diff options
Diffstat (limited to 'collectors/freeipmi.plugin')
| -rw-r--r-- | collectors/freeipmi.plugin/README.md | 141 | ||||
| -rw-r--r-- | collectors/freeipmi.plugin/freeipmi_plugin.c | 2512 | ||||
| -rw-r--r-- | collectors/freeipmi.plugin/metrics.csv | 18 | ||||
| -rw-r--r-- | collectors/freeipmi.plugin/multi_metadata.yaml | 199 |
4 files changed, 1587 insertions, 1283 deletions
diff --git a/collectors/freeipmi.plugin/README.md b/collectors/freeipmi.plugin/README.md index 47decd7ff..5a9fd93c0 100644 --- a/collectors/freeipmi.plugin/README.md +++ b/collectors/freeipmi.plugin/README.md @@ -11,7 +11,10 @@ learn_rel_path: "Integrations/Monitor/Devices" Netdata has a [freeipmi](https://www.gnu.org/software/freeipmi/) plugin. -> FreeIPMI provides in-band and out-of-band IPMI software based on the IPMI v1.5/2.0 specification. The IPMI specification defines a set of interfaces for platform management and is implemented by a number vendors for system management. The features of IPMI that most users will be interested in are sensor monitoring, system event monitoring, power control, and serial-over-LAN (SOL). +> FreeIPMI provides in-band and out-of-band IPMI software based on the IPMI v1.5/2.0 specification. The IPMI +> specification defines a set of interfaces for platform management and is implemented by a number vendors for system +> management. The features of IPMI that most users will be interested in are sensor monitoring, system event monitoring, +> power control, and serial-over-LAN (SOL). ## Installing the FreeIPMI plugin @@ -22,7 +25,8 @@ installed automatically due to the large number of dependencies it requires. When using a static build of Netdata, the FreeIPMI plugin will be included and installed automatically, though you will still need to have FreeIPMI installed on your system to be able to use the plugin. -When using a local build of Netdata, you need to ensure that the FreeIPMI development packages (typically called `libipmimonitoring-dev`, `libipmimonitoring-devel`, or `freeipmi-devel`) are installed when building Netdata. +When using a local build of Netdata, you need to ensure that the FreeIPMI development packages (typically +called `libipmimonitoring-dev`, `libipmimonitoring-devel`, or `freeipmi-devel`) are installed when building Netdata. ### Special Considerations @@ -30,7 +34,9 @@ Accessing IPMI requires root access, so the FreeIPMI plugin is automatically ins FreeIPMI does not work correctly on IBM POWER systems, thus Netdata’s FreeIPMI plugin is not usable on such systems. -If you have not previously used IPMI on your system, you will probably need to run the `ipmimonitoring` command as root to initiailze IPMI settings so that the Netdata plugin works correctly. It should return information about available seensors on the system. +If you have not previously used IPMI on your system, you will probably need to run the `ipmimonitoring` command as root +to initiailze IPMI settings so that the Netdata plugin works correctly. It should return information about available +seensors on the system. In some distributions `libipmimonitoring.pc` is located in a non-standard directory, which can cause building the plugin to fail when building Netdata from source. In that case you @@ -38,37 +44,68 @@ should find the file and link it to the standard pkg-config directory. Usually, /usr/lib/$(uname -m)-linux-gnu/pkgconfig/libipmimonitoring.pc/libipmimonitoring.pc /usr/lib/pkgconfig/libipmimonitoring.pc` resolves this issue. -## Netdata use +## Metrics -The plugin creates (up to) 8 charts, based on the information collected from IPMI: +The plugin does a speed test when it starts, to find out the duration needed by the IPMI processor to respond. Depending +on the speed of your IPMI processor, charts may need several seconds to show up on the dashboard. -1. number of sensors by state -2. number of events in SEL -3. Temperatures CELSIUS -4. Temperatures FAHRENHEIT -5. Voltages -6. Currents -7. Power -8. Fans +Metrics grouped by *scope*. -It also adds 2 alarms: +The scope defines the instance that the metric belongs to. An instance is uniquely identified by a set of labels. -1. Sensors in non-nominal state (i.e. warning and critical) -2. SEL is non empty +### global - +These metrics refer to the monitored host. -The plugin does a speed test when it starts, to find out the duration needed by the IPMI processor to respond. Depending on the speed of your IPMI processor, charts may need several seconds to show up on the dashboard. +This scope has no labels. -## `freeipmi.plugin` configuration +Metrics: + +| Metric | Dimensions | Unit | +|----------|:----------:|:------:| +| ipmi.sel | events | events | + +### sensor + +These metrics refer to the sensor. + +Labels: + +| Label | Description | +|-----------|-----------------------------------------------------------------------------------------------------------------| +| sensor | Sensor name. Same value as the "Name" column in the `ipmi-sensors` output. | +| type | Sensor type. Same value as the "Type" column in the `ipmi-sensors` output. | +| component | General sensor component. Identified by Netdata based on sensor name and type (e.g. System, Processor, Memory). | + +Metrics: + +| Metric | Dimensions | Unit | +|-----------------------------|:-----------------------------------:|:----------:| +| ipmi.sensor_state | nominal, critical, warning, unknown | state | +| ipmi.sensor_temperature_c | temperature | Celsius | +| ipmi.sensor_temperature_f | temperature | Fahrenheit | +| ipmi.sensor_voltage | voltage | Volts | +| ipmi.sensor_ampere | ampere | Amps | +| ipmi.sensor_fan_speed | rotations | RPM | +| ipmi.sensor_power | power | Watts | +| ipmi.sensor_reading_percent | percentage | % | + +## Alarms + +There are 2 alarms: + +- The sensor is in a warning or critical state. +- System Event Log (SEL) is non-empty. + +## Configuration The plugin supports a few options. To see them, run: ```text -# /usr/libexec/netdata/plugins.d/freeipmi.plugin -h +# ./freeipmi.plugin --help - netdata freeipmi.plugin 1.8.0-546-g72ce5d6b_rolling - Copyright (C) 2016-2017 Costa Tsaousis <costa@tsaousis.gr> + netdata freeipmi.plugin v1.40.0-137-gf162c25bd + Copyright (C) 2023 Netdata Inc. Released under GNU General Public License v3 or later. All rights reserved. @@ -86,17 +123,49 @@ The plugin supports a few options. To see them, run: no-sel enable/disable SEL collection default: enabled + reread-sdr-cache re-read SDR cache on every iteration + default: disabled + + interpret-oem-data attempt to parse OEM data + default: disabled + + assume-system-event-record + tread illegal SEL events records as normal + default: disabled + + ignore-non-interpretable-sensors + do not read sensors that cannot be interpreted + default: disabled + + bridge-sensors bridge sensors not owned by the BMC + default: disabled + + shared-sensors enable shared sensors, if found + default: disabled + + no-discrete-reading do not read sensors that their event/reading type code is invalid + default: enabled + + ignore-scanning-disabled + Ignore the scanning bit and read sensors no matter what + default: disabled + + assume-bmc-owner assume the BMC is the sensor owner no matter what + (usually bridging is required too) + default: disabled + hostname HOST username USER password PASS connect to remote IPMI host default: local IPMI processor + no-auth-code-check noauthcodecheck don't check the authentication codes returned - driver-type IPMIDRIVER - Specify the driver type to use instead of doing an auto selection. - The currently available outofband drivers are LAN and LAN_2_0, - which perform IPMI 1.5 and IPMI 2.0 respectively. + driver-type IPMIDRIVER + Specify the driver type to use instead of doing an auto selection. + The currently available outofband drivers are LAN and LAN_2_0, + which perform IPMI 1.5 and IPMI 2.0 respectively. The currently available inband drivers are KCS, SSIF, OPENIPMI and SUNBMC. sdr-cache-dir PATH directory for SDR cache files @@ -105,9 +174,15 @@ The plugin supports a few options. To see them, run: sensor-config-file FILE filename to read sensor configuration default: system default + sel-config-file FILE filename to read sel configuration + default: system default + ignore N1,N2,N3,... sensor IDs to ignore default: none + ignore-status N1,N2,N3,... sensor IDs to ignore status (nominal/warning/critical) + default: none + -v -V version print version and exit @@ -131,13 +206,17 @@ You can set these options in `/etc/netdata/netdata.conf` at this section: command options = ``` -Append to `command options =` the settings you need. The minimum `update every` is 5 (enforced internally by the plugin). IPMI is slow and CPU hungry. So, once every 5 seconds is pretty acceptable. +Append to `command options =` the settings you need. The minimum `update every` is 5 (enforced internally by the +plugin). IPMI is slow and CPU hungry. So, once every 5 seconds is pretty acceptable. ## Ignoring specific sensors -Specific sensor IDs can be excluded from freeipmi tools by editing `/etc/freeipmi/freeipmi.conf` and setting the IDs to be ignored at `ipmi-sensors-exclude-record-ids`. **However this file is not used by `libipmimonitoring`** (the library used by Netdata's `freeipmi.plugin`). +Specific sensor IDs can be excluded from freeipmi tools by editing `/etc/freeipmi/freeipmi.conf` and setting the IDs to +be ignored at `ipmi-sensors-exclude-record-ids`. **However this file is not used by `libipmimonitoring`** (the library +used by Netdata's `freeipmi.plugin`). -So, `freeipmi.plugin` supports the option `ignore` that accepts a comma separated list of sensor IDs to ignore. To configure it, edit `/etc/netdata/netdata.conf` and set: +So, `freeipmi.plugin` supports the option `ignore` that accepts a comma separated list of sensor IDs to ignore. To +configure it, edit `/etc/netdata/netdata.conf` and set: ``` [plugin:freeipmi] @@ -196,7 +275,9 @@ You can also permanently set the above setting by creating the file `/etc/modpro options ipmi_si kipmid_max_busy_us=10 ``` -This instructs the kernel IPMI module to pause for a tick between checking IPMI. Querying IPMI will be a lot slower now (e.g. several seconds for IPMI to respond), but `kipmi` will not use any noticeable CPU. You can also use a higher number (this is the number of microseconds to poll IPMI for a response, before waiting for a tick). +This instructs the kernel IPMI module to pause for a tick between checking IPMI. Querying IPMI will be a lot slower +now (e.g. several seconds for IPMI to respond), but `kipmi` will not use any noticeable CPU. You can also use a higher +number (this is the number of microseconds to poll IPMI for a response, before waiting for a tick). If you need to disable IPMI for Netdata, edit `/etc/netdata/netdata.conf` and set: diff --git a/collectors/freeipmi.plugin/freeipmi_plugin.c b/collectors/freeipmi.plugin/freeipmi_plugin.c index a2251891a..bfd867cc9 100644 --- a/collectors/freeipmi.plugin/freeipmi_plugin.c +++ b/collectors/freeipmi.plugin/freeipmi_plugin.