Trouble Shooting

This troubleshooting guide can help you diagnose and resolve problems that you might encounter when running JMP Clinical software. If problems persist after troubleshooting, contact JMP Technical Support at support@jmp.com . Select Genomics > Documentation and Help > Contacting Technical Support or Clinical > Documentation and Help > Contacting Technical Support in JMP, for additional information.

Report	Problem	Suggested Cause/Resolution
Installation of JMP Clinical	The message: Existing Client Found is displayed in the Install Shield Wizard , indicating that a preexisting copy of SAS has been found on a network server.	A pre-existing copy of SAS has been found that is configured to run as a thin Client from a network server. JMP Genomics and JMP Clinical will work with a personal copy of SAS loaded on the same Client machine only and configured to work locally. Contact JMP Technical Support ( support@jmp.com ) for instructions and assistance in resolving this problem.
Opening JMP Clinical software	A message states that your license has expired.	Your license to run this software has expired. You must obtain updated license files to renew your license. You need two license files for JMP Clinical software: one to update the JMP license and one to update the SAS license. Most likely, you have received these files in an email. If not, send an email to support@jmp.com to request updated license files. The license files must be saved to your local hard drive. To update the JMP license, complete the following steps: 1 Start JMP. 2 JMP opens a dialog asking if you want to update your license. Click Yes to open the license renewal tool. 3 Click Open License . 4 Navigate to the folder in which you saved the license files and click Open . To update the SAS license, complete the following steps: 1 Click Start > Programs > SAS > SAS 9.3 License Renewal & Utilities > Renew SAS Software to open the renewal tool. 2 Navigate to the folder in which you saved the license files and click OK . 3 When prompted click Renew . 4 Click OK in the confirmation .
Moving data between JMP Clinical (32 bit) and JMP Genomics/Clinical (64 bit)	Processes take significantly more time to run than previously observed.	Apply the hot fix found at http://ftp.sas.com/techsup/download/hotfix/HF2/B25.html#B25026 .
Add Study from SDD	An ERROR message indicates that the user cannot be authenticated.	This error is shown when either the specified SDD user credentials or the specified SDD URL are incorrect. To resolve this problem, you should: • Verify that your SDD credentials are correct. Correct if necessary. • Make sure that the SDD URL, specified using the Configure Life Sciences Settings report, is correct. Click File > Life Sciences Settings and click the Integrations tab to inspect the specified URL. Correct if necessary.
Any report in using SAS DATA composed in Microsoft Word and pasted into JMP	A syntax error is displayed.	Microsoft Word uses a Smart Quote feature that converts quotation marks from the format needed by SAS into a format more appropriate for English language text. SAS DATA step commands should either be composed in a simple text editor and then pasted into the JMP Genomics/Clinical DATA Step input fields or directly in the DATA Step input fields. DATA Step commands should never be composed using Microsoft Word.
Any JMP report that uses one or more SAS programs	A SAS log is displayed in your JMP session along with a message preceded by ERROR .	The generated SAS code might not complete successfully because of mis-specified parameters. Most of the error messages should be self-explanatory and provide some idea about what to do next. If not, examine the broader context provided by the SAS log to determine the problem. If this fails, consult and search the SAS documentation for the SAS code generating the error by clicking Help > SAS Documentation - Local or Help > SAS Documentation - Web . There is also the possibility of a bug in the SAS macro code. If you have found what appears to be a bug, please send the SAS log and explanation to support@jmp.com . Please describe your procedure in sufficient detail for us to reproduce the problem. If you are a SAS programmer, you might wish to view and even edit the original SAS code in the ProcessLibrary and/or MacroLib folders. Please also feel free to send suggested changes to the code to support@jmp.com .
	A WARNING dialog appears, telling you that SAS is connected and a report is already running.	JMP can run one only report at a time and does not queue jobs. Click OK in the dialog to wait, disregard the second Run, and let the current report continue running. Click View Log to view the current SAS log to get information about the current report. Click Stop to stop the current report. If the SAS process does not stop in a short period of time, it is okay to kill the sas.exe process directly from s Task Manager, and then click Stop again.
	A report runs longer than expected or produces no output.	In this situation, perform the following steps: 1 Click Run again. A WARNING: SAS is Connected should appear. 2 Click View Log . If any SAS ERROR messages appear, click Stop to stop the current report. If not, proceed to the step below. 3 View the SAS log that is displayed in the JMP Log to see the most recently executed code. You can continue to click View Log as many times as you like to check the status of the SAS program. Alternatively, you can monitor generated file activity in the SAS working folder. To open this folder while the analysis is running, click Open SAS Temporary Folder . You should see various files being generated as the report runs. On s, press F5 to refresh the folder while you are monitoring it. 4 Open the s Task Manager (press <Ctrl+Alt+Delete> ), select the PROCESSES tab and watch sas.exe for activity (Click View>Select Columns... to add I/O Reads , I/O Writes , and I/O Other as columns to view). If these steps do not help, try running the report in the SAS 9.3 Display Manager as described below.
