add utils/collect_os_info.sh script

diff --git a/utils/collect_os_info.sh b/utils/collect_os_info.sh
new file mode 100755 (executable)
index 0000000..15ea6c3
--- /dev/null
+++ b/utils/collect_os_info.sh
@@ -0,0 +1,94 @@
+#!/usr/bin/env bash
+
+Description="The purpose of this script is to collect system information for bug reports.\n
+Submit issues to https://github.com/osm-search/Nominatim/issues"
+
+
+####### Gather the Information ##################################################
+# Separate the information gathering from the report generation.  Dividing these
+# makes it easier to make trivial changes by not have to learn the other portion
+# of this script.
+
+# Nominatium version
+# NOTE: Getting this version will NOT work if it is being ran from in another
+# folder than Nominatim/utils.  It call python3 to import version.py locally and
+# prints it in the version format. 
+NominatimVersion=`cd ../nominatim/ && python3 -c "import version; print('{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(version.NOMINATIM_VERSION))"`
+
+# PostgreSQL version
+PostgreSQLVersion=`postgres --version`
+if [ "$?" -ne "0" ]
+then
+  PostgreSQLVersion="Not installed"
+fi
+
+# - PostGIS version:
+# The command for this should look something like this:
+#       psql -U nominatim -d mydatabase -c 'SELECT PostGIS_full_version();'
+# ASSUME the username is nominatim
+# This needs to be ran under the account with the appropriate permissions.
+# This has been left blank.
+PostGISVersion=
+
+# There are different ways to getting the Linux OS information.
+# https://www.cyberciti.biz/faq/how-to-check-os-version-in-linux-command-line/
+# /etc/os-release has a number of representations of the OS
+# PRETTY_NAME is pity.
+OperatingSystem=`grep '^PRETTY_NAME' /etc/os-release | cut -d'=' -f2`
+
+RAM=`grep ^MemTotal /proc/meminfo | cut -d':' -f2`
+
+# In /proc/cupinfo: siblings seems to refer to total cores like hyperthreaded cores.
+# The hyperthreaded cores could be included if that is needed.
+NumCPUs=`grep '^cpu cores' /proc/cpuinfo | head -1 | cut -d':' -f2`
+
+
+# - type and size of disks:
+# could use `sudo fdisk -l` or `mount` to print this, but snaps have made this
+# worse than useless with loop devices on Ubuntu.  
+# `df -h` - show the free space on drives
+# `lsblk` - this tell you what the server has not necessarily this machine.  So in a container environment
+#  (like docker) this wouldn't be the correct report.
+# This guide shows ways to get various storage device information: https://www.cyberciti.biz/faq/find-hard-disk-hardware-specs-on-linux/
+
+# - bare metal/AWS/other cloud service:
+# Unsure of how to detect this, but it might be useful for reporting disk storage.
+# One options would be to prompt the user something like this:
+# Enter system configuration (1) bare metal (2) AWS (3) Other Cloud (4) Docker (5) Other: _
+
+# ------ What do these commands do? -------------------------------------------
+# "cut -d':' -f2"      command take the line and splits it at the semicolon(:)
+#                      and returns the portion in the second (2nd) "field"
+#
+# "head -1"            returns the first line that matches
+#
+
+####### Print the Markdown Report ######################################################
+# 1>&2 redirects echo to print to stderr instead of stdout
+
+echo 1>&2
+echo -e $Description 1>&2
+echo Copy and paste or redirect the output of the file:  1>&2
+echo "     \$ ./collect_os_info.sh > report.md" 1>&2
+echo 1>&2
+
+
+echo "**Software Environment (please complete the following information):**"
+echo - Nominatim version: $NominatimVersion
+echo - PostgreSQL version: $PostgreSQLVersion 
+echo - PostGIS version: $PostGISVersion 
+echo - OS: $OperatingSystem
+echo
+echo
+
+
+echo "**Hardware Configuration (please correct the following information):**"
+echo - RAM: $RAM
+echo - number of CPUs: $NumCPUs
+echo - type and size of disks:
+echo - bare metal/AWS/other cloud service: 
+echo
+echo
+echo **Postgresql Configuration:**
+echo
+