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-3.0-or-later /* * netdata freeipmi.plugin - * Copyright (C) 2017 Costa Tsaousis + * Copyright (C) 2023 Netdata Inc. * GPL v3+ * * Based on: @@ -15,9 +15,59 @@ * UCRL-CODE-222073 */ +// ---------------------------------------------------------------------------- +// BEGIN NETDATA CODE + +// #define NETDATA_TIMING_REPORT 1 #include "libnetdata/libnetdata.h" #include "libnetdata/required_dummies.h" +// component names, based on our patterns +#define NETDATA_SENSOR_COMPONENT_MEMORY_MODULE "Memory Module" +#define NETDATA_SENSOR_COMPONENT_MEMORY "Memory" +#define NETDATA_SENSOR_COMPONENT_PROCESSOR "Processor" +#define NETDATA_SENSOR_COMPONENT_IPU "Image Processor" +#define NETDATA_SENSOR_COMPONENT_STORAGE "Storage" +#define NETDATA_SENSOR_COMPONENT_MOTHERBOARD "Motherboard" +#define NETDATA_SENSOR_COMPONENT_NETWORK "Network" +#define NETDATA_SENSOR_COMPONENT_POWER_SUPPLY "Power Supply" +#define NETDATA_SENSOR_COMPONENT_SYSTEM "System" +#define NETDATA_SENSOR_COMPONENT_PERIPHERAL "Peripheral" + +// netdata plugin defaults +#define SENSORS_DICT_KEY_SIZE 2048 // the max size of the key for the dictionary of sensors +#define SPEED_TEST_ITERATIONS 5 // how many times to repeat data collection to decide latency +#define IPMI_SENSORS_DASHBOARD_PRIORITY 90000 // the priority of the sensors charts on the dashboard +#define IPMI_SEL_DASHBOARD_PRIORITY 99000 // the priority of the SEL events chart on the dashboard +#define IPMI_SENSORS_MIN_UPDATE_EVERY 5 // the minimum data collection frequency for sensors +#define IPMI_SEL_MIN_UPDATE_EVERY 30 // the minimum data collection frequency for SEL events +#define IPMI_ENABLE_SEL_BY_DEFAULT true // true/false, to enable/disable SEL by default +#define IPMI_RESTART_EVERY_SECONDS 14400 // restart the plugin every this many seconds + // this is to prevent possible bugs/leaks in ipmi libraries +#define IPMI_RESTART_IF_SENSORS_DONT_ITERATE_EVERY_SECONDS (10 * 60) // stale data collection detection time + +// forward definition of functions and structures +struct netdata_ipmi_state; +static void netdata_update_ipmi_sensor_reading( + int record_id + , int sensor_number + , int sensor_type + , int sensor_state + , int sensor_units + , int sensor_reading_type + , char *sensor_name + , void *sensor_reading + , int event_reading_type_code + , int sensor_bitmask_type + , int sensor_bitmask + , char **sensor_bitmask_strings + , struct netdata_ipmi_state *state +); +static void netdata_update_ipmi_sel_events_count(struct netdata_ipmi_state *state, uint32_t events); + +// END NETDATA CODE +// ---------------------------------------------------------------------------- + #include <stdio.h> #include <stdlib.h> #include <stdint.h> @@ -27,22 +77,9 @@ #include <unistd.h> #include <sys/time.h> -#define IPMI_PARSE_DEVICE_LAN_STR "lan" -#define IPMI_PARSE_DEVICE_LAN_2_0_STR "lan_2_0" -#define IPMI_PARSE_DEVICE_LAN_2_0_STR2 "lan20" -#define IPMI_PARSE_DEVICE_LAN_2_0_STR3 "lan_20" -#define IPMI_PARSE_DEVICE_LAN_2_0_STR4 "lan2_0" -#define IPMI_PARSE_DEVICE_LAN_2_0_STR5 "lanplus" -#define IPMI_PARSE_DEVICE_KCS_STR "kcs" -#define IPMI_PARSE_DEVICE_SSIF_STR "ssif" -#define IPMI_PARSE_DEVICE_OPENIPMI_STR "openipmi" -#define IPMI_PARSE_DEVICE_OPENIPMI_STR2 "open" -#define IPMI_PARSE_DEVICE_SUNBMC_STR "sunbmc" -#define IPMI_PARSE_DEVICE_SUNBMC_STR2 "bmc" -#define IPMI_PARSE_DEVICE_INTELDCMI_STR "inteldcmi" - #include <ipmi_monitoring.h> #include <ipmi_monitoring_bitmasks.h> +#include <ipmi_monitoring_offsets.h> /* Communication Configuration - Initialize accordingly */ @@ -50,53 +87,38 @@ char *hostname = NULL; /* In-band Communication Configuration */ -int driver_type = -1; // IPMI_MONITORING_DRIVER_TYPE_KCS; /* or -1 for default */ -int disable_auto_probe = 0; /* probe for in-band device */ -unsigned int driver_address = 0; /* not used if probing */ -unsigned int register_spacing = 0; /* not used if probing */ -char *driver_device = NULL; /* not used if probing */ +int driver_type = -1; // IPMI_MONITORING_DRIVER_TYPE_KCS, etc. or -1 for default +int disable_auto_probe = 0; /* probe for in-band device */ +unsigned int driver_address = 0; /* not used if probing */ +unsigned int register_spacing = 0; /* not used if probing */ +char *driver_device = NULL; /* not used if probing */ /* Out-of-band Communication Configuration */ -int protocol_version = -1; //IPMI_MONITORING_PROTOCOL_VERSION_1_5; /* or -1 for default */ -char *username = "foousername"; -char *password = "foopassword"; -unsigned char *ipmi_k_g = NULL; -unsigned int ipmi_k_g_len = 0; -int privilege_level = -1; // IPMI_MONITORING_PRIVILEGE_LEVEL_USER; /* or -1 for default */ -int authentication_type = -1; // IPMI_MONITORING_AUTHENTICATION_TYPE_MD5; /* or -1 for default */ -int cipher_suite_id = 0; /* or -1 for default */ +int protocol_version = -1; // IPMI_MONITORING_PROTOCOL_VERSION_1_5, etc. or -1 for default +char *username = ""; +char *password = ""; +unsigned char *k_g = NULL; +unsigned int k_g_len = 0; +int privilege_level = -1; // IPMI_MONITORING_PRIVILEGE_LEVEL_USER, etc. or -1 for default +int authentication_type = -1; // IPMI_MONITORING_AUTHENTICATION_TYPE_MD5, etc. or -1 for default +int cipher_suite_id = -1; /* 0 or -1 for default */ int session_timeout = 0; /* 0 for default */ int retransmission_timeout = 0; /* 0 for default */ /* Workarounds - specify workaround flags if necessary */ unsigned int workaround_flags = 0; -/* Initialize w/ record id numbers to only monitor specific record ids */ -unsigned int record_ids[] = {0}; -unsigned int record_ids_length = 0; - -/* Initialize w/ sensor types to only monitor specific sensor types - * see ipmi_monitoring.h sensor types list. - */ -unsigned int sensor_types[] = {0}; -unsigned int sensor_types_length = 0; - /* Set to an appropriate alternate if desired */ char *sdr_cache_directory = "/tmp"; +char *sdr_sensors_cache_format = ".netdata-freeipmi-sensors-%H-on-%L.sdr"; +char *sdr_sel_cache_format = ".netdata-freeipmi-sel-%H-on-%L.sdr"; char *sensor_config_file = NULL; +char *sel_config_file = NULL; -/* Set to 1 or 0 to enable these sensor reading flags - * - See ipmi_monitoring.h for descriptions of these flags. - */ -int reread_sdr_cache = 0; -int ignore_non_interpretable_sensors = 0; -int bridge_sensors = 0; -int interpret_oem_data = 0; -int shared_sensors = 0; -int discrete_reading = 1; -int ignore_scanning_disabled = 0; -int assume_bmc_owner = 0; -int entity_sensor_names = 0; +// controlled via command line options +unsigned int global_sel_flags = IPMI_MONITORING_SEL_FLAGS_REREAD_SDR_CACHE; +unsigned int global_sensor_reading_flags = IPMI_MONITORING_SENSOR_READING_FLAGS_DISCRETE_READING|IPMI_MONITORING_SENSOR_READING_FLAGS_REREAD_SDR_CACHE; +bool remove_reread_sdr_after_first_use = true; /* Initialization flags * @@ -106,26 +128,10 @@ int entity_sensor_names = 0; */ unsigned int ipmimonitoring_init_flags = 0; -int errnum; - -// ---------------------------------------------------------------------------- -// SEL only variables - -/* Initialize w/ date range to only monitoring specific date range */ -char *date_begin = NULL; /* use MM/DD/YYYY format */ -char *date_end = NULL; /* use MM/DD/YYYY format */ - -int assume_system_event_record = 0; - -char *sel_config_file = NULL; - - // ---------------------------------------------------------------------------- // functions common to sensors and SEL -static void -_init_ipmi_config (struct ipmi_monitoring_ipmi_config *ipmi_config) -{ +static void initialize_ipmi_config (struct ipmi_monitoring_ipmi_config *ipmi_config) { fatal_assert(ipmi_config); ipmi_config->driver_type = driver_type; @@ -137,8 +143,8 @@ _init_ipmi_config (struct ipmi_monitoring_ipmi_config *ipmi_config) ipmi_config->protocol_version = protocol_version; ipmi_config->username = username; ipmi_config->password = password; - ipmi_config->k_g = ipmi_k_g; - ipmi_config->k_g_len = ipmi_k_g_len; + ipmi_config->k_g = k_g; + ipmi_config->k_g_len = k_g_len; ipmi_config->privilege_level = privilege_level; ipmi_config->authentication_type = authentication_type; ipmi_config->cipher_suite_id = cipher_suite_id; @@ -148,414 +154,566 @@ _init_ipmi_config (struct ipmi_monitoring_ipmi_config *ipmi_config) ipmi_config->workaround_flags = workaround_flags; } -#ifdef NETDATA_COMMENTED -static const char * -_get_sensor_type_string (int sensor_type) -{ - switch (sensor_type) - { +static const char *netdata_ipmi_get_sensor_type_string (int sensor_type, const char **component) { + switch (sensor_type) { case IPMI_MONITORING_SENSOR_TYPE_RESERVED: return ("Reserved"); + case IPMI_MONITORING_SENSOR_TYPE_TEMPERATURE: return ("Temperature"); + case IPMI_MONITORING_SENSOR_TYPE_VOLTAGE: return ("Voltage"); + case IPMI_MONITORING_SENSOR_TYPE_CURRENT: return ("Current"); + case IPMI_MONITORING_SENSOR_TYPE_FAN: return ("Fan"); + case IPMI_MONITORING_SENSOR_TYPE_PHYSICAL_SECURITY: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Physical Security"); + case IPMI_MONITORING_SENSOR_TYPE_PLATFORM_SECURITY_VIOLATION_ATTEMPT: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Platform Security Violation Attempt"); + case IPMI_MONITORING_SENSOR_TYPE_PROCESSOR: + *component = NETDATA_SENSOR_COMPONENT_PROCESSOR; return ("Processor"); + case IPMI_MONITORING_SENSOR_TYPE_POWER_SUPPLY: + *component = NETDATA_SENSOR_COMPONENT_POWER_SUPPLY; return ("Power Supply"); + case IPMI_MONITORING_SENSOR_TYPE_POWER_UNIT: + *component = NETDATA_SENSOR_COMPONENT_POWER_SUPPLY; return ("Power Unit"); + case IPMI_MONITORING_SENSOR_TYPE_COOLING_DEVICE: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Cooling Device"); + case IPMI_MONITORING_SENSOR_TYPE_OTHER_UNITS_BASED_SENSOR: return ("Other Units Based Sensor"); + case IPMI_MONITORING_SENSOR_TYPE_MEMORY: + *component = NETDATA_SENSOR_COMPONENT_MEMORY; return ("Memory"); + case IPMI_MONITORING_SENSOR_TYPE_DRIVE_SLOT: + *component = NETDATA_SENSOR_COMPONENT_STORAGE; return ("Drive Slot"); + case IPMI_MONITORING_SENSOR_TYPE_POST_MEMORY_RESIZE: + *component = NETDATA_SENSOR_COMPONENT_MEMORY; return ("POST Memory Resize"); + case IPMI_MONITORING_SENSOR_TYPE_SYSTEM_FIRMWARE_PROGRESS: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("System Firmware Progress"); + case IPMI_MONITORING_SENSOR_TYPE_EVENT_LOGGING_DISABLED: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Event Logging Disabled"); + case IPMI_MONITORING_SENSOR_TYPE_WATCHDOG1: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Watchdog 1"); + case IPMI_MONITORING_SENSOR_TYPE_SYSTEM_EVENT: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("System Event"); + case IPMI_MONITORING_SENSOR_TYPE_CRITICAL_INTERRUPT: return ("Critical Interrupt"); + case IPMI_MONITORING_SENSOR_TYPE_BUTTON_SWITCH: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Button/Switch"); + case IPMI_MONITORING_SENSOR_TYPE_MODULE_BOARD: return ("Module/Board"); + case IPMI_MONITORING_SENSOR_TYPE_MICROCONTROLLER_COPROCESSOR: + *component = NETDATA_SENSOR_COMPONENT_PROCESSOR; return ("Microcontroller/Coprocessor"); + case IPMI_MONITORING_SENSOR_TYPE_ADD_IN_CARD: return ("Add In Card"); + case IPMI_MONITORING_SENSOR_TYPE_CHASSIS: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Chassis"); + case IPMI_MONITORING_SENSOR_TYPE_CHIP_SET: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Chip Set"); + case IPMI_MONITORING_SENSOR_TYPE_OTHER_FRU: return ("Other Fru"); + case IPMI_MONITORING_SENSOR_TYPE_CABLE_INTERCONNECT: return ("Cable/Interconnect"); + case IPMI_MONITORING_SENSOR_TYPE_TERMINATOR: return ("Terminator"); + case IPMI_MONITORING_SENSOR_TYPE_SYSTEM_BOOT_INITIATED: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("System Boot Initiated"); + case IPMI_MONITORING_SENSOR_TYPE_BOOT_ERROR: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Boot Error"); + case IPMI_MONITORING_SENSOR_TYPE_OS_BOOT: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("OS Boot"); + case IPMI_MONITORING_SENSOR_TYPE_OS_CRITICAL_STOP: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("OS Critical Stop"); + case IPMI_MONITORING_SENSOR_TYPE_SLOT_CONNECTOR: return ("Slot/Connector"); + case IPMI_MONITORING_SENSOR_TYPE_SYSTEM_ACPI_POWER_STATE: return ("System ACPI Power State"); + case IPMI_MONITORING_SENSOR_TYPE_WATCHDOG2: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Watchdog 2"); + case IPMI_MONITORING_SENSOR_TYPE_PLATFORM_ALERT: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Platform Alert"); + case IPMI_MONITORING_SENSOR_TYPE_ENTITY_PRESENCE: return ("Entity Presence"); + case IPMI_MONITORING_SENSOR_TYPE_MONITOR_ASIC_IC: return ("Monitor ASIC/IC"); + case IPMI_MONITORING_SENSOR_TYPE_LAN: + *component = NETDATA_SENSOR_COMPONENT_NETWORK; return ("LAN"); + case IPMI_MONITORING_SENSOR_TYPE_MANAGEMENT_SUBSYSTEM_HEALTH: + *component = NETDATA_SENSOR_COMPONENT_SYSTEM; return ("Management Subsystem Health"); + case IPMI_MONITORING_SENSOR_TYPE_BATTERY: return ("Battery"); + case IPMI_MONITORING_SENSOR_TYPE_SESSION_AUDIT: return ("Session Audit"); + case IPMI_MONITORING_SENSOR_TYPE_VERSION_CHANGE: return ("Version Change"); + case IPMI_MONITORING_SENSOR_TYPE_FRU_STATE: return ("FRU State"); - } - - return ("Unrecognized"); -} -#endif // NETDATA_COMMENTED + case IPMI_MONITORING_SENSOR_TYPE_UNKNOWN: + return ("Unknown"); -// ---------------------------------------------------------------------------- -// BEGIN NETDATA CODE + default: + if(sensor_type >= IPMI_MONITORING_SENSOR_TYPE_OEM_MIN && sensor_type <= IPMI_MONITORING_SENSOR_TYPE_OEM_MAX) + return ("OEM"); -static int debug = 0; + return ("Unrecognized"); + } +} -static int netdata_update_every = 5; // this is the minimum update frequency -static int netdata_priority = 90000; -static int netdata_do_sel = 1; +#define netdata_ipmi_get_value_int(var, func, ctx) do { \ + (var) = func(ctx); \ + if( (var) < 0) { \ + collector_error("%s(): call to " #func " failed: %s", \ + __FUNCTION__, ipmi_monitoring_ctx_errormsg(ctx)); \ + goto cleanup; \ + } \ + timing_step(TIMING_STEP_FREEIPMI_READ_ ## var); \ +} while(0) + +#define netdata_ipmi_get_value_ptr(var, func, ctx) do { \ + (var) = func(ctx); \ + if(!(var)) { \ + collector_error("%s(): call to " #func " failed: %s", \ + __FUNCTION__, ipmi_monitoring_ctx_errormsg(ctx)); \ + goto cleanup; \ + } \ + timing_step(TIMING_STEP_FREEIPMI_READ_ ## var); \ +} while(0) + +#define netdata_ipmi_get_value_no_check(var, func, ctx) do { \ + (var) = func(ctx); \ + timing_step(TIMING_STEP_FREEIPMI_READ_ ## var); \ +} while(0) + +static int netdata_read_ipmi_sensors(struct ipmi_monitoring_ipmi_config *ipmi_config, struct netdata_ipmi_state *state) { + timing_init(); -static size_t netdata_sensors_updated = 0; -static size_t netdata_sensors_collected = 0; -static size_t netdata_sel_events = 0; -static size_t netdata_sensors_states_nominal = 0; -static size_t netdata_sensors_states_warning = 0; -static size_t netdata_sensors_states_critical = 0; + ipmi_monitoring_ctx_t ctx = NULL; + unsigned int sensor_reading_flags = global_sensor_reading_flags; + int i; + int sensor_count; + int rv = -1; -struct sensor { - int record_id; - int sensor_number; - int sensor_type; - int sensor_state; - int sensor_units; - char *sensor_name; + if (!(ctx = ipmi_monitoring_ctx_create ())) { + collector_error("ipmi_monitoring_ctx_create()"); + goto cleanup; + } - int sensor_reading_type; - union { - uint8_t bool_value; - uint32_t uint32_value; - double double_value; - } sensor_reading; + timing_step(TIMING_STEP_FREEIPMI_CTX_CREATE); - int sent; - int ignore; - int exposed; - int updated; - struct sensor *next; -} *sensors_root = NULL; + if (sdr_cache_directory) { + if (ipmi_monitoring_ctx_sdr_cache_directory (ctx, sdr_cache_directory) < 0) { + collector_error("ipmi_monitoring_ctx_sdr_cache_directory(): %s\n", ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } + } + if (sdr_sensors_cache_format) { + if (ipmi_monitoring_ctx_sdr_cache_filenames(ctx, sdr_sensors_cache_format) < 0) { + collector_error("ipmi_monitoring_ctx_sdr_cache_filenames(): %s\n", ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } + } -static void netdata_mark_as_not_updated() { - struct sensor *sn; - for(sn = sensors_root; sn ;sn = sn->next) - sn->updated = sn->sent = 0; + timing_step(TIMING_STEP_FREEIPMI_DSR_CACHE_DIR); - netdata_sensors_updated = 0; - netdata_sensors_collected = 0; - netdata_sel_events = 0; + // Must call otherwise only default interpretations ever used + // sensor_config_file can be NULL + if (ipmi_monitoring_ctx_sensor_config_file (ctx, sensor_config_file) < 0) { + collector_error( "ipmi_monitoring_ctx_sensor_config_file(): %s\n", ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } - netdata_sensors_states_nominal = 0; - netdata_sensors_states_warning = 0; - netdata_sensors_states_critical = 0; -} + timing_step(TIMING_STEP_FREEIPMI_SENSOR_CONFIG_FILE); + + if ((sensor_count = ipmi_monitoring_sensor_readings_by_record_id (ctx, + hostname, + ipmi_config, + sensor_reading_flags, + NULL, + 0, + NULL, + NULL)) < 0) { + collector_error( "ipmi_monitoring_sensor_readings_by_record_id(): %s", + ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } -static void send_chart_to_netdata_for_units(int units) { - struct sensor *sn, *sn_stored; - int dupfound, multiplier; + timing_step(TIMING_STEP_FREEIPMI_SENSOR_READINGS_BY_X); - switch(units) { - case IPMI_MONITORING_SENSOR_UNITS_CELSIUS: - printf("CHART ipmi.temperatures_c '' 'System Celsius Temperatures read by IPMI' 'Celsius' 'temperatures' 'ipmi.temperatures_c' 'line' %d %d\n" - , netdata_priority + 10 - , netdata_update_every - ); - break; + for (i = 0; i < sensor_count; i++, ipmi_monitoring_sensor_iterator_next (ctx)) { + int record_id, sensor_number, sensor_type, sensor_state, sensor_units, + sensor_bitmask_type, sensor_bitmask, event_reading_type_code, sensor_reading_type; - case IPMI_MONITORING_SENSOR_UNITS_FAHRENHEIT: - printf("CHART ipmi.temperatures_f '' 'System Fahrenheit Temperatures read by IPMI' 'Fahrenheit' 'temperatures' 'ipmi.temperatures_f' 'line' %d %d\n" - , netdata_priority + 11 - , netdata_update_every - ); - break; + char **sensor_bitmask_strings = NULL; + char *sensor_name = NULL; + void *sensor_reading; - case IPMI_MONITORING_SENSOR_UNITS_VOLTS: - printf("CHART ipmi.volts '' 'System Voltages read by IPMI' 'Volts' 'voltages' 'ipmi.voltages' 'line' %d %d\n" - , netdata_priority + 12 - , netdata_update_every - ); - break; + netdata_ipmi_get_value_int(record_id, ipmi_monitoring_sensor_read_record_id, ctx); + netdata_ipmi_get_value_int(sensor_number, ipmi_monitoring_sensor_read_sensor_number, ctx); + netdata_ipmi_get_value_int(sensor_type, ipmi_monitoring_sensor_read_sensor_type, ctx); + netdata_ipmi_get_value_ptr(sensor_name, ipmi_monitoring_sensor_read_sensor_name, ctx); + netdata_ipmi_get_value_int(sensor_state, ipmi_monitoring_sensor_read_sensor_state, ctx); + netdata_ipmi_get_value_int(sensor_units, ipmi_monitoring_sensor_read_sensor_units, ctx); + netdata_ipmi_get_value_int(sensor_bitmask_type, ipmi_monitoring_sensor_read_sensor_bitmask_type, ctx); + netdata_ipmi_get_value_int(sensor_bitmask, ipmi_monitoring_sensor_read_sensor_bitmask, ctx); + // it's ok for this to be NULL, i.e. sensor_bitmask == IPMI_MONITORING_SENSOR_BITMASK_TYPE_UNKNOWN + netdata_ipmi_get_value_no_check(sensor_bitmask_strings, ipmi_monitoring_sensor_read_sensor_bitmask_strings, ctx); + netdata_ipmi_get_value_int(sensor_reading_type, ipmi_monitoring_sensor_read_sensor_reading_type, ctx); + // whatever we read from the sensor, it is ok + netdata_ipmi_get_value_no_check(sensor_reading, ipmi_monitoring_sensor_read_sensor_reading, ctx); + netdata_ipmi_get_value_int(event_reading_type_code, ipmi_monitoring_sensor_read_event_reading_type_code, ctx); + + netdata_update_ipmi_sensor_reading( + record_id, sensor_number, sensor_type, sensor_state, sensor_units, sensor_reading_type, sensor_name, + sensor_reading, event_reading_type_code, sensor_bitmask_type, sensor_bitmask, sensor_bitmask_strings, + state + ); - case IPMI_MONITORING_SENSOR_UNITS_AMPS: - printf("CHART ipmi.amps '' 'System Current read by IPMI' 'Amps' 'current' 'ipmi.amps' 'line' %d %d\n" - , netdata_priority + 13 - , netdata_update_every - ); - break; +#ifdef NETDATA_COMMENTED + /* It is possible you may want to monitor specific event + * conditions that may occur. If that is the case, you may want + * to check out what specific bitmask type and bitmask events + * occurred. See ipmi_monitoring_bitmasks.h for a list of + * bitmasks and types. + */ - case IPMI_MONITORING_SENSOR_UNITS_RPM: - printf("CHART ipmi.rpm '' 'System Fans read by IPMI' 'RPM' 