Configuration Parameters

As soon as the DBMS is installed, the system parameters can be modified by the DB administrator. There are two configuration files:
  • /opt/eventos/ontoquad/ontoquad.conf and
  • /opt/eventos/ontoquad/ontoquad-logging.conf.
    It is not recommended to change parameter values in the configuration file /opt/eventos/ontoquad/ontoquad-logging.conf. A full list and description of ontoquad.conf configuration parameters are given in Table 1.

Table 1. List and specification of ontoquad.conf configuration parameters

Configuration parameter Parameters values
index-type Value type of database index structure:
index-type = polymorphic2
index-type = polymorphic6
index-type = polymorphic6monolith
index-type = polymorphic24
polymorphic2 is used for triplet (SPO) index configuration and polymorphic6, polymorphic6monolith, polymorphic24 are used for quadruplet (SPOG) index configuration
By default: index-type = polymorphic2
transactions Database transactional mode (values on/off or true/false).
By default:
transactions = on
or
transactions = true
threads Number of threads for parallel query processing.
By default: threads = 2
vocabulary-upload-threads Number of threads for parallel vocabulary uploading
Allows to increase the performance of data loading for batch loading
Possible value is in the diapason of from 1 to 3
cachesize Maximum size in bytes of Uncompressed Database pages Cache:
cachesize = 1073741824
compressed-page-cachesize Maximum size in bytes of Compressed Database pages Cache:
compressed-page-cachesize = 13958643712
-1 (default) means no page compression
dataset-cachesize Cache for temporary tables
Temporary tables are used, for example, for storing the final result of the request, calculation of distinct, datasets sorting, ordering etc.
max-inmemory-rows Dynamic memory buffer which serves for database sorting operations. It defines Max number of dataset rows that can be acoomodated by the dynamic memory. For example:
max-inmemory-rows = 1000000
literal-size-threshold The maximum length of a text literal within which the identity of values is checked
By default:
literal-size-threshold = 64
sparql-endpoint-port Port used for sparql endpoint.
By default: sparql-endpoint-port = 8080
sparql-endpoint-address Address used for sparql endpoint, for example:
sparql-endpoint-address = 172.16.4.2
By default: sparql-endpoint-address = 127.0.0.1
sparul Allows (true/false) SparUL queries mode.
By default: sparul = false
admin-interface Allow /admin and /upload sites. If "true", then the bulk database download by HTTP request can be performed
By default: admin-interface = true
inactivity_timeout Time interval allowed for session inactivity (in millisecs), for example:
inactivity_timeout = 600000
indexmaxfilesize Maximal size in bytes of an index file fragment, allows splitting the index file to smaller fragments. Size 0 means unlimited file size (default value). For example:
indexmaxfilesize = 4294967296
[btree]
star
Index file type, set it up before launching database for the first time. Value
star = on
setups B*-tree type. Value
star = off
setups B-tree type.
[api]
port
Remote API server’s listening port, the absence or "-1" means the listening is turned off. For example:
port = 2777
[api]
threads
The number of threads that handle Remote API calls. For example:
threads = 10
[api]
tcpconnections
Max number of accepted API connections at a time. For example:
tcpconnections = 100
[http]
ssl
true/false, shows if Secure Socket Layer (SSL) coding is applied
Default value is false
[http]
authentication
true/false, shows if authentication is applied
Default value is false
[ngram]
enable
true/false, Allows (true/false) n-gram indexing mode.
By default: enable = false
[ngram]
size
Number of characters in single n-gram
[ngram]
max-literal-size
Max size of a literal in bytes to be splitted into n-grams and indexed.
Default value is 32
[ngram]
oversize-handler
Oversized literal handler type. Literal may be ignored or cut first max-literal-size bytes to index or the update operation may be aborted (ignore, split, abort). Default value is split
[ngram]
predicates
Path to a new line separated predicate list file. Only object literals related to these predicates are n-gram indexed.
[ngram]
predicate-suffix
Predicate suffix for n-gram predicates.
Default value is -ngram

A full list and description of the configuration parameters of ontoquad-logging.conf are given in the Table 2.

