settings.conf

# This file enumerates the Basic and Advanced config settings.
# The Basic settings are intended for users with basic knowledge about AnzoGraph
# or databases in general. The advanced settings are intended for system administrators
# or users with an advanced level of knowledge about AnzoGraph or databases.
# 
# After changing this file, restart AnzoGraph to apply the configuration changes.
#

####################################################
# Basic Settings                                   #
####################################################

# Turn support for increment incremental persistence (V3) (load)
# enable_persistence=true

# Enable the SPARQL HTTP protocol (overall)
 enable_sparql_protocol=true

# Enable the secure SPARQL HTTPS protocol (overall)
 enable_ssl_protocol=true

# Location of the database's operational cache files (overall)
# internal_directory=

# The amount of memory (in MB) available for AnzoGraph (default is system-based) (memory)
# max_memory=0

# The default output format for AnzoGraph responses - xml, json or csv (overall)
# output_format=xml

# Location AnzoGraph saves loads/inserts/deletes across restarts (overall)
# persistence_directory=

# The port to open for HTTP (SPARQL end-point) clients (overall)
# sparql_protocol_port=7070

# If 'true': no FROM clause means ALL graphs, else: just Un-named (DEFAULT) graph (overall)
# sparql_spec_default_graph=false

# Location of the disk based query temporary files (usually SSD) (overall)
# spill_directory=

# The port to open AZG on for HTTPS (SPARQL end-point) clients (overall)
# ssl_protocol_port=8256

# Indicates how verbose the startup message is - 0-quiet, 1-ready, 2-ports, 3-more (overall)
# startup_info=1

# Amount of time to wait for queries to complete during a -stop request (overall)
# stop_timeout=30

# Truncate large strings to the max string size (1MB) (load,overall)
 truncate_clob=true

# Use custom SSL files containing fully qualified domain names (overall)
# use_custom_ssl_files=false

# Sets the limit on the number of query pipelines that can run concurrently (overall)
# user_queues=40

####################################################
# Advanced Settings                                #
####################################################

# Port to open AZG on for gRPC clients (overall)
# anzo_protocol_port=5700

# Location of the database's operational cache files (overall)
# auto_restart_directory=

# This will automatically restart the db. For now, this is irrespective of whether persistence is set to true/false (overall)
# auto_restart_max_attempts=5

# If auto_restart > 0, this var limits auto-restart attempts within the given time in seconds (overall)
# auto_restart_time=600

# Sets the logging verbosity of AWS operations. (debug,load)
# aws_log_level=2

# Regions to search for AWS S3 buckets (for LOAD command) (load)
# aws_search_regions=

# Enable client authentication (overall)
# azgmgrd_client_auth=false

# System manager password (overall)
# azgmgrd_password=azgmgrd

# Log base 2 of maximum unique predicates supported (load,memory,overall)
# bits_per_pred_index=16

# Log base 2 of maximun unique subjects supported (load,memory,overall)
# bits_per_uri_index=32

# Default name basis for blank nodes (overall)
# blank_node_name=genid

# Whether to check an internet server to see if this matches current download version (load,overall)
# call_home_for_updates=false

# The internal fabric communications port (clustering)
# comm_port_base=9100

# Max number of concurrent compilations (performance)
# compile_concurrent=8

# The maximum amount of memory (in MB) to use for compiling generated code before switching to non-optimized compile (memory,performance)
# compile_max_memory=500

# The maximum number of seconds to spend compiling generated code before switching to non-optimized compile (performance)
# compile_max_seconds=30

# Whether to use optimized compile (performance)
# compile_optimized=background

# Nominal file sizes to create, in megabytes (load)
# copy_file_size=5

# Graphs to control periodic query execution thread (overall)
# cron_graphs=

# How often to check to see if any cron graphs changed (overall)
# cron_graphs_recheck=10

# In csvdelim mode, the field separator (load)
# csv_delim=,,,,

# In csvdelim mode, the row ender (rather than newline) (load)
# csv_ender=!!!!

# Commonly referenced variable in load commands, for files already on the AnzoGraph server (load,overall)
# dflt_load_dir=etcdir

# Percentage of query working memory per node to allocate to each segment (memory)
# disk_mem_pct=75

# Whether to allow roles and access control (overall)
# enable_acl=false

# Whether to allow roles and access control through reverse proxy (overall)
# enable_external_auth=false

# Enable fast but memory intensive PageRank computation (performance)
# enable_fast_pagerank=false

# Whether to log historical system statistics in the sth_* tables at all (turning this off will disable certain GUI features) (xray)
# enable_logging=true

# OpenCypher language suppport ()
# enable_ocx=true

# Collect stats based on OWL ontology triples ()
# enable_owlstats=true

# Whether to enable quadstore mode (load,memory,overall)
enable_quadstore_mode=true

# Whether to refresh statistics after update to graph data (debug,performance)
# enable_refresh_stats_on_update=true

# Whether to allow running as root (overall)
enable_root_user=true

# Treat functions with 'always-incorrect' argtypes as unbound, not as errors (planner)
# enable_unbound_functions=true

# Treat missing variables as unbound variables, not as errors (planner)
# enable_unbound_variables=false

# Whether SPARQL/Turtle IRIs must be true URIs (load,memory,overall,performance)
# enforce_uris=false

# Directory on leader node to scan for extension libraries (load)
# extensions_dir=./lib/udx

# Decimal digits after the radix point for returned floats (default is 6) (overall)
# float_decimals=6

# Format specifier for formatting returned floats: one of {f,F,e,E,g,G} (default is g) (overall)
# float_format=g

# Default separator for gconcat (overall)
# gconcat_default_separator= 

# Default HDFS load mode (load)
# hdfs_default_load_mode=global

# How long does AnzoGraph wait for data before it times out (min 5, max 30) (performance)
# http_select_timeout=5

# If set to true, this will allow the execution of all queries, even if they were previously denied-listed (overall)
# ignore_deniedlist_queries=false