'fans' 'ipmi.rpm' 'line' %d %d\n" - , netdata_priority + 14 - , netdata_update_every - ); - break; + if (sensor_bitmask_type != IPMI_MONITORING_SENSOR_BITMASK_TYPE_UNKNOWN) + printf (", %Xh", sensor_bitmask); + else + printf (", N/A"); - case IPMI_MONITORING_SENSOR_UNITS_WATTS: - printf("CHART ipmi.watts '' 'System Power read by IPMI' 'Watts' 'power' 'ipmi.watts' 'line' %d %d\n" - , netdata_priority + 5 - , netdata_update_every - ); - break; + if (sensor_bitmask_type != IPMI_MONITORING_SENSOR_BITMASK_TYPE_UNKNOWN + && sensor_bitmask_strings) + { + unsigned int i = 0; - case IPMI_MONITORING_SENSOR_UNITS_PERCENT: - printf("CHART ipmi.percent '' 'System Metrics read by IPMI' '%%' 'other' 'ipmi.percent' 'line' %d %d\n" - , netdata_priority + 15 - , netdata_update_every - ); - break; + printf (","); - default: - for(sn = sensors_root; sn; sn = sn->next) - if(sn->sensor_units == units) - sn->ignore = 1; - return; - } + while (sensor_bitmask_strings[i]) + { + printf (" "); - for(sn = sensors_root; sn; sn = sn->next) { - dupfound = 0; - if(sn->sensor_units == units && sn->updated && !sn->ignore) { - sn->exposed = 1; - multiplier = 1; - - switch(sn->sensor_reading_type) { - case IPMI_MONITORING_SENSOR_READING_TYPE_DOUBLE: - multiplier = 1000; - // fallthrough - case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER8_BOOL: - case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER32: - for (sn_stored = sensors_root; sn_stored; sn_stored = sn_stored->next) { - if (sn_stored == sn) continue; - // If the name is a duplicate, append the sensor number - if ( !strcmp(sn_stored->sensor_name, sn->sensor_name) ) { - dupfound = 1; - printf("DIMENSION i%d_n%d_r%d '%s i%d' absolute 1 %d\n" - , sn->sensor_number - , sn->record_id - , sn->sensor_reading_type - , sn->sensor_name - , sn->sensor_number - , multiplier - ); - break; - } - } - // No duplicate name was found, display it just with Name - if (!dupfound) { - // display without ID - printf("DIMENSION i%d_n%d_r%d '%s' absolute 1 %d\n" - , sn->sensor_number - , sn->record_id - , sn->sensor_reading_type - , sn->sensor_name - , multiplier - ); - } - break; + printf ("'%s'", + sensor_bitmask_strings[i]); - default: - sn->ignore = 1; - break; + i++; } } + else + printf (", N/A"); + + printf ("\n"); +#endif // NETDATA_COMMENTED } -} -static void send_metrics_to_netdata_for_units(int units) { - struct sensor *sn; + rv = 0; - switch(units) { - case IPMI_MONITORING_SENSOR_UNITS_CELSIUS: - printf("BEGIN ipmi.temperatures_c\n"); - break; +cleanup: + if (ctx) + ipmi_monitoring_ctx_destroy (ctx); - case IPMI_MONITORING_SENSOR_UNITS_FAHRENHEIT: - printf("BEGIN ipmi.temperatures_f\n"); - break; + timing_report(); - case IPMI_MONITORING_SENSOR_UNITS_VOLTS: - printf("BEGIN ipmi.volts\n"); - break; + if(remove_reread_sdr_after_first_use) + global_sensor_reading_flags &= ~(IPMI_MONITORING_SENSOR_READING_FLAGS_REREAD_SDR_CACHE); - case IPMI_MONITORING_SENSOR_UNITS_AMPS: - printf("BEGIN ipmi.amps\n"); - break; + return (rv); +} - case IPMI_MONITORING_SENSOR_UNITS_RPM: - printf("BEGIN ipmi.rpm\n"); - break; - case IPMI_MONITORING_SENSOR_UNITS_WATTS: - printf("BEGIN ipmi.watts\n"); - break; +static int netdata_get_ipmi_sel_events_count(struct ipmi_monitoring_ipmi_config *ipmi_config, struct netdata_ipmi_state *state) { + timing_init(); - case IPMI_MONITORING_SENSOR_UNITS_PERCENT: - printf("BEGIN ipmi.percent\n"); - break; + ipmi_monitoring_ctx_t ctx = NULL; + unsigned int sel_flags = global_sel_flags; + int sel_count; + int rv = -1; - default: - for(sn = sensors_root; sn; sn = sn->next) - if(sn->sensor_units == units) - sn->ignore = 1; - return; + if (!(ctx = ipmi_monitoring_ctx_create ())) { + collector_error("ipmi_monitoring_ctx_create()"); + goto cleanup; } - for(sn = sensors_root; sn; sn = sn->next) { - if(sn->sensor_units == units && sn->updated && !sn->sent && !sn->ignore) { - netdata_sensors_updated++; + if (sdr_cache_directory) { + if (ipmi_monitoring_ctx_sdr_cache_directory (ctx, sdr_cache_directory) < 0) { + collector_error( "ipmi_monitoring_ctx_sdr_cache_directory(): %s", ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } + } + if (sdr_sel_cache_format) { + if (ipmi_monitoring_ctx_sdr_cache_filenames(ctx, sdr_sel_cache_format) < 0) { + collector_error("ipmi_monitoring_ctx_sdr_cache_filenames(): %s\n", ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } + } - sn->sent = 1; + // Must call otherwise only default interpretations ever used + // sel_config_file can be NULL + if (ipmi_monitoring_ctx_sel_config_file (ctx, sel_config_file) < 0) { + collector_error( "ipmi_monitoring_ctx_sel_config_file(): %s", + ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } - switch(sn->sensor_reading_type) { - case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER8_BOOL: - printf("SET i%d_n%d_r%d = %u\n" - , sn->sensor_number - , sn->record_id - , sn->sensor_reading_type - , sn->sensor_reading.bool_value - ); - break; + if ((sel_count = ipmi_monitoring_sel_by_record_id (ctx, + hostname, + ipmi_config, + sel_flags, + NULL, + 0, + NULL, + NULL)) < 0) { + collector_error( "ipmi_monitoring_sel_by_record_id(): %s", + ipmi_monitoring_ctx_errormsg (ctx)); + goto cleanup; + } - case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER32: - printf("SET i%d_n%d_r%d = %u\n" - , sn->sensor_number - , sn->record_id - , sn->sensor_reading_type - , sn->sensor_reading.uint32_value - ); - break; + netdata_update_ipmi_sel_events_count(state, sel_count); - case IPMI_MONITORING_SENSOR_READING_TYPE_DOUBLE: - printf("SET i%d_n%d_r%d = %lld\n" - , sn->sensor_number - , sn->record_id - , sn->sensor_reading_type - , (long long int)(sn->sensor_reading.double_value * 1000) - ); - break; + rv = 0; - default: - sn->ignore = 1; - break; - } - } - } +cleanup: + if (ctx) + ipmi_monitoring_ctx_destroy (ctx); + + timing_report(); - printf("END\n"); + if(remove_reread_sdr_after_first_use) + global_sel_flags &= ~(IPMI_MONITORING_SEL_FLAGS_REREAD_SDR_CACHE); + + return (rv); } -static void send_metrics_to_netdata() { - static int sel_chart_generated = 0, sensors_states_chart_generated = 0; - struct sensor *sn; +// ---------------------------------------------------------------------------- +// copied from freeipmi codebase commit 8dea6dec4012d0899901e595f2c868a05e1cefed +// added netdata_ in-front to not overwrite library functions + +// FROM: common/miscutil/network.c +static int netdata_host_is_localhost (const char *host) { + /* Ordered by my assumption of most popular */ + if (!strcasecmp (host, "localhost") + || !strcmp (host, "127.0.0.1") + || !strcasecmp (host, "ipv6-localhost") + || !strcmp (host, "::1") + || !strcasecmp (host, "ip6-localhost") + || !strcmp (host, "0:0:0:0:0:0:0:1")) + return (1); - if(netdata_do_sel && !sel_chart_generated) { - sel_chart_generated = 1; - printf("CHART ipmi.events '' 'IPMI Events' 'events' 'events' ipmi.sel area %d %d\n" - , netdata_priority + 2 - , netdata_update_every - ); - printf("DIMENSION events '' absolute 1 1\n"); - } + return (0); +} - if(!sensors_states_chart_generated) { - sensors_states_chart_generated = 1; - printf("CHART ipmi.sensors_states '' 'IPMI Sensors State' 'sensors' 'states' ipmi.sensors_states line %d %d\n" - , netdata_priority + 1 - , netdata_update_every - ); - printf("DIMENSION nominal '' absolute 1 1\n"); - printf("DIMENSION critical '' absolute 1 1\n"); - printf("DIMENSION warning '' absolute 1 1\n"); - } +// FROM: common/parsecommon/parse-common.h +#define IPMI_PARSE_DEVICE_LAN_STR "lan" +#define IPMI_PARSE_DEVICE_LAN_2_0_STR "lan_2_0" +#define IPMI_PARSE_DEVICE_LAN_2_0_STR2 "lan20" +#define IPMI_PARSE_DEVICE_LAN_2_0_STR3 "lan_20" +#define IPMI_PARSE_DEVICE_LAN_2_0_STR4 "lan2_0" +#define IPMI_PARSE_DEVICE_LAN_2_0_STR5 "lanplus" +#define IPMI_PARSE_DEVICE_KCS_STR "kcs" +#define IPMI_PARSE_DEVICE_SSIF_STR "ssif" +#define IPMI_PARSE_DEVICE_OPENIPMI_STR "openipmi" +#define IPMI_PARSE_DEVICE_OPENIPMI_STR2 "open" +#define IPMI_PARSE_DEVICE_SUNBMC_STR "sunbmc" +#define IPMI_PARSE_DEVICE_SUNBMC_STR2 "bmc" +#define IPMI_PARSE_DEVICE_INTELDCMI_STR "inteldcmi" - // generate the CHART/DIMENSION lines, if we have to - for(sn = sensors_root; sn; sn = sn->next) - if(sn->updated && !sn->exposed && !sn->ignore) - send_chart_to_netdata_for_units(sn->sensor_units); +// FROM: common/parsecommon/parse-common.c +// changed the return values to match ipmi_monitoring.h +static int netdata_parse_outofband_driver_type (const char *str) { + if (strcasecmp (str, IPMI_PARSE_DEVICE_LAN_STR) == 0) + return (IPMI_MONITORING_PROTOCOL_VERSION_1_5); - if(netdata_do_sel) { - printf( - "BEGIN ipmi.events\n" - "SET events = %zu\n" - "END\n" - , netdata_sel_events - ); - } + /* support "lanplus" for those that might be used to ipmitool. + * support typo variants to ease. + */ + else if (strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR) == 0 + || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR2) == 0 + || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR3) == 0 + || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR4) == 0 + || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR5) == 0) + return (IPMI_MONITORING_PROTOCOL_VERSION_2_0); - printf( - "BEGIN ipmi.sensors_states\n" - "SET nominal = %zu\n" - "SET warning = %zu\n" - "SET critical = %zu\n" - "END\n" - , netdata_sensors_states_nominal - , netdata_sensors_states_warning - , netdata_sensors_states_critical - ); - - // send metrics to netdata - for(sn = sensors_root; sn; sn = sn->next) - if(sn->updated && sn->exposed && !sn->sent && !sn->ignore) - send_metrics_to_netdata_for_units(sn->sensor_units); + return (-1); +} +// FROM: common/parsecommon/parse-common.c +// changed the return values to match ipmi_monitoring.h +static int netdata_parse_inband_driver_type (const char *str) { + if (strcasecmp (str, IPMI_PARSE_DEVICE_KCS_STR) == 0) + return (IPMI_MONITORING_DRIVER_TYPE_KCS); + else if (strcasecmp (str, IPMI_PARSE_DEVICE_SSIF_STR) == 0) + return (IPMI_MONITORING_DRIVER_TYPE_SSIF); + /* support "open" for those that might be used to + * ipmitool. + */ + else if (strcasecmp (str, IPMI_PARSE_DEVICE_OPENIPMI_STR) == 0 + || strcasecmp (str, IPMI_PARSE_DEVICE_OPENIPMI_STR2) == 0) + return (IPMI_MONITORING_DRIVER_TYPE_OPENIPMI); + /* support "bmc" for those that might be used to + * ipmitool. + */ + else if (strcasecmp (str, IPMI_PARSE_DEVICE_SUNBMC_STR) == 0 + || strcasecmp (str, IPMI_PARSE_DEVICE_SUNBMC_STR2) == 0) + return (IPMI_MONITORING_DRIVER_TYPE_SUNBMC); + +#ifdef IPMI_MONITORING_DRIVER_TYPE_INTELDCMI + else if (strcasecmp (str, IPMI_PARSE_DEVICE_INTELDCMI_STR) == 0) + return (IPMI_MONITORING_DRIVER_TYPE_INTELDCMI); +#endif // IPMI_MONITORING_DRIVER_TYPE_INTELDCMI + + return (-1); } +// ---------------------------------------------------------------------------- +// BEGIN NETDATA CODE + +typedef enum __attribute__((packed)) { + IPMI_COLLECT_TYPE_SENSORS = (1 << 0), + IPMI_COLLECT_TYPE_SEL = (1 << 1), +} IPMI_COLLECTION_TYPE; + +struct sensor { + int sensor_type; + int sensor_state; + int sensor_units; + char *sensor_name; + + int sensor_reading_type; + union { + uint8_t bool_value; + uint32_t uint32_value; + double double_value; + } sensor_reading; + + // netdata provided + const char *context; + const char *title; + const char *units; + const char *family; + const char *chart_type; + const char *dimension; + int priority; + + const char *type; + const char *component; + + int multiplier; + bool do_metric; + bool do_state; + bool metric_chart_sent; + bool state_chart_sent; + usec_t last_collected_metric_ut; + usec_t last_collected_state_ut; +}; + +typedef enum __attribute__((packed)) { + ICS_INIT, + ICS_INIT_FAILED, + ICS_RUNNING, + ICS_FAILED, +} IPMI_COLLECTOR_STATUS; + +struct netdata_ipmi_state { + bool debug; + + struct { + IPMI_COLLECTOR_STATUS status; + usec_t last_iteration_ut; + size_t collected; + usec_t now_ut; + usec_t freq_ut; + int priority; + DICTIONARY *dict; + } sensors; + + struct { + IPMI_COLLECTOR_STATUS status; + usec_t last_iteration_ut; + size_t events; + usec_t now_ut; + usec_t freq_ut; + int priority; + } sel; + + struct { + usec_t now_ut; + } updates; +}; + +// ---------------------------------------------------------------------------- +// excluded record ids maintenance (both for sensor data and state) + static int *excluded_record_ids = NULL; size_t excluded_record_ids_length = 0; -static void excluded_record_ids_parse(const char *s) { +static void excluded_record_ids_parse(const char *s, bool debug) { if(!s) return; while(*s) { @@ -567,18 +725,14 @@ static void excluded_record_ids_parse(const char *s) { s = e; if(n != 0) { - excluded_record_ids = realloc(excluded_record_ids, (excluded_record_ids_length + 1) * sizeof(int)); - if(!excluded_record_ids) { - fprintf(stderr, "freeipmi.plugin: failed to allocate memory. Exiting."); - exit(1); - } + excluded_record_ids = reallocz(excluded_record_ids, (excluded_record_ids_length + 1) * sizeof(int)); excluded_record_ids[excluded_record_ids_length++] = (int)n; } } } if(debug) { - fprintf(stderr, "freeipmi.plugin: excluded record ids:"); + fprintf(stderr, "%s: excluded record ids:", program_name); size_t i; for(i = 0; i < excluded_record_ids_length; i++) { fprintf(stderr, " %d", excluded_record_ids[i]); @@ -590,7 +744,7 @@ static void excluded_record_ids_parse(const char *s) { static int *excluded_status_record_ids = NULL; size_t excluded_status_record_ids_length = 0; -static void excluded_status_record_ids_parse(const char *s) { +static void excluded_status_record_ids_parse(const char *s, bool debug) { if(!s) return; while(*s) { @@ -602,18 +756,14 @@ static void excluded_status_record_ids_parse(const char *s) { s = e; if(n != 0) { - excluded_status_record_ids = realloc(excluded_status_record_ids, (excluded_status_record_ids_length + 1) * sizeof(int)); - if(!excluded_status_record_ids) { - fprintf(stderr, "freeipmi.plugin: failed to allocate memory. Exiting."); - exit(1); - } + excluded_status_record_ids = reallocz(excluded_status_record_ids, (excluded_status_record_ids_length + 1) * sizeof(int)); excluded_status_record_ids[excluded_status_record_ids_length++] = (int)n; } } } if(debug) { - fprintf(stderr, "freeipmi.plugin: excluded status record ids:"); + fprintf(stderr, "%s: excluded status record ids:", program_name); size_t i; for(i = 0; i < excluded_status_record_ids_length; i++) { fprintf(stderr, " %d", excluded_status_record_ids[i]); @@ -645,959 +795,649 @@ static int excluded_status_record_ids_check(int record_id) { return 0; } -static void netdata_get_sensor( - int record_id - , int sensor_number - , int sensor_type - , int sensor_state - , int sensor_units - , int sensor_reading_type - , char *sensor_name - , void *sensor_reading -) { - // find the sensor record - struct sensor *sn; - for(sn = sensors_root; sn ;sn = sn->next) - if( sn->record_id == record_id && - sn->sensor_number == sensor_number && - sn->sensor_reading_type == sensor_reading_type && - sn->sensor_units == sensor_units && - !strcmp(sn->sensor_name, sensor_name) - ) - break; +// ---------------------------------------------------------------------------- +// data collection functions - if(!sn) { - // not found, create it - // check if it is excluded - if(excluded_record_ids_check(record_id)) { - if(debug) fprintf(stderr, "Sensor '%s' is excluded by excluded_record_ids_check()\n", sensor_name); - return; - } +struct { + const char *search; + SIMPLE_PATTERN *pattern; + const char *label; +} sensors_component_patterns[] = { - if(debug) fprintf(stderr, "Allocating new sensor data record for sensor '%s', id %d, number %d, type %d, state %d, units %d, reading_type %d\n", sensor_name, record_id, sensor_number, sensor_type, sensor_state, sensor_units, sensor_reading_type); + // The order is important! + // They are evaluated top to bottom + // The first the matches is used - sn = calloc(1, sizeof(struct sensor)); - if(!sn) { - fatal("cannot allocate %zu bytes of memory.", sizeof(struct sensor)); - } + { + .search = "*DIMM*|*_DIM*|*VTT*|*VDDQ*|*ECC*|*MEM*CRC*|*MEM*BD*", + .label = NETDATA_SENSOR_COMPONENT_MEMORY_MODULE, + }, + { + .search = "*CPU*|SOC_*|*VDDCR*|P*_VDD*|*_DTS|*VCORE*|*PROC*", + .label = NETDATA_SENSOR_COMPONENT_PROCESSOR, + }, + { + .search = "IPU*", + .label = NETDATA_SENSOR_COMPONENT_IPU, + }, + { + .search = "M2_*|*SSD*|*HSC*|*HDD*|*NVME*", + .label = NETDATA_SENSOR_COMPONENT_STORAGE, + }, + { + .search = "MB_*|*PCH*|*VBAT*|*I/O*BD*|*IO*BD*", + .label = NETDATA_SENSOR_COMPONENT_MOTHERBOARD, + }, + { + .search = "Watchdog|SEL|SYS_*|*CHASSIS*", + .label = NETDATA_SENSOR_COMPONENT_SYSTEM, + }, + { + .search = "PS*|P_*|*PSU*|*PWR*|*TERMV*|*D2D*", + .label = NETDATA_SENSOR_COMPONENT_POWER_SUPPLY, + }, - sn->record_id = record_id; - sn->sensor_number = sensor_number; - sn->sensor_type = sensor_type; - sn->sensor_state = sensor_state; - sn->sensor_units = sensor_units; - sn->sensor_reading_type = sensor_reading_type; - sn->sensor_name = strdup(sensor_name); - if(!sn->sensor_name) { - fatal("cannot allocate %zu bytes of memory.", strlen(sensor_name)); + // fallback components + { + .search = "VR_P*|*VRMP*", + .label = NETDATA_SENSOR_COMPONENT_PROCESSOR, + }, + { + .search = "*VSB*|*PS*", + .label = NETDATA_SENSOR_COMPONENT_POWER_SUPPLY, + }, + { + .search = "*MEM*|*MEM*RAID*", + .label = NETDATA_SENSOR_COMPONENT_MEMORY, + }, + { + .search = "*RAID*", // there is also "Memory RAID", so keep this after memory + .label = NETDATA_SENSOR_COMPONENT_STORAGE, + }, + { + .search = "*PERIPHERAL*|*USB*", + .label = NETDATA_SENSOR_COMPONENT_PERIPHERAL, + }, + { + .search = "*FAN*|*12V*|*VCC*|*PCI*|*CHIPSET*|*AMP*|*BD*", + .label = NETDATA_SENSOR_COMPONENT_SYSTEM, + }, + + // terminator + { + .search = NULL, + .label = NULL, } +}; - sn->next = sensors_root; - sensors_root = sn; - } - else { - if(debug) fprintf(stderr, "Reusing sensor record for sensor '%s', id %d, number %d, type %d, state %d, units %d, reading_type %d\n", sensor_name, record_id, sensor_number, sensor_type, sensor_state, sensor_units, sensor_reading_type); +static const char *netdata_sensor_name_to_component(const char *sensor_name) { + for(int i = 0; sensors_component_patterns[i].search ;i++) { + if(!sensors_component_patterns[i].pattern) + sensors_component_patterns[i].pattern = simple_pattern_create(sensors_component_patterns[i].search, "|", SIMPLE_PATTERN_EXACT, false); + + if(simple_pattern_matches(sensors_component_patterns[i].pattern, sensor_name)) + return sensors_component_patterns[i].label; } - switch(sensor_reading_type) { + return "Other"; +} + +const char *netdata_collect_type_to_string(IPMI_COLLECTION_TYPE type) { + if((type & (IPMI_COLLECT_TYPE_SENSORS|IPMI_COLLECT_TYPE_SEL)) == (IPMI_COLLECT_TYPE_SENSORS|IPMI_COLLECT_TYPE_SEL)) + return "sensors,sel"; + if(type & IPMI_COLLECT_TYPE_SEL) + return "sel"; + if(type & IPMI_COLLECT_TYPE_SENSORS) + return "sensors"; + + return "unknown"; +} + +static void netdata_sensor_set_value(struct sensor *sn, void *sensor_reading, struct netdata_ipmi_state *state __maybe_unused) { + switch(sn->sensor_reading_type) { case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER8_BOOL: sn->sensor_reading.bool_value = *((uint8_t *)sensor_reading); - sn->updated = 1; - netdata_sensors_collected++; break; case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER32: sn->sensor_reading.uint32_value = *((uint32_t *)sensor_reading); - sn->updated = 1; - netdata_sensors_collected++; break; case IPMI_MONITORING_SENSOR_READING_TYPE_DOUBLE: sn->sensor_reading.double_value = *((double *)sensor_reading); - sn->updated = 1; - netdata_sensors_collected++; break; default: - if(debug) fprintf(stderr, "Unknown reading type - Ignoring sensor record for sensor '%s', id %d, number %d, type %d, state %d, units %d, reading_type %d\n", sensor_name, record_id, sensor_number, sensor_type, sensor_state, sensor_units, sensor_reading_type); - sn->ignore = 1; + case IPMI_MONITORING_SENSOR_READING_TYPE_UNKNOWN: + sn->do_metric = false; break; } +} - // check if it is excluded - if(excluded_status_record_ids_check(record_id)) { - if(debug) fprintf(stderr, "Sensor '%s' is excluded for status check, by excluded_status_record_ids_check()\n", sensor_name); +static void netdata_update_ipmi_sensor_reading( + int record_id + , int sensor_number + , int sensor_type + , int sensor_state + , int sensor_units + , int sensor_reading_type + , char *sensor_name + , void *sensor_reading + , int event_reading_type_code __maybe_unused + , int sensor_bitmask_type __maybe_unused + , int sensor_bitmask __maybe_unused + , char **sensor_bitmask_strings __maybe_unused + , struct netdata_ipmi_state *state +) { + if(unlikely(sensor_state == IPMI_MONITORING_STATE_UNKNOWN && + sensor_type == IPMI_MONITORING_SENSOR_TYPE_UNKNOWN && + sensor_units == IPMI_MONITORING_SENSOR_UNITS_UNKNOWN && + sensor_reading_type == IPMI_MONITORING_SENSOR_READING_TYPE_UNKNOWN && + (!sensor_name || !*sensor_name))) + // we can't do anything about this sensor - everything is unknown return; - } - switch(sensor_state) { - case IPMI_MONITORING_STATE_NOMINAL: - netdata_sensors_states_nominal++; - break; + if(unlikely(!sensor_name || !