Skip to content

Latest commit

 

History

History
176 lines (160 loc) · 32.9 KB

malcolm-config.md

File metadata and controls

176 lines (160 loc) · 32.9 KB
Raw

Malcolm Configuration

Malcolm's runtime settings are stored (with a few exceptions) as environment variables in configuration files ending with a .env suffix in the ./config directory. The ./scripts/configure script can help users configure and tune these settings.

Run ./scripts/configure and answer the questions to configure Malcolm. For an in-depth treatment of these configuration questions, see the Configuration section in End-to-end Malcolm and Hedgehog Linux ISO Installation.

Environment variable files

Although the configuration script automates many of the following configuration and tuning parameters, some environment variables of particular interest are listed here for reference.

  • arkime.env and arkime-secret.env - settings for Arkime
    • ARKIME_AUTO_ANALYZE_PCAP_THREADS – the number of threads available to Arkime for analyzing PCAP files (default 1)
    • ARKIME_PASSWORD_SECRET - the password hash secret for the Arkime viewer cluster (see passwordSecret in Arkime INI Settings) used to secure the connection used when Arkime viewer retrieves a PCAP payload for display in its user interface
    • ARKIME_ROTATE_INDEX - how often (based on network traffic timestamp) to create a new index in OpenSearch
    • ARKIME_QUERY_ALL_INDICES - whether or not Arkime should query all indices instead of trying to calculate which ones pertain to the search time frame (default false)
    • ARKIME_SPI_DATA_MAX_INDICES - the maximum number of indices for querying SPI data, or set to -1 to disable any max. The Arkime documentation warns "OpenSearch/Elasticsearch MAY blow up if we ... search too many indices." (default 7)
    • MANAGE_PCAP_FILES and ARKIME_FREESPACEG - these variables deal with PCAP deletion by Arkime, see Managing disk usage below
    • MAXMIND_GEOIP_DB_LICENSE_KEY - Malcolm uses MaxMind's free GeoLite2 databases for GeoIP lookups. As of December 30, 2019, these databases are no longer available for download via a public URL. Instead, they must be downloaded using a MaxMind license key (available without charge from MaxMind). The license key can be specified here for GeoIP database downloads during build- and run-time.
    • The following variables configure Arkime's use of OpenSearch Index State Management (ISM) or Elasticsearch Index Lifecycle Management (ILM):
      • INDEX_MANAGEMENT_ENABLED - if set to true, Malcolm's instance of Arkime will use these features when indexing data
      • INDEX_MANAGEMENT_OPTIMIZATION_PERIOD - the period in hours or days that Arkime will keep records in the hot state (default 30d)
      • INDEX_MANAGEMENT_RETENTION_TIME - the period in hours or days that Arkime will keep records before deleting them (default 90d)
      • INDEX_MANAGEMENT_OLDER_SESSION_REPLICAS - the number of replicas for older sessions indices (default 0)
      • INDEX_MANAGEMENT_HISTORY_RETENTION_WEEKS - the retention time period (weeks) for Arkime history data (default 13)
      • INDEX_MANAGEMENT_SEGMENTS - the number of segments Arlime will use to optimize sessions (default 1)
      • INDEX_MANAGEMENT_HOT_WARM_ENABLED - whether or not Arkime should use a hot/warm design (storing non-session data in a warm index); setting up hot/warm index policies also requires configuration on the local nodes in accordance with the Arkime documentation
  • arkime-live.env - settings for live traffic capture with Arkime
    • See Tuning Arkime for variables related to managing Arkime's performance and resource utilization during live capture.
  • auth-common.env - authentication-related settings
  • auth.env - stores the Malcolm administrator's username and password hash for its nginx reverse proxy
  • beats-common.env - settings for interactions between Logstash and Filebeat
    • BEATS_SSL – if set to true, Logstash will use require encrypted communications for any external Beats-based forwarders from which it will accept logs (default true)
    • LOGSTASH_HOST – the host and port at which Beats-based forwarders will connect to Logstash (default logstash:5044); see MALCOLM_PROFILE below