Table 2. List and specification of ontoquad-logging.conf configuration parameters

Configuration parameter Parameters values
[Core]
DisableLogging
Logging mode switcher (true/false), by default:
DisableLogging = false
[Core]
Filter
Record severity-based global filter, allowed values: trace, debug, info, warning, error, fatal. By default:
Filter = "%Severity% >= info"
[Sink:CompleteSink] Sink settings sections. Text-file sink for all records, regardless severity
[Sink:CompleteSink]
Destination
Sink destination type (Console/TextFile)
Destination = Console
[Sink:CompleteSink]
FileName
The file name pattern for the sink backend. This parameter is mandatory. Destination should be set in TextFile
FileName = ontoquad_complete_%5N.log
[Sink:CompleteSink]
Filter
Sink-specific filter. Optional, by default no filter is applied
Filter = "%MethodName% contains "DocumentIndex""
[Sink:CompleteSink]
Format
Formatter string. Optional, by default only log record message text is written. Supported attributes: LineID, TimeStamp, ProcessID, ThreadID, Severity, RemoteAddress, MethodName, ApplicationName, ApplicationType
Format = "%LineID% Severity [%TimeStamp%][%Name%][%Type%][%RemoteAddress%] [%MethodName%] - _"
[Sink:CompleteSink]
Asynchronous
The flag shows whether the sink should be asynchronous (true/false)
Asynchronous = false
[Sink:CompleteSink]
AutoFlush
Enables automatic stream flush after each log record. Asynchronous should be set in true.
AutoFlush = true
[Sink:CompleteSink]
Target
Target directory name, in which the rotated files will be stored. If this parameter is specified, the collection of rotated files is enabled. Otherwise the feature is not enabled, and all corresponding parameters are ignored
Target = logs
[Sink:CompleteSink]
MaxSize
Total size of files in the Target directory, in bytes; when reached, the oldest file will be deleted.
If not specified, no size-based file cleanup will be performed
MaxSize = 10485760
[Sink:CompleteSink]
MinFreeSpace
Minimum free space in the Target directory, in bytes; when reached, the oldest file will be deleted.
If not specified, no space-based file cleanup will be performed.
MinFreeSpace = 1048576
[Sink:CompleteSink]
RotationSize
File size, in bytes, upon which the file rotation will be performed. If not specified, no size-based rotation will be made
RotationSize = 10485760
[Sink:CompleteSink]
RotationInterval
Time interval, in seconds, upon which the file rotation will be performed. If not specified, no time-based rotation will be done.
RotationInterval = 604800
[Sink:CompleteSink]
ScanForFiles
Mode of scanning for old files in the target directory. If specified in string form, the following values are permitted: All/Matching. If not specified, no scanning will be performed.
ScanForFiles = Matching
[Sink:ImportantSink] Text-file sink for all important records, i.e. severity equals to warning, error or fatal
[Sink:ImportantSink]
Destination
Sink destination type (Console/TextFile)
Destination = TextFile
[Sink:ImportantSink]
FileName
The file name pattern for the sink backend. This parameter is mandatory. Destination should be set in TextFile
FileName = ontoquad_important_%5N.log
[Sink:ImportantSink]
Filter
Record severity-based global filter, allowed values: trace, debug, info, warning, error, fatal. By default:
Filter = "%Severity% >= warning"
[Sink:ImportantSink]
Format
Formatter string. Optional, by default only log record message text is written. Supported attributes: LineID, TimeStamp, ProcessID, ThreadID, Severity, RemoteAddress, MethodName, ApplicationName, ApplicationType
Format = "%LineID% Severity [%TimeStamp%][%Name%][%Type%][%RemoteAddress%] [%MethodName%] - _"