ExaChk and AHF commands for Exadata

99 commands, grouped by what they do.

Check status and configuration

  • tfactl status

    Reports the running status of the Trace File Analyzer (TFA) on each node, whether the daemon is up, its port, and version. Read-only health check used to confirm TFA is collecting before relying on it for diagnostics.

    Read only

  • tfactl print hosts

    Lists hosts in the Oracle Trace File Analyzer configuration. Read-only command used to confirm the TFA cluster view before collection, troubleshooting, or comparing host coverage.

    Read only19c

  • ahfctl version -all

    Reports the installed Oracle Autonomous Health Framework version along with the versions of its bundled components (TFA, EXAchk/orachk, etc.). Read-only inventory command used to confirm the AHF level before running checks or comparing against a known-good baseline.

    Read only

  • tfactl print status

    Shows Oracle Trace File Analyzer status, including whether TFA is running and how it sees the local or cluster configuration. Read-only status command used before collecting diagnostics or troubleshooting TFA communication.

    Read only19c

  • tfactl summary -help

    Displays help for the TFA summary command. Use it to check available summary options on the installed TFA version.

    Read only19c

  • tfactl print actions

    Lists Oracle Trace File Analyzer action history, with optional filters available for status or age. Read-only diagnostic metadata used to see what TFA has recently done before deeper collection or troubleshooting.

    Read only19c

  • tfactl diagcollection -srdc <srdc_type> -help

    Shows help for a specific SRDC diagnostic collection type. Use it to understand required parameters before collecting SRDC data.

    Read only19c

  • tfactl managelogs -show usage [-gi|-database]

    Shows diagnostic log usage information for Grid Infrastructure, database, or all managed homes. Use it to size cleanup before purging.

    Read only19c

  • tfactl print config [-node all|local|<node_list> -name <parameter>]

    Prints TFA configuration values from the Berkeley database. Use it to inspect local, all-node, or named configuration parameters.

    Read only19c

  • tfactl diagnosetfa [-repo <repository>] [-tag <tag_name>] [-local]

    Collects diagnostic data about Oracle Trace File Analyzer itself from the local node or requested repository/tag. Use it when TFA is failing or support asks for TFA diagnostics.

    Safe19c

  • tfactl print directories

    Lists directories registered in Oracle Trace File Analyzer for diagnostic monitoring and collection. Read-only; useful for confirming which paths TFA knows about before troubleshooting missing logs or configuring collection scope.

    Read only19c

  • tfactl changes [options]

    Reports recent changes detected by Oracle Trace File Analyzer. Use it to correlate environment changes with incidents or performance shifts.

    Read only19c

  • tfactl events [options]

    Reports event information collected or analyzed by Oracle Trace File Analyzer. Use it to review incident-adjacent events from TFA metadata.

    Read only19c

  • tfactl print repository

    Displays the TFA repository location and related repository information. It is read-only and commonly used before checking disk usage or collection output.

    Read only19c

  • tfactl availability -list

    Lists resources configured for Availability Score enablement or disablement. Use it before changing availability tracking for a resource.

    Read only19c

  • tfactl search -json '<json_query>'

    Searches Oracle Trace File Analyzer indexed metadata using a JSON query and returns matching metadata. This is useful for machine-readable searches across events and diagnostics.

    Read only19c

  • tfactl access lsusers

    Lists users and groups currently allowed to access Oracle Trace File Analyzer. Use it to audit delegated TFA access.

    Read only19c

  • tfactl search -showdatatypes

    Lists available TFA search datatypes stored in the metadata index. This helps build valid JSON search queries.

    Read only19c

  • tfactl search -showdatatypes|-json [json_details]

    Searches TFA indexed metadata or displays datatype metadata, with optional JSON details. This documented grammar covers both datatype discovery and JSON search mode.

    Read only19c

  • tfactl access reset

    Resets the TFA access list to its default state. Use it when access entries have drifted and should be rebuilt.

    Caution19c

  • tfactl access removeall [-local]

    Removes all users and groups from the TFA access list, including defaults. Use -local to restrict the change to the local node when supported.