  • dashboards.env and dashboards-helper.env - settings for the containers that configure and maintain OpenSearch and OpenSearch Dashboards
    • DASHBOARDS_URL - used primarily when OPENSEARCH_PRIMARY is set to elasticsearch-remote (see OpenSearch and Elasticsearch instances), this variable stores the URL for the Kibana instance into which Malcolm's dashboard's and index templates will be imported
    • DASHBOARDS_PREFIX – a string to prepend to the titles of Malcolm's prebuilt dashboards prior upon import during Malcolm's initialization (default is an empty string)
    • DASHBOARDS_DARKMODE – if set to true, OpenSearch Dashboards will be set to dark mode upon initialization (default true)
    • OPENSEARCH_INDEX_SIZE_PRUNE_LIMIT - the maximum cumulative size of OpenSearch indices are allowed to consume before the oldest indices are deleted, see Managing disk usage below
  • filebeat.env - settings specific to Filebeat, particularly for how Filebeat watches for new log files to parse and how it receives and stores third-Party logs
    • LOG_CLEANUP_MINUTES and ZIP_CLEANUP_MINUTES - these variables deal cleaning up already-processed log files, see Managing disk usage below
  • logstash.env - settings specific to Logstash
    • LOGSTASH_OUI_LOOKUP – if set to true, Logstash will map MAC addresses to vendors for all source and destination MAC addresses when analyzing Zeek logs (default true)
    • LOGSTASH_REVERSE_DNS – if set to true, Logstash will perform a reverse DNS lookup for all external source and destination IP address values when analyzing Zeek logs (default false)
    • LOGSTASH_SEVERITY_SCORING - if set to true, Logstash will perform severity scoring when analyzing Zeek logs (default true)
    • LS_JAVA_OPTS - part of LogStash's JVM settings, the -Xmx and -Xms values set the size of LogStash's Java heap (we recommend somewhere between 1500m and 4g)
  • lookup-common.env - settings for enrichment lookups, including those used for customizing event severity scoring
    • CONNECTION_SECONDS_SEVERITY_THRESHOLD - when severity scoring is enabled, this variable indicates the duration threshold (in seconds) for assigning severity to long connections (default 3600)
    • FREQ_LOOKUP - if set to true, domain names (from DNS queries and SSL server names) will be assigned entropy scores as calculated by freq (default false)
    • FREQ_SEVERITY_THRESHOLD - when severity scoring is enabled, this variable indicates the entropy threshold for assigning severity to events with entropy scores calculated by freq; a lower value will only assign severity scores to fewer domain names with higher entropy (e.g., 2.0 for NQZHTFHRMYMTVBQJE.COM), while a higher value will assign severity scores to more domain names with lower entropy (e.g., 7.5 for naturallanguagedomain.example.org) (default 2.0)
    • SENSITIVE_COUNTRY_CODES - when severity scoring is enabled, this variable defines a comma-separated list of sensitive countries (using ISO 3166-1 alpha-2 codes) (default 'AM,AZ,BY,CN,CU,DZ,GE,HK,IL,IN,IQ,IR,KG,KP,KZ,LY,MD,MO, PK,RU,SD,SS,SY,TJ,TM,TW,UA,UZ', taken from the U.S. Department of Energy Sensitive Country List)
    • TOTAL_MEGABYTES_SEVERITY_THRESHOLD - when severity scoring is enabled, this variable indicates the size threshold (in megabytes) for assigning severity to large connections or file transfers (default 1000)
  • netbox-common.env, netbox.env, netbox-secret.env, netbox-postgres.env, netbox-redis-cache.env and netbox-redis.env - settings related to NetBox and Asset Interaction Analysis
  • nginx.env - settings specific to Malcolm's nginx reverse proxy
    • NGINX_LOG_ACCESS_AND_ERRORS - if set to true, all access to Malcolm via its web interfaces will be logged to OpenSearch (default false)
    • NGINX_SSL - if set to true, require HTTPS connections to Malcolm's nginx-proxy container (default); if set to false, use unencrypted HTTP connections (using unsecured HTTP connections is NOT recommended unless you are running Malcolm behind another reverse proxy such as Traefik, Caddy, etc.)
  • opensearch.env - settings specific to OpenSearch
    • OPENSEARCH_JAVA_OPTS - one of OpenSearch's most important settings, the -Xmx and -Xms values set the size of OpenSearch's Java heap (we recommend setting this value to half of system RAM, up to 32 gigabytes)