*sensor_name)) + sensor_name = "UNNAMED"; - case IPMI_MONITORING_STATE_WARNING: - netdata_sensors_states_warning++; - break; + state->sensors.collected++; - case IPMI_MONITORING_STATE_CRITICAL: - netdata_sensors_states_critical++; - break; + char key[SENSORS_DICT_KEY_SIZE + 1]; + snprintfz(key, SENSORS_DICT_KEY_SIZE, "i%d_n%d_t%d_u%d_%s", + record_id, sensor_number, sensor_reading_type, sensor_units, sensor_name); - default: - break; - } -} + // find the sensor record + const DICTIONARY_ITEM *item = dictionary_get_and_acquire_item(state->sensors.dict, key); + if(likely(item)) { + // recurring collection -static void netdata_get_sel( - int record_id - , int record_type_class - , int sel_state -) { - (void)record_id; - (void)record_type_class; - (void)sel_state; + if(state->debug) + fprintf(stderr, "%s: reusing sensor record for sensor '%s', id %d, number %d, type %d, state %d, units %d, reading_type %d\n", + program_name, sensor_name, record_id, sensor_number, sensor_type, sensor_state, sensor_units, sensor_reading_type); - netdata_sel_events++; -} + struct sensor *sn = dictionary_acquired_item_value(item); + if(sensor_reading) { + netdata_sensor_set_value(sn, sensor_reading, state); + sn->last_collected_metric_ut = state->sensors.now_ut; + } -// END NETDATA CODE -// ---------------------------------------------------------------------------- + sn->sensor_state = sensor_state; + sn->last_collected_state_ut = state->sensors.now_ut; -static int -_ipmimonitoring_sensors (struct ipmi_monitoring_ipmi_config *ipmi_config) -{ - ipmi_monitoring_ctx_t ctx = NULL; - unsigned int sensor_reading_flags = 0; - int i; - int sensor_count; - int rv = -1; + dictionary_acquired_item_release(state->sensors.dict, item); - if (!(ctx = ipmi_monitoring_ctx_create ())) { - collector_error("ipmi_monitoring_ctx_create()"); - goto cleanup; + return; } - if (sdr_cache_directory) - { - if (ipmi_monitoring_ctx_sdr_cache_directory (ctx, - sdr_cache_directory) < 0) - { - collector_error("ipmi_monitoring_ctx_sdr_cache_directory(): %s\n", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - } + if(state->debug) + fprintf(stderr, "Allocating new sensor data record for sensor '%s', id %d, number %d, type %d, state %d, units %d, reading_type %d\n", + sensor_name, record_id, sensor_number, sensor_type, sensor_state, sensor_units, sensor_reading_type); - /* Must call otherwise only default interpretations ever used */ - if (sensor_config_file) - { - if (ipmi_monitoring_ctx_sensor_config_file (ctx, - sensor_config_file) < 0) - { - collector_error( "ipmi_monitoring_ctx_sensor_config_file(): %s\n", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + // check if it is excluded + bool excluded_metric = excluded_record_ids_check(record_id); + bool excluded_state = excluded_status_record_ids_check(record_id); + + if(excluded_metric) { + if(state->debug) + fprintf(stderr, "Sensor '%s' is excluded by excluded_record_ids_check()\n", sensor_name); } - else - { - if (ipmi_monitoring_ctx_sensor_config_file (ctx, NULL) < 0) - { - collector_error( "ipmi_monitoring_ctx_sensor_config_file(): %s\n", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + + if(excluded_state) { + if(state->debug) + fprintf(stderr, "Sensor '%s' is excluded for status check, by excluded_status_record_ids_check()\n", sensor_name); } - if (reread_sdr_cache) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_REREAD_SDR_CACHE; + struct sensor t = { + .sensor_type = sensor_type, + .sensor_state = sensor_state, + .sensor_units = sensor_units, + .sensor_reading_type = sensor_reading_type, + .sensor_name = strdupz(sensor_name), + .component = netdata_sensor_name_to_component(sensor_name), + .do_state = !excluded_state, + .do_metric = !excluded_metric, + }; - if (ignore_non_interpretable_sensors) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_IGNORE_NON_INTERPRETABLE_SENSORS; + t.type = netdata_ipmi_get_sensor_type_string(t.sensor_type, &t.component); - if (bridge_sensors) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_BRIDGE_SENSORS; + switch(t.sensor_units) { + case IPMI_MONITORING_SENSOR_UNITS_CELSIUS: + t.dimension = "temperature"; + t.context = "ipmi.sensor_temperature_c"; + t.title = "IPMI Sensor Temperature Celsius"; + t.units = "Celsius"; + t.family = "temperatures"; + t.chart_type = "line"; + t.priority = state->sensors.priority + 10; + break; - if (interpret_oem_data) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_INTERPRET_OEM_DATA; + case IPMI_MONITORING_SENSOR_UNITS_FAHRENHEIT: + t.dimension = "temperature"; + t.context = "ipmi.sensor_temperature_f"; + t.title = "IPMI Sensor Temperature Fahrenheit"; + t.units = "Fahrenheit"; + t.family = "temperatures"; + t.chart_type = "line"; + t.priority = state->sensors.priority + 20; + break; - if (shared_sensors) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_SHARED_SENSORS; + case IPMI_MONITORING_SENSOR_UNITS_VOLTS: + t.dimension = "voltage"; + t.context = "ipmi.sensor_voltage"; + t.title = "IPMI Sensor Voltage"; + t.units = "Volts"; + t.family = "voltages"; + t.chart_type = "line"; + t.priority = state->sensors.priority + 30; + break; - if (discrete_reading) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_DISCRETE_READING; + case IPMI_MONITORING_SENSOR_UNITS_AMPS: + t.dimension = "ampere"; + t.context = "ipmi.sensor_ampere"; + t.title = "IPMI Sensor Current"; + t.units = "Amps"; + t.family = "current"; + t.chart_type = "line"; + t.priority = state->sensors.priority + 40; + break; - if (ignore_scanning_disabled) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_IGNORE_SCANNING_DISABLED; + case IPMI_MONITORING_SENSOR_UNITS_RPM: + t.dimension = "rotations"; + t.context = "ipmi.sensor_fan_speed"; + t.title = "IPMI Sensor Fans Speed"; + t.units = "RPM"; + t.family = "fans"; + t.chart_type = "line"; + t.priority = state->sensors.priority + 50; + break; - if (assume_bmc_owner) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_ASSUME_BMC_OWNER; + case IPMI_MONITORING_SENSOR_UNITS_WATTS: + t.dimension = "power"; + t.context = "ipmi.sensor_power"; + t.title = "IPMI Sensor Power"; + t.units = "Watts"; + t.family = "power"; + t.chart_type = "line"; + t.priority = state->sensors.priority + 60; + break; -#ifdef IPMI_MONITORING_SENSOR_READING_FLAGS_ENTITY_SENSOR_NAMES - if (entity_sensor_names) - sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_ENTITY_SENSOR_NAMES; -#endif // IPMI_MONITORING_SENSOR_READING_FLAGS_ENTITY_SENSOR_NAMES + case IPMI_MONITORING_SENSOR_UNITS_PERCENT: + t.dimension = "percentage"; + t.context = "ipmi.sensor_reading_percent"; + t.title = "IPMI Sensor Reading Percentage"; + t.units = "%%"; + t.family = "other"; + t.chart_type = "line"; + t.priority = state->sensors.priority + 70; + break; - if (!record_ids_length && !sensor_types_length) - { - if ((sensor_count = ipmi_monitoring_sensor_readings_by_record_id (ctx, - hostname, - ipmi_config, - sensor_reading_flags, - NULL, - 0, - NULL, - NULL)) < 0) - { - collector_error( "ipmi_monitoring_sensor_readings_by_record_id(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - } - else if (record_ids_length) - { - if ((sensor_count = ipmi_monitoring_sensor_readings_by_record_id (ctx, - hostname, - ipmi_config, - sensor_reading_flags, - record_ids, - record_ids_length, - NULL, - NULL)) < 0) - { - collector_error( "ipmi_monitoring_sensor_readings_by_record_id(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - } - else - { - if ((sensor_count = ipmi_monitoring_sensor_readings_by_sensor_type (ctx, - hostname, - ipmi_config, - sensor_reading_flags, - sensor_types, - sensor_types_length, - NULL, - NULL)) < 0) - { - collector_error( "ipmi_monitoring_sensor_readings_by_sensor_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + default: + t.priority = state->sensors.priority + 80; + t.do_metric = false; + break; } -#ifdef NETDATA_COMMENTED - printf ("%s, %s, %s, %s, %s, %s, %s, %s, %s, %s\n", - "Record ID", - "Sensor Name", - "Sensor Number", - "Sensor Type", - "Sensor State", - "Sensor Reading", - "Sensor Units", - "Sensor Event/Reading Type Code", - "Sensor Event Bitmask", - "Sensor Event String"); -#endif // NETDATA_COMMENTED - - for (i = 0; i < sensor_count; i++, ipmi_monitoring_sensor_iterator_next (ctx)) - { - int record_id, sensor_number, sensor_type, sensor_state, sensor_units, - sensor_reading_type; - -#ifdef NETDATA_COMMENTED - int sensor_bitmask_type, sensor_bitmask, event_reading_type_code; - char **sensor_bitmask_strings = NULL; - const char *sensor_type_str; - const char *sensor_state_str; -#endif // NETDATA_COMMENTED + switch(sensor_reading_type) { + case IPMI_MONITORING_SENSOR_READING_TYPE_DOUBLE: + t.multiplier = 1000; + break; - char *sensor_name = NULL; - void *sensor_reading; + case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER8_BOOL: + case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER32: + t.multiplier = 1; + break; - if ((record_id = ipmi_monitoring_sensor_read_record_id (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_record_id(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + default: + t.do_metric = false; + break; + } - if ((sensor_number = ipmi_monitoring_sensor_read_sensor_number (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_sensor_number(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + if(sensor_reading) { + netdata_sensor_set_value(&t, sensor_reading, state); + t.last_collected_metric_ut = state->sensors.now_ut; + } + t.last_collected_state_ut = state->sensors.now_ut; - if ((sensor_type = ipmi_monitoring_sensor_read_sensor_type (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_sensor_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + dictionary_set(state->sensors.dict, key, &t, sizeof(t)); +} - if (!