into the list >> and executed commands in R again – nothing has changed so far. Any idea? >> >> >> >> Thanks, >> >> Matthew >> >> >> >> From: Bill Denney [mailto:[email protected]] >> Sent: Thursday, December 21, 2017 11:11 AM >> To: HUI, Ka Ho <[email protected]> >> Cc: [email protected] >> Subject: Re: [NMusers] Failure to execute NONMEM from R >> >> >> >> Hi Matthew, >> >> >> >> Since everything is working when run directly from the command line >> (cmd), I would guess that it's something like a path issue. >> >> >> >> Check your path at the command prompt (type "path") and in RStudio >> ("Sys.getenv("PATH")"). >> >> >> >> Thanks, >> >> >> >> Bill >> >> >> On Dec 20, 2017, at 21:54, HUI, Ka Ho <[email protected]> >> wrote: >> >> Dear NMusers, >> >> >> >> Recently I need to execute NONMEM 7.4 from Rstudio terminal. I have >> tried the following scripts: >> >> `system2(“execute FIT_01.mod”)` #Through PsN, >> using `system2` >> `shell(“execute FIT_01.mod”)` #Through >> PsN, using `shell` >> `write(“execute FIT_01.mod”, “FIT_01.bat”)` `shell(“FIT_01.bat”)` >> #Through PsN, using `shell` to call a batch file `write(“execute >> FIT_01.mod”, “FIT_01.bat”)` `shell.exec(“FIT_01.bat”)` #Through PsN, >> using `shell.exec` to call a batch file The above commands, but call >> through C:/nm74g64/run/nmfe74.bat directly >> >> >> >> The above were also tested using the test dataset `CONTROL5` >> >> >> >> Results: >> >> gave me the `running command '"execute FIT_01.mod"' had status 127` >> warning without any execution >> >> >> >> (2)-(4) gave me the following messages: >> >> Starting 1 NONMEM executions. 1 in parallel. >> S:1 .. >> All executions started. >> Starting NMTRAN >> NMtran failed. There is no output for model 1. Contents of FMSG: >> >> Not restarting this model. >> F:1 .. >> execute done >> >> (5) similar to (1) but the status/error code became 107 >> >> >> >> None of the above has led to a successful NONMEM run. But the above >> command lines work when outside R (i.e. through `cmd` +/- PsN) >> >> Is there anyone experiencing the same issue? >> >> >> >> System/software info: >> >> OS: Windows 10 >> >> NONMEM v7.4 >> >> R3.4.2 >> >> Rstudio v1.1.383 >> >> >> >> Best regards, >> >> Matthew Hui >> >> PhD Student >> >> School of Pharmacy >> >> Faculty of Medicine >> >> The Chinese University of Hong Kong >> >>

From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of Smith, Mike K Sent: Tuesday, January 02, 2018 2:52 AM To: HUI, Ka Ho; Tarj Sahota; Devin Pastoor Cc: Bill Denney; [email protected]<mailto:[email protected]> Subject: RE: [NMusers] Failure to execute NONMEM from R Likewise, glad to report that Sys.unsetenv("GFORTRAN_STDOUT_UNIT") works. You need to submit this command (preferably at the top of an R script) each time you start R. The question remains *WHY*??? And what is different between NONMEM-7.3 and NONMEM-7.4 (to Devin's point). I've managed to get NONMEM-7.4 to run using system, system2 and shell. system2(command = 'c:/nm74g64/run/nmfe74.bat', args = c('run1.mod', 'run1.lst'), stdout = TRUE, stderr = TRUE,wait = TRUE) cmd <- paste("c:/nm74g64/run/nmfe74.bat run1.mod run1.lst") system( command = cmd) or using PsN Execute: system2(command = 'c:/strawberry/perl/bin/execute-4.7.9.bat', args = c('run1.mod', '-nm_version="nonmem-7.4"'), stdout = TRUE, stderr = TRUE,wait = TRUE) -----Original Message----- From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of HUI, Ka Ho Sent: 22 December 2017 02:27 To: Tarj Sahota; Devin Pastoor Cc: Bill Denney; [email protected]<mailto:[email protected]> Subject: [EXTERNAL] RE: [NMusers] Failure to execute NONMEM from R Thanks Tarj for your advice and it works! sys.unsetenv("GFORTRAN_STDOUT_UNIT") is the line that is needed. Matthew -----Original Message----- From: Tarj Sahota [mailto:[email protected]] Sent: Thursday, December 21, 2017 10:18 PM To: Devin Pastoor <[email protected]<mailto:[email protected]>> Cc: HUI, Ka Ho <[email protected]<mailto:[email protected]>>; Bill Denney <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Subject: Re: [NMusers] Failure to execute NONMEM from R I get the same errors as Matthew (with NM 7.4.1+gfortran on windows) - I forgot to mention that in order to get it working on my machine I also need to use shell() rather than system()/system2()/... along with Sys.unsetenv("GFORTRAN_STDOUT_UNIT"). Not sure if that helps you, but it fixes the issue with my windows machine. Tarj. On Thu, Dec 21, 2017 at 11:49 AM, Devin Pastoor <[email protected]<mailto:[email protected]>> wrote: > I believe the problem is actually that is cross referenced > http://discuss.go-isop.org/t/psn-execute-failed-to-start-when-called-f > rom-rstudio/1119 > It is not a path issue > > Mike and I have had a small amount of offline exploration and can > confirm it a pervasive issue with NM74 being called from Rstudio. This > is not just PsN, I also use a separate language (go) to manage the > nmfe commands and invoking that from Rstudio via system2/processx/sys > all fails in the same fashion. At this point, I believe it is likely a > permissions issue, but I am unsure of if the (primary) culprit is nonmem or > Rstudio. > > We will need to get Bob's input whether there was a change in the > mechanics of how nonmem is invoked from 73 --> 74 to hopefully narrow > the surface area of what could be causing the problem. > > On Wed, Dec 20, 2017 at 10:47 PM HUI, Ka Ho > <[email protected]<mailto:[email protected]>> > wrote: >> >> Thanks for attending the problem, Bill. >> >> >> >> Exactly which path should I be looking for? Am I to check if the >> desired path exists in the list or not? >> >> I tried adding the paths for NONMEM, PsN, R and Rstudio into the list >> and executed commands in R again – nothing has changed so far. Any idea? >> >> >> >> Thanks, >> >> Matthew >> >> >> >> From: Bill Denney [mailto:[email protected]] >> Sent: Thursday, December 21, 2017 11:11 AM >> To: HUI, Ka Ho >> <[email protected]<mailto:[email protected]>> >> Cc: [email protected]<mailto:[email protected]> >> Subject: Re: [NMusers] Failure to execute NONMEM from R >> >> >> >> Hi Matthew, >> >> >> >> Since everything is working when run directly from the command line >> (cmd), I would guess that it's something like a path issue. >> >> >> >> Check your path at the command prompt (type "path") and in RStudio >> ("Sys.getenv("PATH")"). >> >> >> >> Thanks, >> >> >> >> Bill >> >> >> On Dec 20, 2017, at 21:54, HUI, Ka Ho >> <[email protected]<mailto:[email protected]>> >> wrote: >> >> Dear NMusers, >> >> >> >> Recently I need to execute NONMEM 7.4 from Rstudio terminal. I have >> tried the following scripts: >> >> `system2(“execute FIT_01.mod”)` #Through PsN, >> using `system2` >> `shell(“execute FIT_01.mod”)` #Through >> PsN, using `shell` >> `write(“execute FIT_01.mod”, “FIT_01.bat”)` `shell(“FIT_01.bat”)` >> #Through PsN, using `shell` to call a batch file `write(“execute >> FIT_01.mod”, “FIT_01.bat”)` `shell.exec(“FIT_01.bat”)` #Through PsN, >> using `shell.exec` to call a batch file The above commands, but call >> through C:/nm74g64/run/nmfe74.bat directly >> >> >> >> The above were also tested using the test dataset `CONTROL5` >> >> >> >> Results: >> >> gave me the `running command '"execute FIT_01.mod"' had status 127` >> warning without any execution >> >> >> >> (2)-(4) gave me the following messages: >> >> Starting 1 NONMEM executions. 1 in parallel. >> S:1 .. >> All executions started. >> Starting NMTRAN >> NMtran failed. There is no output for model 1. Contents of FMSG: >> >> Not restarting this model. >> F:1 .. >> execute done >> >> (5) similar to (1) but the status/error code became 107 >> >> >> >> None of the above has led to a successful NONMEM run. But the above >> command lines work when outside R (i.e. through `cmd` +/- PsN) >> >> Is there anyone experiencing the same issue? >> >> >> >> System/software info: >> >> OS: Windows 10 >> >> NONMEM v7.4 >> >> R3.4.2 >> >> Rstudio v1.1.383 >> >> >> >> Best regards, >> >> Matthew Hui >> >> PhD Student >> >> School of Pharmacy >> >> Faculty of Medicine >> >> The Chinese University of Hong Kong >> >>