    • OPENSEARCH_PRIMARY - one of opensearch-local, opensearch-remote, or elasticsearch-remote, to determine the OpenSearch or Elasticsearch instance Malcolm will use (default opensearch-local)
    • OPENSEARCH_URL - when using Malcolm's internal OpenSearch instance (i.e., OPENSEARCH_PRIMARY is opensearch-local) this should be http://opensearch:9200, otherwise this value specifies the primary remote instance URL in the format protocol://host:port (default http://opensearch:9200)
    • OPENSEARCH_SSL_CERTIFICATE_VERIFICATION - if set to true, connections to the primary remote OpenSearch instance will require full TLS certificate validation (this may fail if using self-signed certificates) (default false)
    • OPENSEARCH_SECONDARY - one of opensearch-local, opensearch-remote, elasticsearch-remote, or blank (unset) to indicate that Malcolm should forward logs to a secondary remote OpenSearch instance in addition to the primary OpenSearch instance (default is unset)
    • OPENSEARCH_SECONDARY_URL - when forwarding to a secondary remote OpenSearch instance (i.e., OPENSEARCH_SECONDARY is set) this value specifies the secondary remote instance URL in the format protocol://host:port
    • OPENSEARCH_SECONDARY_SSL_CERTIFICATE_VERIFICATION - if set to true, connections to the secondary remote OpenSearch instance will require full TLS certificate validation (this may fail if using self-signed certificates) (default false)
    • The following variables control the OpenSearch indices to which network traffic metadata are written. Changing them from their defaults may cause logs from non-Arkime data sources (i.e., Zeek, Suricata) to not show up correctly in Arkime.
      • MALCOLM_NETWORK_INDEX_PATTERN - Index pattern for network traffic logs written via Logstash (default is arkime_sessions3-*)
      • MALCOLM_NETWORK_INDEX_TIME_FIELD - Default time field to use for network traffic logs in Logstash and Dashboards (default is firstPacket)
      • MALCOLM_NETWORK_INDEX_SUFFIX - Suffix used to create index to which network traffic logs are written (supports Ruby strftime strings in ％{}) (e.g., hourly: ％{％y％m％dh％H}, twice daily: ％{％P％y％m％d}, daily (default): ％{％y％m％d}, weekly: ％{％yw％U}, monthly: ％{％ym％m})
    • The following variables control the OpenSearch indices to which other logs (third-party logs, resource utilization reports from network sensors, etc.) are written.
      • MALCOLM_OTHER_INDEX_PATTERN - Index pattern for other logs written via Logstash (default is malcolm_beats_*)
      • MALCOLM_OTHER_INDEX_TIME_FIELD - Default time field to use for other logs in Logstash and Dashboards (default is @timestamp)
      • MALCOLM_OTHER_INDEX_SUFFIX - Suffix used to create index to which other logs are written (supports Ruby strftime strings in ％{}) (default is ％{％y％m％d})
  • pcap-capture.env - settings specific to capturing traffic for live traffic analysis
    • PCAP_ENABLE_NETSNIFF – if set to true, Malcolm will capture network traffic on the local network interface(s) indicated in PCAP_IFACE using netsniff-ng
    • PCAP_ENABLE_TCPDUMP – if set to true, Malcolm will capture network traffic on the local network interface(s) indicated in PCAP_IFACE using tcpdump; there is no reason to enable both PCAP_ENABLE_NETSNIFF and PCAP_ENABLE_TCPDUMP
    • PCAP_FILTER – specifies a tcpdump-style filter expression for local packet capture; leave blank to capture all traffic
    • PCAP_IFACE – used to specify the network interface(s) for local packet capture if PCAP_ENABLE_NETSNIFF, PCAP_ENABLE_TCPDUMP, ZEEK_LIVE_CAPTURE or SURICATA_LIVE_CAPTURE are enabled; for multiple interfaces, separate the interface names with a comma (e.g., 'enp0s25' or 'enp10s0,enp11s0')
    • PCAP_IFACE_TWEAK - if set to true, Malcolm will [use ethtool]({{ site.github.repository_url }}/blob/{{ site.github.build_revision }}/shared/bin/nic-capture-setup.sh) to disable NIC hardware offloading features and adjust ring buffer sizes for capture interface(s); this should be true if the interface(s) are being used for capture only, false if they are being used for management/communication
