Server Properties
Server properties are displayed on the Settings > Server Properties page.
Properties are grouped into multiple section. Each property has a tooltip which describes the setting.
Default Values
The Default Value column shows the original value for a given property.
Whenever a value deviates from the default, the database displays the field highlighted in blue.
To reset a property to the default value, leave the value blank and click Save.
Changing Properties
Modifications performed on the Server Properties page apply instantly after clicking Save.
Disabled properties highlighted in grey or blue and grey cannot be changed in the user interface and must be modified in the /opt/atsd/atsd/conf/server.properties
file. These modifications require a database restart.
The server.properties
file can also be changed on the Settings > Configuration Files page by users with administrative permissions, however a database restart is still required.
Configuring Scheduled Tasks
To disable a task in the Scheduled Tasks table, enter No or use the drop-down list to select Never for the selected task.
To disable the entity.group.update
or the search.indexing.incremental
task, set the count to zero (0
).
Exporting Properties
Export the current properties, as well as operating system settings and metrics, in JSON format by clicking Export Configuration at the bottom of the page.
Property Reference
Bold property names can only be modified in the server.properties
file by a qualified user and require a database restart.
HBase
Property | Default Value | Description |
---|---|---|
entity.countToUseFilter | 2 | Applies to Data API queries with entity wildcard (*). If entity count is greater than entity.countToUseFilter , one scan with skips is used. Otherwise a separate scan is executed for each query. |
hbase.compaction.list | d properties message li tag forecast | List of HBase table names for which the major compaction is performed by the task triggered with hbase.compaction.schedule . The list contains ATSD data tables. |
hbase.compression.type | none | HBase-level compression. Supported options: none, gz , lzo , snappy , lz4 , and zstd . |
hbase.compression.type.raw | none | HBase-level compression applied to raw time series data. Supported options are: none, gz , lzo , snappy , lz4 , and zstd . |
hbase.table.prefix | atsd_ | Prefix added to all HBase tables created by ATSD. |
htable.executor.corePoolSize | 16 | Minimum number of worker threads performing HBase queries such as scans. This pool does not include series and properties write threads. |
htable.executor.maxPoolSize | 32 | Maximum number of worker threads performing HBase queries such as scans. This pool does not include series and properties write threads. |
last.insert.cache.max.size | 100000 | Maximum number of series records kept in the Last Insert cache. If the limit is reached by the cache, records older than ten minutes, based on received time, are removed. If the cache remains more than 75% full, messages older than one minute are removed. |
last.insert.write.period.seconds | 15 | Interval at which updated series from the Last Insert cache are persisted to the database. |
messages.timeToLive | 31536000 | Number of seconds after which records are deleted from the atsd_messages table. The retention is based on the insertion time, not the record timestamp. The default is 365 days (365*24*3600 ). |
properties.batch.size | 512 | Number of property records sent to HBase in each PUT command. |
properties.queue.limit | 8192 | Maximum number of property records waiting in memory to be inserted into HBase. |
properties.queue.pool.size | 4 | Maximum number of work threads sending property records into HBase. |
properties.queue.rejection.policy | BLOCK | Policy that controls the behavior when the number of property records in memory reaches the properties.queue.limit . BLOCK policy slows down clients. DISCARD policy deletes most recent records in case of overflow. |
scan.caching.size | 2048 | Number of rows sent to ATSD HBase in each RPC request. Setting the property to a low value causes unnecessary round-trips to the region server, while setting it to a large value requires more memory on the client and causes unnecessary rows to be sent to the client. |
series.batch.size | 1024 | Number of series samples sent to HBase in each PUT command. |
series.queue.limit | 32768 | Maximum number of series samples waiting in memory to be inserted into HBase. |
series.queue.pool.size | 4 | Maximum number of worker threads sending series samples into HBase. |
series.queue.rejection.policy | BLOCK | Policy that controls the behavior when the number of series samples in memory reaches the series.queue.limit . BLOCK policy slows down clients, DISCARD policy deletes most recent samples in case of overflow. |
Scheduled Tasks
Property | Default Value | Description |
---|---|---|
data.compaction.schedule | 0 0 22 * * * | Schedule for launching a data compaction task. The task compresses raw values into a smaller number of daily rows. |
delete.schedule | 0 0 21 * * * | Schedule for running a task to remove deleted entities, metrics, and properties. The task masks deleted data by placing Delete markers that hide the data from reading scans. |
entity.group.update.interval | 1 minute | Interval at which entity groups with expression-based memberships are re-evaluated to add new entities and remove non-matching members from the group. Specify the interval in seconds, minutes, or hours. |
expired.data.removal.schedule | 0 0 1 * * * | Schedule for running a task to automatically delete series samples for metrics with a configured retention interval. This task masks deleted data by placing Delete markers that hide the data from reading scans. |