    Caution19c

Collect diagnostics

  • exachk

    Runs Oracle EXAchk to perform a health check across Exadata software and hardware components using the default command behavior. It produces findings and reports for proactive risk identification and support-ready diagnostics.

    Safe

  • exachk -a

    Runs Oracle EXAchk with the all-checks flag, producing a broader Exadata health-check report than the default invocation. The command is diagnostic and does not intentionally change database, Grid Infrastructure, or cell configuration.

    Safe

  • tfactl diagcollect

    Collects a TFA diagnostic bundle for Oracle stack troubleshooting. It reads logs and traces and packages them for analysis; it does not intentionally change database or cluster configuration.

    Safe

  • exachk -v

    Prints the installed Oracle EXAchk (AHF) version and exits. Pure read-only invocation used to confirm which EXAchk build is on the host before running a full health check or matching against a known-good release.

    Read only

  • exachk -checkdiscovery

    Checks whether cached discovery data exists for EXAchk. It is used to validate discovery-cache state before using or refreshing it.

    Read only19c

  • tfactl diagcollect -srdc ORA-00600

    Runs a Service Request Data Collection (SRDC) tailored to an ORA-00600 internal error, gathering the specific traces, logs, and dumps Oracle Support expects for that error class across the relevant nodes. It reads diagnostic data and writes a compressed collection bundle.

    Caution

  • tfactl diagcollect -database <db_name>[,<db_name>] -last <duration>

    Collects Oracle Trace File Analyzer diagnostics for one or more databases over a recent time window. The command writes a diagnostic collection bundle for later review or support upload.

    Safe19c

  • tfactl analyze -last 1h

    Analyzes Oracle Trace File Analyzer diagnostic data for the last hour and reports relevant events or patterns. It is a diagnostic read/analysis operation, though it can consume CPU, I/O, and space for generated output.

    Read only19c

  • tfactl diagcollect -from "2024-01-01 02:00:00" -to "2024-01-01 06:00:00"

    Collects a TFA (Trace File Analyzer) diagnostic bundle restricted to the given time window, gathering logs and traces from across the cluster for that interval. Read-mostly: it reads and packages diagnostics into an archive; it does not change database or cluster state.

    Safe

  • tfactl upload -sr <sr_number> [-user <user_id> | -wallet] <file> [<file> ...]

    Uploads diagnostic collections or files to an Oracle Support Service Request. The wallet form uses saved support credentials; the user form prompts or supplies a MOS user.

    Caution19c

  • tfactl run orachk [-autostop]

    Runs ORAchk through the TFACTL framework. It launches a diagnostic health check and writes report output for review, without directly changing database configuration.

    Safe19c

  • tfactl ips show incidents

    Lists incidents known to IPS through TFA. Use it to find incident ids before packing, adding, or removing incident diagnostics.

    Read only19c

  • tfactl ips pack incident <incident_id>

    Creates an IPS package for a specific ADR incident. It collects the diagnostic evidence tied to that incident so it can be reviewed or uploaded with a support request.

    Safe19c

  • exachk -profile exatier1

    Runs Oracle EXAchk using the exatier1 profile, limiting the run to the most critical Exadata checks. This is a diagnostic health-check run; it produces findings and reports without intentionally changing platform configuration.

    Safe

  • exachk -refreshdiscovery

    Refreshes cached EXAchk discovery data. It updates the discovery cache used by later health-check runs.

    Safe19c

  • tfactl ips show problems

    Lists problems known to IPS through TFA. It is read-only and helps choose the problem id to package or inspect.

    Read only19c

  • tfactl ips GENERATE PACKAGE <package_id> [IN <path>] [COMPLETE|INCREMENTAL]

    Generates the physical ZIP file for an IPS package in the current or specified directory. COMPLETE includes all package files, while INCREMENTAL includes changes since the last generation.

    Caution19c

  • tfactl ips pack {incident <incident_id> | problem <problem_id> | time <start_time> to <end_time>} [package <package_id>]

    Creates or updates an IPS package from incidents, problems, or a time range. It gathers ADR diagnostic material into a package for later review or upload to Oracle Support.