(sensor_name = ipmi_monitoring_sensor_read_sensor_name (ctx))) - { - collector_error( "ipmi_monitoring_sensor_read_sensor_name(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } +static void netdata_update_ipmi_sel_events_count(struct netdata_ipmi_state *state, uint32_t events) { + state->sel.events = events; +} - if ((sensor_state = ipmi_monitoring_sensor_read_sensor_state (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_sensor_state(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } +int netdata_ipmi_collect_data(struct ipmi_monitoring_ipmi_config *ipmi_config, IPMI_COLLECTION_TYPE type, struct netdata_ipmi_state *state) { + errno = 0; - if ((sensor_units = ipmi_monitoring_sensor_read_sensor_units (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_sensor_units(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + if(type & IPMI_COLLECT_TYPE_SENSORS) { + state->sensors.collected = 0; + state->sensors.now_ut = now_monotonic_usec(); -#ifdef NETDATA_COMMENTED - if ((sensor_bitmask_type = ipmi_monitoring_sensor_read_sensor_bitmask_type (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_sensor_bitmask_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - if ((sensor_bitmask = ipmi_monitoring_sensor_read_sensor_bitmask (ctx)) < 0) - { - collector_error("ipmi_monitoring_sensor_read_sensor_bitmask(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + if (netdata_read_ipmi_sensors(ipmi_config, state) < 0) return -1; + } - /* it's ok for this to be NULL, i.e. sensor_bitmask == - * IPMI_MONITORING_SENSOR_BITMASK_TYPE_UNKNOWN - */ - sensor_bitmask_strings = ipmi_monitoring_sensor_read_sensor_bitmask_strings (ctx); - - - -#endif // NETDATA_COMMENTED + if(type & IPMI_COLLECT_TYPE_SEL) { + state->sel.events = 0; + state->sel.now_ut = now_monotonic_usec(); + if(netdata_get_ipmi_sel_events_count(ipmi_config, state) < 0) return -2; + } - if ((sensor_reading_type = ipmi_monitoring_sensor_read_sensor_reading_type (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_sensor_reading_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + return 0; +} - sensor_reading = ipmi_monitoring_sensor_read_sensor_reading (ctx); +int netdata_ipmi_detect_speed_secs(struct ipmi_monitoring_ipmi_config *ipmi_config, IPMI_COLLECTION_TYPE type, struct netdata_ipmi_state *state) { + int i, checks = SPEED_TEST_ITERATIONS, successful = 0; + usec_t total = 0; -#ifdef NETDATA_COMMENTED - if ((event_reading_type_code = ipmi_monitoring_sensor_read_event_reading_type_code (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sensor_read_event_reading_type_code(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } -#endif // NETDATA_COMMENTED + for(i = 0 ; i < checks ; i++) { + if(unlikely(state->debug)) + fprintf(stderr, "%s: checking %s data collection speed iteration %d of %d\n", + program_name, netdata_collect_type_to_string(type), i + 1, checks); - netdata_get_sensor( - record_id - , sensor_number - , sensor_type - , sensor_state - , sensor_units - , sensor_reading_type - , sensor_name - , sensor_reading - ); + // measure the time a data collection needs + usec_t start = now_realtime_usec(); -#ifdef NETDATA_COMMENTED - if (!strlen (sensor_name)) - sensor_name = "N/A"; - - sensor_type_str = _get_sensor_type_string (sensor_type); - - printf ("%d, %s, %d, %s", - record_id, - sensor_name, - sensor_number, - sensor_type_str); - - if (sensor_state == IPMI_MONITORING_STATE_NOMINAL) - sensor_state_str = "Nominal"; - else if (sensor_state == IPMI_MONITORING_STATE_WARNING) - sensor_state_str = "Warning"; - else if (sensor_state == IPMI_MONITORING_STATE_CRITICAL) - sensor_state_str = "Critical"; - else - sensor_state_str = "N/A"; + if(netdata_ipmi_collect_data(ipmi_config, type, state) < 0) + continue; - printf (", %s", sensor_state_str); + usec_t end = now_realtime_usec(); - if (sensor_reading) - { - const char *sensor_units_str; - - if (sensor_reading_type == IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER8_BOOL) - printf (", %s", - (*((uint8_t *)sensor_reading) ? "true" : "false")); - else if (sensor_reading_type == IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER32) - printf (", %u", - *((uint32_t *)sensor_reading)); - else if (sensor_reading_type == IPMI_MONITORING_SENSOR_READING_TYPE_DOUBLE) - printf (", %.2f", - *((double *)sensor_reading)); - else - printf (", N/A"); - - if (sensor_units == IPMI_MONITORING_SENSOR_UNITS_CELSIUS) - sensor_units_str = "C"; - else if (sensor_units == IPMI_MONITORING_SENSOR_UNITS_FAHRENHEIT) - sensor_units_str = "F"; - else if (sensor_units == IPMI_MONITORING_SENSOR_UNITS_VOLTS) - sensor_units_str = "V"; - else if (sensor_units == IPMI_MONITORING_SENSOR_UNITS_AMPS) - sensor_units_str = "A"; - else if (sensor_units == IPMI_MONITORING_SENSOR_UNITS_RPM) - sensor_units_str = "RPM"; - else if (sensor_units == IPMI_MONITORING_SENSOR_UNITS_WATTS) - sensor_units_str = "W"; - else if (sensor_units == IPMI_MONITORING_SENSOR_UNITS_PERCENT) - sensor_units_str = "%"; - else - sensor_units_str = "N/A"; - - printf (", %s", sensor_units_str); - } - else - printf (", N/A, N/A"); + successful++; - printf (", %Xh", event_reading_type_code); + if(unlikely(state->debug)) + fprintf(stderr, "%s: %s data collection speed was %llu usec\n", + program_name, netdata_collect_type_to_string(type), end - start); - /* It is possible you may want to monitor specific event - * conditions that may occur. If that is the case, you may want - * to check out what specific bitmask type and bitmask events - * occurred. See ipmi_monitoring_bitmasks.h for a list of - * bitmasks and types. - */ + // add it to our total + total += end - start; - if (sensor_bitmask_type != IPMI_MONITORING_SENSOR_BITMASK_TYPE_UNKNOWN) - printf (", %Xh", sensor_bitmask); - else - printf (", N/A"); + // wait the same time + // to avoid flooding the IPMI processor with requests + sleep_usec(end - start); + } - if (sensor_bitmask_type != IPMI_MONITORING_SENSOR_BITMASK_TYPE_UNKNOWN - && sensor_bitmask_strings) - { - unsigned int i = 0; + if(!successful) + return 0; - printf (","); + // so, we assume it needed 2x the time + // we find the average in microseconds + // and we round-up to the closest second - while (sensor_bitmask_strings[i]) - { - printf (" "); + return (int)(( total * 2 / successful / USEC_PER_SEC ) + 1); +} - printf ("'%s'", - sensor_bitmask_strings[i]); +// ---------------------------------------------------------------------------- +// data collection threads - i++; - } - } - else - printf (", N/A"); +struct ipmi_collection_thread { + struct ipmi_monitoring_ipmi_config ipmi_config; + int freq_s; + bool debug; + IPMI_COLLECTION_TYPE type; + SPINLOCK spinlock; + struct netdata_ipmi_state state; +}; - printf ("\n"); -#endif // NETDATA_COMMENTED - } +void *netdata_ipmi_collection_thread(void *ptr) { + struct ipmi_collection_thread *t = ptr; - rv = 0; - cleanup: - if (ctx) - ipmi_monitoring_ctx_destroy (ctx); - return (rv); -} + if(t->debug) fprintf(stderr, "%s: calling initialize_ipmi_config() for %s\n", + program_name, netdata_collect_type_to_string(t->type)); + initialize_ipmi_config(&t->ipmi_config); -static int -_ipmimonitoring_sel (struct ipmi_monitoring_ipmi_config *ipmi_config) -{ - ipmi_monitoring_ctx_t ctx = NULL; - unsigned int sel_flags = 0; - int i; - int sel_count; - int rv = -1; + if(t->debug) fprintf(stderr, "%s: detecting IPMI minimum update frequency for %s...\n", + program_name, netdata_collect_type_to_string(t->type)); - if (!(ctx = ipmi_monitoring_ctx_create ())) - { - collector_error("ipmi_monitoring_ctx_create()"); - goto cleanup; - } - - if (sdr_cache_directory) - { - if (ipmi_monitoring_ctx_sdr_cache_directory (ctx, - sdr_cache_directory) < 0) - { - collector_error( "ipmi_monitoring_ctx_sdr_cache_directory(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; + int freq_s = netdata_ipmi_detect_speed_secs(&t->ipmi_config, t->type, &t->state); + if(!freq_s) { + if(t->type & IPMI_COLLECT_TYPE_SENSORS) { + t->state.sensors.status = ICS_INIT_FAILED; + t->state.sensors.last_iteration_ut = 0; } - } - /* Must call otherwise only default interpretations ever used */ - if (sel_config_file) - { - if (ipmi_monitoring_ctx_sel_config_file (ctx, - sel_config_file) < 0) - { - collector_error( "ipmi_monitoring_ctx_sel_config_file(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - } - else - { - if (ipmi_monitoring_ctx_sel_config_file (ctx, NULL) < 0) - { - collector_error( "ipmi_monitoring_ctx_sel_config_file(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; + if(t->type & IPMI_COLLECT_TYPE_SEL) { + t->state.sel.status = ICS_INIT_FAILED; + t->state.sel.last_iteration_ut = 0; } - } - if (reread_sdr_cache) - sel_flags |= IPMI_MONITORING_SEL_FLAGS_REREAD_SDR_CACHE; - - if (interpret_oem_data) - sel_flags |= IPMI_MONITORING_SEL_FLAGS_INTERPRET_OEM_DATA; - - if (assume_system_event_record) - sel_flags |= IPMI_MONITORING_SEL_FLAGS_ASSUME_SYSTEM_EVENT_RECORD; - -#ifdef IPMI_MONITORING_SEL_FLAGS_ENTITY_SENSOR_NAMES - if (entity_sensor_names) - sel_flags |= IPMI_MONITORING_SEL_FLAGS_ENTITY_SENSOR_NAMES; -#endif // IPMI_MONITORING_SEL_FLAGS_ENTITY_SENSOR_NAMES - - if (record_ids_length) - { - if ((sel_count = ipmi_monitoring_sel_by_record_id (ctx, - hostname, - ipmi_config, - sel_flags, - record_ids, - record_ids_length, - NULL, - NULL)) < 0) - { - collector_error( "ipmi_monitoring_sel_by_record_id(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - } - else if (sensor_types_length) - { - if ((sel_count = ipmi_monitoring_sel_by_sensor_type (ctx, - hostname, - ipmi_config, - sel_flags, - sensor_types, - sensor_types_length, - NULL, - NULL)) < 0) - { - collector_error( "ipmi_monitoring_sel_by_sensor_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - } - else if (date_begin - || date_end) - { - if ((sel_count = ipmi_monitoring_sel_by_date_range (ctx, - hostname, - ipmi_config, - sel_flags, - date_begin, - date_end, - NULL, - NULL)) < 0) - { - collector_error( "ipmi_monitoring_sel_by_sensor_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + return ptr; } - else - { - if ((sel_count = ipmi_monitoring_sel_by_record_id (ctx, - hostname, - ipmi_config, - sel_flags, - NULL, - 0, - NULL, - NULL)) < 0) - { - collector_error( "ipmi_monitoring_sel_by_record_id(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - } - -#ifdef NETDATA_COMMENTED - printf ("%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s\n", - "Record ID", - "Record Type", - "SEL State", - "Timestamp", - "Sensor Name", - "Sensor Type", - "Event Direction", - "Event Type Code", - "Event Data", - "Event Offset", - "Event Offset String"); -#endif // NETDATA_COMMENTED - - for (i = 0; i < sel_count; i++, ipmi_monitoring_sel_iterator_next (ctx)) - { - int record_id, record_type, sel_state, record_type_class; -#ifdef NETDATA_COMMENTED - int sensor_type, sensor_number, event_direction, - event_offset_type, event_offset, event_type_code, manufacturer_id; - unsigned int timestamp, event_data1, event_data2, event_data3; - char *event_offset_string = NULL; - const char *sensor_type_str; - const char *event_direction_str; - const char *sel_state_str; - char *sensor_name = NULL; - unsigned char oem_data[64]; - int oem_data_len; - unsigned int j; -#endif // NETDATA_COMMENTED - - if ((record_id = ipmi_monitoring_sel_read_record_id (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_record_id(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; + else { + if(t->type & IPMI_COLLECT_TYPE_SENSORS) { + t->state.sensors.status = ICS_RUNNING; } - if ((record_type = ipmi_monitoring_sel_read_record_type (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_record_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; + if(t->type & IPMI_COLLECT_TYPE_SEL) { + t->state.sel.status = ICS_RUNNING; } + } - if ((record_type_class = ipmi_monitoring_sel_read_record_type_class (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_record_type_class(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + t->freq_s = freq_s = MAX(t->freq_s, freq_s); - if ((sel_state = ipmi_monitoring_sel_read_sel_state (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_sel_state(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + if(t->debug) { + fprintf(stderr, "%s: IPMI minimum update frequency of %s was calculated to %d seconds.