    • PCAP_ROTATE_MEGABYTES – used to specify how large a locally captured PCAP file can become (in megabytes) before it is closed for processing and a new PCAP file created
    • PCAP_ROTATE_MINUTES – used to specify a time interval (in minutes) after which a locally-captured PCAP file will be closed for processing and a new PCAP file created
  • process.env - settings for how the processes running inside Malcolm containers are executed
    • PUID and PGID - Docker runs all its containers as the privileged root user by default. For better security, Malcolm immediately drops to non-privileged user accounts for executing internal processes wherever possible. The PUID (process user ID) and PGID (process group ID) environment variables allow Malcolm to map internal non-privileged user accounts to a corresponding user account on the host. Note a few (including the logstash and netbox containers) may take a few extra minutes during startup if PUID and PGID are set to values other than the default 1000. This is expected and should not affect operation after the initial startup.
    • MALCOLM_PROFILE - Specifies the profile which determines the Malcolm containers to run (malcolm to run all containers, hedgehog to run only capture-related containers)
  • ssl.env - TLS-related settings used by many containers
  • suricata.env, suricata-live.env and suricata-offline.env - settings for Suricata
    • SURICATA_AUTO_ANALYZE_PCAP_FILES – if set to true, all PCAP files imported into Malcolm will automatically be analyzed by Suricata, and the resulting logs will also be imported (default false)
    • SURICATA_AUTO_ANALYZE_PCAP_THREADS – the number of threads available to Malcolm for analyzing Suricata logs (default 1)
    • SURICATA_CUSTOM_RULES_ONLY – if set to true, Malcolm will bypass the default Suricata ruleset and use only user-defined rules (./suricata/rules/*.rules).
    • SURICATA_UPDATE_RULES – if set to true, Suricata signatures will periodically be updated (default false)
    • SURICATA_LIVE_CAPTURE - if set to true, Suricata will monitor live traffic on the local interface(s) defined by PCAP_FILTER
    • SURICATA_ROTATED_PCAP - if set to true, Suricata can analyze PCAP files captured by netsniff-ng or tcpdump (see PCAP_ENABLE_NETSNIFF and PCAP_ENABLE_TCPDUMP, as well as SURICATA_AUTO_ANALYZE_PCAP_FILES); if SURICATA_LIVE_CAPTURE is true, this should be false; otherwise Suricata will see duplicate traffic
    • SURICATA_DISABLE_ICS_ALL - if set to true, this variable can be used to disable Malcolm's [built-in Suricata rules for Operational Technology/Industrial Control Systems (OT/ICS) vulnerabilities and exploits]({{ site.github.repository_url }}/blob/{{ site.github.build_revision }}/suricata/rules-default/OT)
    • See Tuning Suricata for other variables related to managing Suricata's performance and resource utilization.
  • upload-common.env - settings for dealing with PCAP files uploaded to Malcolm for analysis
    • AUTO_TAG – if set to true, Malcolm will automatically create Arkime sessions and Zeek logs with tags based on the filename, as described in Tagging (default true)
    • EXTRA_TAGS – a comma-separated list of default tags for data generated by Malcolm (default is an empty string)
    • PCAP_NODE_NAME - specifies the node name to associate with network traffic metadata
  • zeek.env, zeek-secret.env, zeek-live.env and zeek-offline.env - settings for Zeek and for scanning extracted files Zeek observes in network traffic
    • EXTRACTED_FILE_CAPA_VERBOSE – if set to true, all Capa rule hits will be logged; otherwise (false) only MITRE ATT&CK® technique classifications will be logged
    • EXTRACTED_FILE_ENABLE_CAPA – if set to true, Zeek-extracted files determined to be PE (portable executable) files will be scanned with Capa
    • EXTRACTED_FILE_ENABLE_CLAMAV – if set to true, Zeek-extracted files will be scanned with ClamAV
    • EXTRACTED_FILE_ENABLE_YARA – if set to true, Zeek-extracted files will be scanned with Yara
    • EXTRACTED_FILE_HTTP_SERVER_ENABLE – if set to true, the directory containing Zeek-extracted files will be served over HTTP at ./extracted-files/ (e.g., https://localhost/extracted-files/ if connecting locally)
    • EXTRACTED_FILE_HTTP_SERVER_ZIP – if to true, the Zeek-extracted files will be archived in a ZIP file upon download