    Safe19c

  • tfactl ips show package <package_id>

    Displays details for an IPS package, including its scope and package state. It is read-only and useful before generating, unpacking, or uploading diagnostics.

    Read only19c

  • tfactl ips CREATE PACKAGE [INCIDENT <incid> | PROBLEM <prob_id> | PROBLEMKEY <prob_key> | SECONDS <seconds> | TIME <start_time> TO <end_time>] [CORRELATE BASIC|TYPICAL|ALL] [MANIFEST <file_spec>] [KEYFILE <file_spec>]

    Creates an ADR Incident Packaging Service package and optionally selects initial contents by incident, problem, problem key, or time range. It can also generate manifest or key files.

    Caution19c

  • exachk -querypatch all

    Lists all EXAchk patches known to the patch query mechanism. It is used before selecting a patch to apply or rollback.

    Read only19c

  • exachk -querypatch <patch_id>

    Queries whether a specific EXAchk patch is available or applicable. It inspects patch information rather than applying it.

    Read only19c

  • exachk -autostart

    Enables or starts EXAchk automatic health-check scheduling depending on the installed AHF behavior. It changes background health-check automation.

    Caution19c

  • tfactl ips add incident <incident_id> package <package_id>

    Adds a specific ADR incident to an existing IPS diagnostic package. It expands the package so the selected incident evidence is included for support or analysis.

    Safe19c

  • tfactl ips unpack file <file_name> package <package_id> [into <directory>]

    Extracts a selected file from an IPS diagnostic package. It writes the unpacked file to the chosen or default location for local inspection.

    Safe19c

  • tfactl ips unpack package <package_id> [into <directory>]

    Unpacks an IPS diagnostic package into files for inspection. It is a local file extraction operation and does not change database or Clusterware state.

    Safe19c

  • exachk -checkdbupload

    Checks the EXAchk database upload configuration. It reports the current upload setting without changing it.

    Read only19c

  • exachk -usediscovery -discoverydir <directory>

    Runs EXAchk using discovery data from an existing cache directory. It avoids rediscovery and relies on the cached configuration snapshot.

    Read only19c

  • tfactl ips show files package <package_id>

    Lists files associated with an IPS diagnostic package. Use it to confirm the package includes the expected traces, logs, and incident files before upload or unpacking.

    Read only19c

  • exachk -applypatch <exachk_patch_zip>

    Applies an EXAchk patch package between AHF releases. It modifies the EXAchk installation rather than only collecting diagnostics.

    Caution19c

  • tfactl ips ADD NEW INCIDENTS <package_id>

    Finds new incidents for the problems already represented in a package and adds the latest ones. This updates package contents based on current ADR incident state.

    Caution19c

  • tfactl ips ADD [INCIDENT <incid> | PROBLEM <prob_id> | PROBLEMKEY <prob_key> | SECONDS <seconds> | TIME <start_time> TO <end_time>] PACKAGE <package_id>

    Adds incidents, problems, problem keys, or time-ranged contents to an existing ADR Incident Packaging Service package. This changes the package contents that can later be finalized or generated.

    Caution19c

  • tfactl ips FINALIZE PACKAGE <package_id>

    Finalizes an IPS package so it is ready for shipping, automatically including correlated contents. Use it after selecting and reviewing package contents.

    Caution19c

  • exachk <option> -h

    Displays usage/help text for EXAchk, including the syntax and meaning of a given option. Read-only; prints help and exits without running any health check or collection.

    Read only

  • exachk -getdbupload

    Displays the current EXAchk database upload setting. It is an inspection command for AHF collection manager configuration.

    Read only19c

  • exachk -discovery -discoverydir <directory>

    Runs EXAchk discovery and writes the collected discovery data to a cache directory. The cache can later be reused to speed health-check collection.

    Safe19c

  • exachk -autostop

    Stops or disables EXAchk automatic health-check scheduling depending on the installed AHF behavior. It affects future automated health-check collection.