expired.series.removal.schedule | 0 0 2 * * * | Schedule for running a task to automatically delete expired time series for metrics with a configured series retention interval. This task masks deleted data by placing Delete markers that hide the data from reading scans. |
hbase.compaction.schedule | 0 0 1 * * * | Schedule for triggering a major compaction in HBase. A major compactions removes Delete markers and that data is hidden. The compaction also re-writes storage files for optimal storage. |
internal.backup.schedule | 0 30 23 * * * | Schedule for backing up ATSD configuration records to XML files in ./atsd/backup directory. |
search.indexing.full.schedule | 0 0 3 * * * | Schedule at which the Series Search index is rebuilt. |
search.indexing.incremental.interval | 5 minute | Interval at which the Series Search index is refreshed. Specify the interval in seconds, minutes, or hours. |
Monitoring
Property | Default Value | Description |
---|---|---|
internal.metrics.dump.path | ../logs/metrics.txt | Absolute or relative path to a file containing current database monitoring parameters. Relative path must be relative to base ATSD directory. The file is refreshed every 15 seconds. |
jmx.access.file | None | Absolute path to JMX access file containing the list of users and their read and write permissions. |
jmx.enabled | No | JMX service status. |
jmx.host | localhost | Hostname returned by the JMX service for Data connections. |
jmx.password.file | None | Absolute path to JMX password file. |
jmx.port | 1099 | Port on which JMX service is listening. |
Network
Property | Default Value | Description |
---|---|---|
http.port | 8088 | Port for plaintext HTTP protocol. |
https.keyManagerPassword | None | Java SSL Key Manager password. |
https.keyStorePassword | None | Java SSL Key Store password. |
https.port | 8443 | Port for secure HTTP protocol. |
https.trustStorePassword | None | Java SSL Trust Store password. |
input.disconnect.on.error | Yes | Stop the TCP connection if a malformed command is received. Applies to Network API series, property, and message commands. |
input.port | 8081 | TCP port for sending Network API commands. |
input.socket.keep-alive | Yes | Enable keepalive for TCP sockets open on input.port |
input.socket.receive-buffer-size | -1 | Network memory buffer for processing commands received on the input.port . |
pickle.port | 8084 | TCP for sending commands serialized in Python pickle format. |
series.processing.pool.size | 2 | Number of thread processing series commands received over TCP. |
udp.input.port | 8082 | UDP port for sending Network API commands. |
SQL
Property | Default Value | Description |
---|---|---|
sql.metric.like.limit | 50 | Maximum number of metrics matched by metric LIKE expression in atsd_series tables. |
sql.tmp.storage.max_rows_in_memory | 51200 | Maximum number of rows to be processed in memory by grouping and ordering queries. The limit is shared by all concurrently executing queries. If a query selects more rows than remain in shared memory, the query is processed by ATSD using a temporary table which results in increased response times. |
Cache
Cache usage statistics are available as cache.size
and cache.used_percent
metrics, and on the Settings > Cache Management page.
Property | Default Value | Description |
---|---|---|
cache.entity.maximum.size | 100000 | Maximum number of entity records to keep in server memory cache. |
cache.metric.maximum.value | 50000 | Maximum number of metric records to keep in server memory cache. |
cache.tagKey.maximum.size | 50000 | Maximum number of tag names to keep in server memory cache. |
cache.tagValue.maximum.size | 100000 | Maximum number of tag values to keep in server memory cache. |
Web Interface
Property | Default Value | Description |
---|---|---|
entity-group.display.tags | None | List of entity group tags displayed on the Entity Group list. |
scollector.ignore.tags | Environment Role | List of tags ignored from series commands received from scollector agents. |
Rule Engine
Property | Default Value | Description |
---|---|---|
cancel.on.rule.change | No | Trigger actions when a rule is saved or deleted. |
system.commands.enabled | Yes | Enable script execution by Rule Engine. |
system.commands.timeout.seconds | 15 | Interval in seconds after which the script execution ceases. |
webdriver.chromebrowser.path | None | Path to the google-chrome binary. |
webdriver.chromedriver.path | None | Path to the chromedriver binary. |
webdriver.phantomjs.path | None | Path to phantomjs binary. |
webdriver.pool.size | 4 | Number of worker threads taking screenshots using Web Driver. |
webdriver.screenshots.enable | Yes | Enable screenshot features in Rule Engine. |
Other
Property | Default Value | Description |
---|---|---|
api.guest.access.enabled | No | Enable anonymous access to Data API and Meta API query methods. |
backup.data.directory | ../backup | Path to a directory where ATSD stores backup files. |
default.addInsertTime | Yes | Include entity/metric last insert time in the response by default. |
delete.sleep.interval.ms | 2000 | Millisecond interval to wait between 10-second deletion tasks. |
delete.total.duration.minutes | 60 | Maximum amount of time in minutes spent executing pending deletion tasks. |
hostname | nurswgvml007 | Hostname of the ATSD server determined automatically by executing the hostname command. |
nmon.data.directory | /tmp/atsd/nmon | Absolute path to a directory where nmon agents are stored. |
search.index.path | /tmp/atsd/lucene | Absolute path to Lucene index directory. |
security.letsencrypt.renew | No | Enable to check Let's Encrypt certificate once a day and attempt to renew it. |
server.url | https://{atsd_hostname}:8443 | Server URL specified in email notification links, by default set to hostname . |