    • EXTRACTED_FILE_HTTP_SERVER_KEY – specifies the password for the ZIP archive if EXTRACTED_FILE_HTTP_SERVER_ZIP is true; otherwise, this specifies the decryption password for encrypted Zeek-extracted files in an openssl enc-compatible format (e.g., openssl enc -aes-256-cbc -d -in example.exe.encrypted -out example.exe)
    • EXTRACTED_FILE_IGNORE_EXISTING – if set to true, files extant in ./zeek-logs/extract_files/ directory will be ignored on startup rather than scanned
    • EXTRACTED_FILE_PRESERVATION – determines behavior for preservation of Zeek-extracted files
    • EXTRACTED_FILE_UPDATE_RULES – if set to true, file scanner engines (e.g., ClamAV, Capa, Yara) will periodically update their rule definitions (default false)
    • EXTRACTED_FILE_YARA_CUSTOM_ONLY – if set to true, Malcolm will bypass the default Yara rulesets (Neo23x0/signature-base, reversinglabs/reversinglabs-yara-rules, and bartblaze/Yara-rules) and use only user-defined rules in ./yara/rules
    • VTOT_API2_KEY – used to specify a VirusTotal Public API v.20 key, which, if specified, will be used to submit hashes of Zeek-extracted files to VirusTotal
    • ZEEK_AUTO_ANALYZE_PCAP_FILES – if set to true, all PCAP files imported into Malcolm will automatically be analyzed by Zeek, and the resulting logs will also be imported (default false)
    • ZEEK_AUTO_ANALYZE_PCAP_THREADS – the number of threads available to Malcolm for analyzing Zeek logs (default 1)
    • ZEEK_JSON - whether Zeek should generate JSON format logs (true) or TSV format logs (false)
    • ZEEK_DISABLE_… - if set to true, each of these variables can be used to disable a certain Zeek function when it analyzes PCAP files (for example, setting ZEEK_DISABLE_LOG_PASSWORDS to true to disable logging of cleartext passwords)
    • ZEEK_…_PORTS - used to specify non-default ports to register certain Zeek analyzers (e.g., ZEEK_SYNCHROPHASOR_PORTS for the ICSNPP-Synchrophasor analyzer, ZEEK_GENISYS_PORTS for the ICSNPP-Genisys analyzer, and ZEEK_ENIP_PORTS for the ICSNPP-Ethernet/IP analyzer) formatted as a comma-separated list of Zeek ports (e.g., 12345/tcp or 4041/tcp,4042/udp)
    • ZEEK_DISABLE_ICS_ALL and ZEEK_DISABLE_ICS_… - if set to true, these variables can be used to disable Zeek's protocol analyzers for Operational Technology/Industrial Control Systems (OT/ICS) protocols
    • ZEEK_DISABLE_BEST_GUESS_ICS - see "Best Guess" Fingerprinting for ICS Protocols
    • ZEEK_EXTRACTOR_MODE – determines the file extraction behavior for file transfers detected by Zeek; see Automatic file extraction and scanning for more details
    • ZEEK_INTEL_FEED_SINCE - when querying a TAXII or MISP feed, only process threat indicators created or modified since the time represented by this value; it may be either a fixed date/time (01/01/2021) or relative interval (30 days ago)
    • ZEEK_INTEL_ITEM_EXPIRATION - specifies the value for Zeek's Intel::item_expiration timeout as used by the Zeek Intelligence Framework (default -1min, which disables item expiration)
    • ZEEK_INTEL_REFRESH_CRON_EXPRESSION - specifies a cron expression indicating the refresh interval for generating the Zeek Intelligence Framework files (defaults to empty, which disables automatic refresh)
    • ZEEK_LIVE_CAPTURE - if set to true, Zeek will monitor live traffic on the local interface(s) defined by PCAP_FILTER
      • See Tuning Zeek for other variables related to managing Zeek's performance and resource utilization.
    • ZEEK_LOCAL_NETS - specifies the value for Zeek's Site::local_nets variable (and networks.cfg for live capture) (e.g., 1.2.3.0/24,5.6.7.0/24); note that by default, Zeek considers IANA-registered private address space such as 10.0.0.0/8 and 192.168.0.0/16 site-local
    • ZEEK_ROTATED_PCAP - if set to true, Zeek can analyze captured PCAP files captured by netsniff-ng or tcpdump (see PCAP_ENABLE_NETSNIFF and PCAP_ENABLE_TCPDUMP, as well as ZEEK_AUTO_ANALYZE_PCAP_FILES); if ZEEK_LIVE_CAPTURE is true, this should be false; otherwise Zeek will see duplicate traffic
    • See Managing disk usage below for a discussion of the variables control automatic threshold-based deletion of the oldest Zeek-extracted files.

