GSoC blog 4

Till July 23

I am glad to say that I have now gotten access to the server managed by the organisation. Some time was spent on setting up my user account, with the appropriate permissions etc.

While setting up the account one issue faced by the mentor and me was the confusion between the adduser and the useradd commands. This was solved promptly with some googling.

Now since the server was going to be used for database storage and data collection, I had to manage the space on the server effectively. On running a few commands, I found that an external drive of 1 TB was mounted onto the server to store the info. This had to be the location of the file storage.

Since the server was recently shifted, programs had yet to be reinstalled on the computer. PostgreSQL server, PostGIS extension had to be installed, and its configuration was changed to store it into the large drive that was mounted at /pgsql.

Till July 26

The server was almost ready and there were some final glitches that I took care of today.

Asking the mentor, I cloned my repository into the server as it would have been easier to make, commit, test and push changes. These changes would then be consolidated at the end of the GSoC period and then pushed together in the PR, since by then they would already have been tested.

There were slight issues in the cloning, regarding the permissions for the retrieverdash user. they were also solved promptly.

Till July 30

The server being ready, I ran the first test to check the the dataset, it ran everything except the current dataset

List of schemas
                      Name                      |     Owner     
------------------------------------------------+---------------
 activity_range_vegetation_improvement          | retrieverdash
 activity_silviculture_timber_stand_improvement | retrieverdash
 activity_timberharvest                         | retrieverdash
 bioclim                                        | retrieverdash
 ecoregions_us                                  | retrieverdash
 fire_occurrence_firestat_yearly                | retrieverdash
 harvard_forest                                 | retrieverdash
 mtbs_burn_area_boundary                        | retrieverdash
 mtbs_fire_occurrence                           | retrieverdash
 national_usfs_finalfire_perimeter              | retrieverdash
 public                                         | postgres
 sample_hdf                                     | retrieverdash
 topology                                       | retrieverdash
 usa_activity_silvreforestation                 | retrieverdash
 worldclim_five                                 | retrieverdash
 worldclim_ten                                  | retrieverdash
 worldclim_thirty                               | retrieverdash
 worldclim_twofive                              | retrieverdash

The above datasets were tested and among these many gave issues:

Upon realising that the error was from the wrongful deletion of a folder, I added a few new lines to switch directories before deleting a particular folder.

This reduced the number of repositories giving errors.

ecoregions-us 
mtbs_burn_area_boundary 
mtbs-fire-occurrence
worldclim_five
worldclim_ten
worldclim_thirty
worldclim_twofive

The above were the spatial datasets that still gave errors .

Till Aug 2

By now I had individually checked the following 3 dataset to check if they were installing properly.

ecoregions-us 
mtbs_burn_area_boundary 
mtbs-fire-occurrence

The worldclim ones still seem to be giving errors, that is probably because the data has changed considerably.

I am also reading up on port forwarding so I can have access to the ports of the server from my local machine. Will research more on how to get it working properly. Only the worldclim ones seem to be giving errors at this point.

Until next time!