    Caution19c

  • tfactl ips ADD FILE <file_spec> PACKAGE <package_id>

    Adds a file from the same ADR base to an existing IPS package. Use it to include a specific trace or diagnostic file in the package.

    Caution19c

  • tfactl ips CHECK REMOTE KEYS <file_spec> PACKAGE <package_id>

    Creates a file with keys matching incidents in a specified IPS package. Use it to validate remote-key matching before exchanging or using key files.

    Read only19c

  • tfactl ips get manifest <package_id>

    Retrieves the manifest for an Incident Packaging Service package managed through TFACTL. It is read-only and helps inspect what diagnostic files and metadata are included before sharing or unpacking a package.

    Read only19c

  • tfactl ips get metadata <package_id>

    Retrieves metadata for a TFA IPS diagnostic package. Use it to inspect package attributes and support details without changing the package contents.

    Read only19c

  • tfactl ips show configuration

    Displays the current IPS configuration used by TFA diagnostic packaging. It is read-only and useful before changing packaging behavior or troubleshooting collection output.

    Read only19c

  • exachk -setdbupload all

    Sets EXAchk database upload behavior for all applicable health-check collections. It changes how collected health-check data is uploaded or consumed.

    Caution19c

  • exachk -rollbackpatch <patch_id>

    Rolls back an EXAchk patch that was previously applied. It changes the EXAchk installation state.

    Caution19c

  • tfactl ips set configuration <parameter> <value>

    Changes IPS configuration used by TFA diagnostic packaging. It affects later package behavior, so it should be treated as an operational configuration change rather than a display command.

    Caution19c

  • tfactl ips COPY IN FILE <file> [TO <new_name>] [OVERWRITE] PACKAGE <package_id> [INCIDENT <incid>]

    Copies an external file into ADR and associates it with an IPS package and optionally an incident. The OVERWRITE option can replace an existing file copy.

    Caution19c

  • exachk -rmdiscovery

    Removes cached EXAchk discovery data. It deletes cache artifacts but does not remove database or cluster resources.

    Safe19c

Start, stop and move

  • tfactl start [tool]

    Starts the Oracle Trace File Analyzer daemon on the local node, or starts a supported tool through TFACTL. It restores diagnostic collection after maintenance or an intentional stop.

    Safe

  • tfactl shutdown

    Shuts down Oracle Trace File Analyzer on the local node. Oracle documents it for Clusterware administration tasks such as host IP changes where TFA must be stopped and restarted.

    Caution26ai

  • tfactl stop [tool]

    Stops the Oracle Trace File Analyzer daemon on the local node, or stops a specific TFA-managed tool. It is used before maintenance that requires diagnostic collectors to be inactive.

    Caution

  • tfactl enable

    Enables automatic restart of the TFA daemon after failure or system reboot. This restores normal TFA service persistence.

    Caution19c

  • tfactl disable

    Stops the TFA daemon and disables automatic restart after failure or system reboot. Use it only when TFA must be taken out of service.

    Caution19c

  • tfactl collection stop <collection_id>

    Stops a running TFA diagnostic collection by collection ID. Obtain the ID from TFA print output before stopping it.

    Caution19c

  • tfactl rest -start

    Starts the TFA REST service. Oracle documents REST management under tfactl rest with start, stop, status, upgrade, uninstall, and configuration options.

    Caution19c

  • tfactl access enable

    Enables non-root access to Oracle Trace File Analyzer for users or groups on the access list. This restores delegated diagnostic access.

    Caution19c

  • tfactl availability enable -type <resource_type> -key <key> -value <value>

    Enables a resource for Availability Score. Use it to restore availability tracking after a planned disablement.

    Caution19c

  • tfactl access disable

    Disables non-root access to Oracle Trace File Analyzer. Root or sudo access remains the recovery path for administration.

    Caution19c

  • tfactl availability disable -type <resource_type> -key <key> -value <value> [-for <n>d|h|m]

    Disables a resource for Availability Score, optionally for a limited duration. This changes how TFA availability scoring treats the specified resource.

    Caution19c

Create and change

  • tfactl dbglevel -set <profile>

    Sets Oracle Grid Infrastructure trace levels using a named debug profile. This changes tracing behavior for GI components.

