Skip to content

Commit

Permalink
Adding Troubleshooting information
Browse files Browse the repository at this point in the history
  • Loading branch information
skarunan committed Feb 7, 2019
1 parent acbccf7 commit c71b453
Show file tree
Hide file tree
Showing 2 changed files with 52 additions and 14 deletions.
22 changes: 8 additions & 14 deletions docs/Installation.rst
Original file line number Diff line number Diff line change
Expand Up @@ -19,15 +19,11 @@ An example command to use RAPID as a conda environment : ::

conda create --name <environment_name> rapid=<version>

This command creates an environment for RAPID. We advise to use conda environment based approach, as this would not disturb your exisiting installations, and use only the compatible versions of dependencies.
This command creates an environment for RAPID. We advise to use conda environment based approach, as this would not disturb your existing installations, and use only the compatible versions of dependencies.

You can find the location of environments in the file ~/.conda/environments.txt

If you wish to test the installation, download the testData folder from the git repository `RAPID <https://github.com/SchulzLab/RAPID>`_.

Move to the test data folder. Edit the *rapid* path variable in the runTest.sh script. Depending on your OS, and conda installation directory, it should look soemthing like ::

rapid=/home/<username>/miniconda2/envs/<environment_name>/bin/ (or) rapid=/Users/<username>/miniconda2/envs/<environment_name>/bin/
Please refer to **TroubleShooting_FAQs** section, if you encounter issues.


First activate the desired conda environment ::
Expand All @@ -38,12 +34,10 @@ Now, simply run ::

bash runTest.sh
Upon succesful completion, there should be two folders TestRapid created by **rapidStats**, and TestCompare from **rapidNorm** in the testData folder.
You should also find the outputs described under the **Visualization** section.

If there are some errors with R modules, Open R by staying in the conda environment, and update the R modules ::
Upon successful completion, there should be two folders TestRapid created by **rapidStats**, and TestCompare from **rapidNorm** in the testData folder.
You should see outputs as shown in the SampleOutput folder. The output descriptions can be found in the **Visualization** section.

update.packages(ask = FALSE, checkBuilt = TRUE)
If you encounter any issues, which is not addressed in the **TroubleShooting_FAQs** section, please report to us.

To move out of the environment, type ::

Expand All @@ -70,10 +64,10 @@ RAPID makes use of the following tools, and requires them to be in your PATH.

Simple Test
-----------
After installation you can try running RAPID using the provided script runTest.sh in the testData folder. Ensure to set the *rapid* path variable in the script.
After installation you can try running RAPID using the provided script runTest.sh (You will have to uncomment respective lines) in the testData folder. Ensure to add the rapid in the PATH variable or provide the scripts with an environment variable for rapid.
Now, simply run ::

bash runTest.sh

Upon succesful completion, there should be two folders TestRapid created by **rapidStats**, and TestCompare from **rapidNorm** in the testData folder.
You should also find the basic statistics output described under the **Visualization** section.
Upon successful completion, there should be two folders TestRapid created by **rapidStats**, and TestCompare from **rapidNorm** in the testData folder.
You should see outputs as shown in the SampleOutput folder. The output descriptions can be found in the **Visualization** section.
44 changes: 44 additions & 0 deletions docs/TroubleShooting_FAQs.rst
Original file line number Diff line number Diff line change
@@ -0,0 +1,44 @@
Troubleshooting
===============

This part of the document will address FAQs, known issues, and work arounds.
Please do report to us of issues not listed here, and we will try to keep this document up-to-date.


Conda related
-----------------

Known issues related to conda based installation/execution.

* Library not loaded (libgfortran.3) *
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

This issue was encountered in the following scenario: ::

Conda version: 4.6.2
OSX version: 10.14.4

Error:

Abort trap: 6
dyld: Library not loaded: @rpath/libgfortran.3.dylib
Referenced from: /Users/...../rapid08/lib/R/lib/libRblas.dylib
Reason: image not found

This seems to be an issue with MacOS, and conda usage after recent updates. For some reason these fortran libraries are getting disabled when you activate the conda environment. The workaround is to change the order of PATH variable to access /usr/local/bin, and /usr/bin folders before the Conda environments bin folder.

For instance, the following PATH variable: ::

/Users/xxx/miniconda2/envs/rapid/bin:/Users/xxx/miniconda2/condabin:/Users/xxx/miniconda2/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/X11/bin

Should be changed to: ::

export PATH=/Users/xxx/miniconda2/condabin:/Users/xxx/miniconda2/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/X11/bin:/Users/xxx/miniconda2/envs/rapid/bin


R - related
-------------

If there are some errors with R modules, Open R by staying in the conda environment, and update the R modules: ::

update.packages(ask = FALSE, checkBuilt = TRUE)

0 comments on commit c71b453

Please sign in to comment.