# Whether to skip unsupported external table types, or error (load)
# ignore_unsupported_types=true

# Number of seconds to wait between iterations of incremental vacuum (load)
# incremental_vacuum_delay=1

# Max length of a string to keep inline - limit 32767 (overall)
# inline_strings_size=32767

# When true, suppresses some of the optimizer's join ordering algorithms (performance,planner)
# join_using_order_presented=false

# Specifies the maximum size of the heap used by the embedded JVM (memory)
# jvm_max_memory=

# Optional arguments with which to start the embedded JVM (';' delimited) (load)
# jvm_options=

# Location of the license file (overall)
# license_file=license.pem

# Timeout between a socket loading client post the load and actually connecting (load)
# load_client_timeout=10

# Graph to write error info to for LOAD SILENT (load)
# load_errors_graph=load_errors

# Change bad datetimes in loads to this value, e.g. 0001-01-01T00:00:00Z (load)
# load_normalize_datetime=

# Whether to persist load for X-ray playback - choices are nothing, all or pload (load)
# load_persistence=

# Whether create the graph as a property (load)
# load_quads_as_rdfstar=false

# Location of the log files (overall)
# log_directory=

# By default, when azgmgrd log file is rotated, the last 5 azgmgrd log files will be stored ()
# log_rotate_counter=5

# Azgmgrd log file rotation size limit in bytes ()
# log_size_limit=1790000

# The limit on the number of rows produced by a cartesian product (performance)
# max_cartesian_product=1000000000

# The limit on the number of rows in the inner table of a cartesian product (performance)
# max_cartesian_product_inner_size=20

# Maximum string/blob/udt size ()
# max_string_size=100000000

# Whether to check for numeric multiplication overflow (performance)
# numeric_check_overflow=true

# Enable relative sort order for UDT (overall)
# order_udt_as_string=false

# Whether alternate kernel vm configurations be permitted (os)
# os_allow_alternate_vm_config=true

# Whether small mmap values are permitted (os)
# os_allow_small_mmap=true

# Whether Transparent Huge Pages are permitted (os)
# os_allow_thp=true

# Percentage of memory to reserve for paged triplestore cache (memory,performance)
# paged_cache_memory_percent=20

# Disk based data loading and use (memory,performance)
# paged_data=false

# Max number of concurrent parallel loaders (e.g. azgcliload clients) (load)
# parallel_load_concurrency=4

# Base port used for shipping the data in parallel loading, each IP address starts here (load)
# parallel_load_portbase=8100

# Ordered list of preprocesors to rewrite query source text prior to compilation (';' delimited) (overall)
# preprocessors=

# Delay (in seconds) before updating query progress (overall)
# progress_monitor_delay=0.5

# Rate (in seconds) to update clients on query progress or 0.0 for none (overall)
# progress_monitor_interval=0.5

# Set internal thread names for ps and top commands (debug,os)
# ps_top_thread_name=true

# The name to assign to a query or group of queries (overall)
# query_label=

# User id to run the system manager, and thus the database as (overall)
# run_as_user=

# Enable scatter-gather to compress URIs (overall)
# sc_compress=false

# Enable scatter-gather to decompress URIs (overall)
# sc_decompress=false

# Used to block all selects (overall)
# selects_blocked=false

# Limit the number of rows returned by the remote sparql endpoint ()
# service_graph_rowset_limit=1000

# Whether to mint a globally unique IRI for the bnode (load,overall)
# skolemize_bnodes=false

# How to treat skolemized bnodes in ISBLANK and ISIRI/ISURI (load,overall)
# skolemize_funcs=false

# The number of slices per node - usually equal to the number of cores per CPU ()
# slices_per_node=0

# Stateful endpoints config parameters (network)
# stateful_endpoints_config=default

# Streaming scatter-gather uri compression/decompression (overall)
# streaming_sc=false

# By default sysmgr port is 5600. Always need to restart the system manager whenever the value is modified (overall)
# sysmgr_port=5600

# Whether to use IPv6 style addresses (default is IPv4) (clustering)
# tcp_ipv6=false

# Log scanned extension meta-data to './internal/log/udx.log' (debug)
# udx_log_metadata=false

# Validate the URI data on restore from persistence (overall)
# validate_uris=true

# Approximate number of seconds to include in xray dumps (xray)
# xray_seconds=2147483647

# Which tables go to presisted spool file (xray)
# xray_sth_spool=sth_session,sth_query,sth_query_tags,sth_errors,sth_save_reload,sth_cron_events,sth_repl_contract

# How long to accumulate spooled sth (xray)
# xray_sth_spool_duration=2days