Command-line arguments

The ./scripts/configure script can also be run noninteractively which can be useful for scripting Malcolm setup. This behavior can be selected by supplying the -d or --defaults option on the command line. Running with the --help option will list the arguments accepted by the script:

$ ./scripts/configure --help
usage: configure <arguments>

Malcolm install script

options:
  -v [true|false], --verbose [true|false]
                        Verbose output
  -d [true|false], --defaults [true|false]
                        Accept defaults to prompts without user interaction
  -c [true|false], --configure [true|false]
                        Only do configuration (not installation)
…

Note that the value for any argument not specified on the command line will be reset to its default (as if for a new Malcolm installation) regardless of the setting's current value in the corresponding .env file. In other words, users who want to use the --defaults option should carefully review all available command-line options and choose all that apply.

Similarly, authentication-related settings can also be set noninteractively by using the command-line arguments for ./scripts/auth_setup.

Managing disk usage

In instances where Malcolm is deployed with the intention of running indefinitely, eventually the question arises of what to do when the file systems used for storing Malcolm's artifacts (e.g., PCAP files, raw logs, OpenSearch indices, extracted files, etc.). Malcolm provides options for tuning the "aging out" (deletion) of old artifacts to make room for newer data.

  • PCAP deletion is configured by environment variables in arkime.env:
    • MANAGE_PCAP_FILES – if set to true, all PCAP files imported into Malcolm will be marked as available for deletion by Arkime if available storage space becomes too low (default false)
    • ARKIME_FREESPACEG - when MANAGE_PCAP_FILES is true, this value is used by Arkime to determine when to delete the oldest PCAP files. Note that this variable represents the amount of free/unused/available desired on the file system: e.g., a value of 5% means "delete PCAP files if the amount of unused storage on the file system falls below 5%" (default 10%).
  • Zeek logs and Suricata logs are temporarily stored on disk as they are parsed, enriched, and indexed, and afterwards are periodically [pruned]({{ site.github.repository_url }}/blob/{{ site.github.build_revision }}/filebeat/scripts/clean-processed-folder.py) from the file system as they age, based on these variables in filebeat.env:
    • LOG_CLEANUP_MINUTES - specifies the age, in minutes, at which already-processed log files should be deleted
    • ZIP_CLEANUP_MINUTES - specifies the age, in minutes, at which the compressed archives containing already-processed log files should be deleted
  • Files extracted by Zeek stored in the ./zeek-logs/extract_files/ directory can be periodically [pruned]({{ site.github.repository_url }}/blob/{{ site.github.build_revision }}/shared/bin/prune_files.sh) based on the following variables in zeek.env. If either of the two threshold limits defined here are met, the oldest extracted files will be deleted until the limit is no longer met. Setting either of the threshold limits to 0 disables that check.
    • EXTRACTED_FILE_PRUNE_THRESHOLD_MAX_SIZE - specifies the maximum size, specified either in gigabytes or as a human-readable data size (e.g., 250G), that the ./zeek-logs/extract_files/ directory is allowed to contain before the prune condition triggers
    • EXTRACTED_FILE_PRUNE_THRESHOLD_TOTAL_DISK_USAGE_PERCENT - specifies a maximum fill percentage for the file system containing the ./zeek-logs/extract_files/; in other words, if the disk is more than this percentage utilized, the prune condition triggers
    • EXTRACTED_FILE_PRUNE_INTERVAL_SECONDS - the interval between checking the prune conditions, in seconds (default 300)
  • Index management policies can be handled via plugins provided as part of the OpenSearch and Elasticsearch platforms, respectively. In addition to those tools, the OPENSEARCH_INDEX_SIZE_PRUNE_LIMIT variable in dashboards-helper.env defines a maximum cumulative that OpenSearch indices are allowed to consume before the oldest indices [are deleted]({{ site.github.repository_url }}/blob/{{ site.github.build_revision }}/shared/bin/opensearch_index_size_prune.py), specified as either as a human-readable data size (e.g., 250G) or as a percentage of the total disk size (e.g., 70%): e.g., a value of 500G means "delete the oldest OpenSearch indices if the total space consumed by Malcolm's indices exceeds five hundred gigabytes."

Similar settings exist on for managing disk usage Hedgehog Linux.