    Caution19c

  • tfactl host add <host_name>

    Adds a host to the Oracle Trace File Analyzer configuration. TFA contacts the other host, authenticates certificates, and synchronizes host lists before cluster-wide commands are active.

    Caution19c

  • tfactl add <dir> -collectall

    Adds a directory for custom TFA collection and marks it collect-all eligible. Oracle documents this legacy shorthand in custom collection examples.

    Caution19c

  • tfactl set tracelevel <COMPONENT> <LEVEL>

    Sets diagnostic trace verbosity for a TFA component. This changes troubleshooting output behavior and can increase diagnostic log volume until the trace level is restored.

    Caution

  • tfactl access add (-user <username> | -group <group_name>) [-local]

    Adds a user or group to the TFA access list and enables access either cluster-wide or locally. Oracle examples show both user and group forms.

    Caution19c

  • tfactl directory add <directory> [-public] [-exclusions|-noexclusions|-collectall] [-node all|<node_list>]

    Adds a directory to the TFA collection configuration. TFA uses this metadata to collect files from custom or undiscovered diagnostic directories.

    Caution19c

  • tfactl directory modify <directory> [-private|-public] [-exclusions|-noexclusions|-collectall]

    Modifies TFA collection policy for an existing directory. Options control visibility and how exclusions apply during collections.

    Caution19c

  • tfactl modify <dir> -collectall

    Changes a TFA collection directory so all files in that directory are collected. This broadens diagnostic collection scope and can increase repository size or include sensitive files.

    Caution19c

Patch and upgrade

  • exachk -upgrade

    Runs the EXAchk upgrade operation for the local EXAchk/AHF tooling. This changes the diagnostic tool installation or command family state rather than merely collecting a report.

    Caution

  • tfactl rest -upgrade

    Upgrades the TFA REST service configuration or deployed service components. Use it after TFA REST service updates require an upgrade step.

    Caution19c

Remove and drop

  • tfactl managelogs -purge -older <n>[m|h|d] [-gi|-database] [-dryrun]

    Purges old Automatic Diagnostic Repository log and trace files managed by TFA. Oracle examples recommend using -dryrun first to preview what would be removed.

    Dangerous19c

  • tfactl purge -older <n>[h|d] [-force]

    Deletes TFA diagnostic collections from the repository when they are older than the specified age. The -force option suppresses confirmation.

    Dangerous19c

  • tfactl ips DELETE PACKAGE <package_id>

    Drops an IPS package and its contents from ADR. Oracle documents this as deleting the package by package ID.

    Dangerous19c

  • tfactl access remove (-user <username> | -group <group_name>) [-local]

    Removes a user or group from the TFA access list. Use -local when only the local node should be changed.

    Caution19c

  • tfactl host [add <host_name> | remove <host_name>]

    Manages hosts in the Oracle Trace File Analyzer configuration. The same command family adds hosts to or removes hosts from the TFA cluster configuration.

    Caution19c

  • tfactl directory remove <directory> [-node all|<node_list>]

    Removes a directory from the TFA collection configuration. It does not delete the operating-system directory; it stops TFA from collecting it through this configuration.

    Caution19c

  • tfactl host remove <host_name>

    Removes a host from the Oracle Trace File Analyzer configuration. Use it when a node is decommissioned or should no longer participate in TFA cluster-wide commands.

    Caution19c

  • tfactl ips remove file <file_name> package <package_id>

    Removes a named file from an IPS package. It is useful for excluding irrelevant or sensitive diagnostic files before generating or uploading the package.

    Safe19c

  • tfactl ips remove incident <incident_id> package <package_id>

    Removes a selected incident from an existing IPS diagnostic package. This trims diagnostic package scope without changing the underlying ADR incident.

    Safe19c

  • tfactl ips remove {incident <incident_id> | problem <problem_id> | file <file_name>} package <package_id>

    Removes selected incidents, problems, or files from an existing IPS diagnostic package. It changes only the diagnostic package contents, not the database or Clusterware configuration.

    Safe19c

Search every command