Any JMP Clinical report	Output of the report does not automatically open.	The output filename might contain the following characters: (), @, ^ and &, any place of output name, or contains [] at the beginning of the name (such as [name], for example). If these characters are present, you can open the output by completing the following steps: 1 Navigate to the specified output folder. 2 Double-click on the sasclean.jsl script in the folder. All of the output should open.
	An ERROR message is displayed stating: User does not have appropriate authorization level for library OUTLIB .	This message is most frequently generated when certain general directories, such as the Desktop or My Documents , are specified as the output folder. To resolve this problem, specify another directory as the output folder and run the report again.
Processes that perform repetitive computations	The SAS log is too long and is truncated.	In either of these cases, an alternative way to debug the report is to open the .sas file in the SAS 9.3 Display Manager (right-click and select Open with SAS 9.3 ) and run it from there by pressing F3 . The SAS Display Manager provides options for saving or deleting sections of long logs. On s operating systems, you can alternatively right-click on a .sas file and select Submit to SAS 9.3 . SAS then runs in batch mode and produce . log and . lst files.
Processes that specify a lot of variables in one macro	The line length can become too long for SAS batch mode.
Processes using wide data sets composed of long lists of variables	Numerous ERROR messages are generated in the SAS log.	The SAS Macro text expression limit of 65534 bytes might have been exceeded. Work-arounds for this situation include the following: 1 Re-create the data set or rename the variables to have the shortest possible names. 2 Modify the report specification to have list-style input for long lists of variables, such as Col1-Col20000 . 3 Reduce the number of variables using K-Means Clustering, as follows. Transpose the data to tall form using Transpose Rectangular , run K-Means Clustering to generate a few thousand or fewer clusters, retain representatives from each cluster to use as the data, and then transpose back to wide form using Transpose Rectangular .
Opening a data file using the File > Open command in any JMP Clinical report	The column names listed in the Available Variables box of a dialog appear different from the original column names in the data set.	SAS uses two ways to name a column: the variable label and the variable name. When a file is opened using the File > Open command from the JMP menu, SAS variable labels are displayed. These might differ from those displayed in the Available Variables list in the JMP report dialogs, which display SAS variable names for the available variables. To solve the problem, open the data file using the Open button on the report dialogs. This displays the table with names the same as those in the Available Variables lists. Alternatively, use the File > Open command from the JMP menu and, in the Open Data File dialog, change File of type to SAS Data Sets and click the Use SAS Variable Names for Column Names check box.
DM DIstribution	Process does not complete (restricted to Chinese or Japanese versions of JMP Clinical on machines running an English-language OS).	A specified filter contains characters coded by double-byte character sets (DBCS). Note : This report cannot be run on machines running English-language operating systems if specified filters contain double-byte character sets (DBCS). You should specify filters coded in single bytes, such as letters or numbers, only.
AE Bayesian Hierarchical Model	Results generated using JMP Clinical (v5.0 and earlier) do not match results expected using published methods.	An error exists in the underlying JMP Clinical code. The code in these versions reads: random Bi ~ binary(P) subject = s monitor =(Bi) ;. Instead, this line of code should be: random Bi ~ binary(P) subject = e monitor =(Bi); . This bug has been corrected in JMP Clinical v5.1. You should update JMP Clinical to v5.1.
AE Time to Event	The following ERROR message is generated: ERROR: NEWRAP Optimization cannot be completed.	You have specified Negative Binomial as the Distribution type for Incidence Density Modeling but the actual distribution of data does not approximate a negative binomial. If this error affects only a small number of subjects, 1 you can opt to ignore it. If the error affects a large number of subjects, you can either change the distribution type to Poisson or increase the Gradient Convergence Criterion value.
Any ANOVA report	After running an ANOVA or One-way ANOVA the following ERROR message is generated when you click Create a Subset with Mean Difference and P-Value Criteria, and choose a single comparison: Note: There is no significant difference found. You can change the cutoff values and recreate the subset table.	This can occur even when significant differences exist. To fix this problem, open web browser and navigate to the JMP Problem Note 46935 page and follow the instructions to download and deploy a patch file.
Uninstalling JMP Clinical software	Files in the SAS Home directory cannot be deleted after running the Uninstaller.	The SAS service might still be running in the background. To stop the SAS service: 1. Open Control panel > Administrative Tools > Services and scroll down and find the first service. It should be named as SASxxx , where xxx indicates an individual SAS service. 2. Double click SASxxx to open the panel, highlight, and copy the service name. 3. Open the Starter , and enter cmd in the text box and click Enter to open a command prompt. 4. At the prompt, type sc delete SASxxx and click Enter to delete SASxxx . 5. Repeat steps 1-4 for all of the listed SAS services. 6. Restart your computer and proceed with the uninstall process. Note : You should delete all the SAS services before uninstalling SAS.

---

1

Examine the SAS log to determine the number of subjects affected.