\n", + program_name, netdata_collect_type_to_string(t->type), t->freq_s); - netdata_get_sel( - record_id - , record_type_class - , sel_state - ); + fprintf(stderr, "%s: starting data collection of %s\n", + program_name, netdata_collect_type_to_string(t->type)); + } -#ifdef NETDATA_COMMENTED - if (sel_state == IPMI_MONITORING_STATE_NOMINAL) - sel_state_str = "Nominal"; - else if (sel_state == IPMI_MONITORING_STATE_WARNING) - sel_state_str = "Warning"; - else if (sel_state == IPMI_MONITORING_STATE_CRITICAL) - sel_state_str = "Critical"; - else - sel_state_str = "N/A"; + size_t iteration = 0, failures = 0; + usec_t step = t->freq_s * USEC_PER_SEC; - printf ("%d, %d, %s", - record_id, - record_type, - sel_state_str); + heartbeat_t hb; + heartbeat_init(&hb); + while(++iteration) { + heartbeat_next(&hb, step); - if (record_type_class == IPMI_MONITORING_SEL_RECORD_TYPE_CLASS_SYSTEM_EVENT_RECORD - || record_type_class == IPMI_MONITORING_SEL_RECORD_TYPE_CLASS_TIMESTAMPED_OEM_RECORD) - { + if(t->debug) + fprintf(stderr, "%s: calling netdata_ipmi_collect_data() for %s\n", + program_name, netdata_collect_type_to_string(t->type)); - if (ipmi_monitoring_sel_read_timestamp (ctx, ×tamp) < 0) - { - collector_error( "ipmi_monitoring_sel_read_timestamp(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + struct netdata_ipmi_state tmp_state = t->state; - /* XXX: This should be converted to a nice date output using - * your favorite timestamp -> string conversion functions. - */ - printf (", %u", timestamp); + if(t->type & IPMI_COLLECT_TYPE_SENSORS) { + tmp_state.sensors.last_iteration_ut = now_monotonic_usec(); + tmp_state.sensors.freq_ut = t->freq_s * USEC_PER_SEC; } - else - printf (", N/A"); - - if (record_type_class == IPMI_MONITORING_SEL_RECORD_TYPE_CLASS_SYSTEM_EVENT_RECORD) - { - /* If you are integrating ipmimonitoring SEL into a monitoring application, - * you may wish to count the number of times a specific error occurred - * and report that to the monitoring application. - * - * In this particular case, you'll probably want to check out - * what sensor type each SEL event is reporting, the - * event offset type, and the specific event offset that occurred. - * - * See ipmi_monitoring_offsets.h for a list of event offsets - * and types. - */ - - if (!(sensor_name = ipmi_monitoring_sel_read_sensor_name (ctx))) - { - collector_error( "ipmi_monitoring_sel_read_sensor_name(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - - if ((sensor_type = ipmi_monitoring_sel_read_sensor_type (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_sensor_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - - if ((sensor_number = ipmi_monitoring_sel_read_sensor_number (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_sensor_number(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - - if ((event_direction = ipmi_monitoring_sel_read_event_direction (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_event_direction(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - - if ((event_type_code = ipmi_monitoring_sel_read_event_type_code (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_event_type_code(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - if (ipmi_monitoring_sel_read_event_data (ctx, - &event_data1, - &event_data2, - &event_data3) < 0) - { - collector_error( "ipmi_monitoring_sel_read_event_data(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + if(t->type & IPMI_COLLECT_TYPE_SEL) { + tmp_state.sel.last_iteration_ut = now_monotonic_usec(); + tmp_state.sel.freq_ut = t->freq_s * USEC_PER_SEC; + } - if ((event_offset_type = ipmi_monitoring_sel_read_event_offset_type (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_event_offset_type(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + if(netdata_ipmi_collect_data(&t->ipmi_config, t->type, &tmp_state) != 0) + failures++; + else + failures = 0; - if ((event_offset = ipmi_monitoring_sel_read_event_offset (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_event_offset(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } + if(failures > 10) { + collector_error("%s() failed to collect %s data for %zu consecutive times, having made %zu iterations.", + __FUNCTION__, netdata_collect_type_to_string(t->type), failures, iteration); - if (!(event_offset_string = ipmi_monitoring_sel_read_event_offset_string (ctx))) - { - collector_error( "ipmi_monitoring_sel_read_event_offset_string(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; + if(t->type & IPMI_COLLECT_TYPE_SENSORS) { + t->state.sensors.status = ICS_FAILED; + t->state.sensors.last_iteration_ut = 0; } - if (!strlen (sensor_name)) - sensor_name = "N/A"; - - sensor_type_str = _get_sensor_type_string (sensor_type); - - if (event_direction == IPMI_MONITORING_SEL_EVENT_DIRECTION_ASSERTION) - event_direction_str = "Assertion"; - else - event_direction_str = "Deassertion"; - - printf (", %s, %s, %d, %s, %Xh, %Xh-%Xh-%Xh", - sensor_name, - sensor_type_str, - sensor_number, - event_direction_str, - event_type_code, - event_data1, - event_data2, - event_data3); - - if (event_offset_type != IPMI_MONITORING_EVENT_OFFSET_TYPE_UNKNOWN) - printf (", %Xh", event_offset); - else - printf (", N/A"); - - if (event_offset_type != IPMI_MONITORING_EVENT_OFFSET_TYPE_UNKNOWN) - printf (", %s", event_offset_string); - else - printf (", N/A"); - } - else if (record_type_class == IPMI_MONITORING_SEL_RECORD_TYPE_CLASS_TIMESTAMPED_OEM_RECORD - || record_type_class == IPMI_MONITORING_SEL_RECORD_TYPE_CLASS_NON_TIMESTAMPED_OEM_RECORD) - { - if (record_type_class == IPMI_MONITORING_SEL_RECORD_TYPE_CLASS_TIMESTAMPED_OEM_RECORD) - { - if ((manufacturer_id = ipmi_monitoring_sel_read_manufacturer_id (ctx)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_manufacturer_id(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; - } - - printf (", Manufacturer ID = %Xh", manufacturer_id); - } - - if ((oem_data_len = ipmi_monitoring_sel_read_oem_data (ctx, oem_data, 1024)) < 0) - { - collector_error( "ipmi_monitoring_sel_read_oem_data(): %s", - ipmi_monitoring_ctx_errormsg (ctx)); - goto cleanup; + if(t->type & IPMI_COLLECT_TYPE_SEL) { + t->state.sel.status = ICS_FAILED; + t->state.sel.last_iteration_ut = 0; } - printf (", OEM Data = "); - - for (j = 0; j < oem_data_len; j++) - printf ("%02Xh ", oem_data[j]); + break; } - else - printf (", N/A, N/A, N/A, N/A, N/A, N/A, N/A"); - printf ("\n"); -#endif // NETDATA_COMMENTED + spinlock_lock(&t->spinlock); + t->state = tmp_state; + spinlock_unlock(&t->spinlock); } - rv = 0; - cleanup: - if (ctx) - ipmi_monitoring_ctx_destroy (ctx); - return (rv); + return ptr; } // ---------------------------------------------------------------------------- -// MAIN PROGRAM FOR NETDATA PLUGIN - -int ipmi_collect_data(struct ipmi_monitoring_ipmi_config *ipmi_config) { - errno = 0; +// sending data to netdata - if (_ipmimonitoring_sensors(ipmi_config) < 0) return -1; +static inline bool is_sensor_updated(usec_t last_collected_ut, usec_t now_ut, usec_t freq) { + return (now_ut - last_collected_ut < freq * 2) ? true : false; +} - if(netdata_do_sel) { - if(_ipmimonitoring_sel(ipmi_config) < 0) return -2; +static size_t send_ipmi_sensor_metrics_to_netdata(struct netdata_ipmi_state *state) { + if(state->sensors.status != ICS_RUNNING) { + if(unlikely(state->debug)) + fprintf(stderr, "%s: %s() sensors state is not RUNNING\n", + program_name, __FUNCTION__ ); + return 0; } - return 0; -} + size_t total_sensors_sent = 0; + int update_every = (int)(state->sensors.freq_ut / USEC_PER_SEC); + struct sensor *sn; -int ipmi_detect_speed_secs(struct ipmi_monitoring_ipmi_config *ipmi_config) { - int i, checks = 10; - unsigned long long total = 0; + // generate the CHART/DIMENSION lines, if we have to + dfe_start_reentrant(state->sensors.dict, sn) { + if(unlikely(!sn->do_metric && !sn->do_state)) + continue; - for(i = 0 ; i < checks ; i++) { - if(debug) fprintf(stderr, "freeipmi.plugin: checking data collection speed iteration %d of %d\n", i+1, checks); + bool did_metric = false, did_state = false; - // measure the time a data collection needs - unsigned long long start = now_realtime_usec(); - if(ipmi_collect_data(ipmi_config) < 0) - fatal("freeipmi.plugin: data collection failed."); + if(likely(sn->do_metric)) { + if(unlikely(!is_sensor_updated(sn->last_collected_metric_ut, state->updates.now_ut, state->sensors.freq_ut))) { + if(unlikely(state->debug)) + fprintf(stderr, "%s: %s() sensor '%s' metric is not UPDATED (last updated %llu, now %llu, freq %llu\n", + program_name, __FUNCTION__, sn->sensor_name, sn->last_collected_metric_ut, state->updates.now_ut, state->sensors.freq_ut); + } + else { + if (unlikely(!sn->metric_chart_sent)) { + sn->metric_chart_sent = true; - unsigned long long end = now_realtime_usec(); + printf("CHART '%s_%s' '' '%s' '%s' '%s' '%s' '%s' %d %d '' '%s' '%s'\n", + sn->context, sn_dfe.name, sn->title, sn->units, sn->family, sn->context, + sn->chart_type, sn->priority + 1, update_every, program_name, "sensors"); - if(debug) fprintf(stderr, "freeipmi.plugin: data collection speed was %llu usec\n", end - start); + printf("CLABEL 'sensor' '%s' 1\n", sn->sensor_name); + printf("CLABEL 'type' '%s' 1\n", sn->type); + printf("CLABEL 'component' '%s' 1\n", sn->component); + printf("CLABEL_COMMIT\n"); - // add it to our total - total += end - start; + printf("DIMENSION '%s' '' absolute 1 %d\n", sn->dimension, sn->multiplier); + } - // wait the same time - // to avoid flooding the IPMI processor with requests - sleep_usec(end - start); - } + printf("BEGIN '%s_%s'\n", sn->context, sn_dfe.name); + + switch (sn->sensor_reading_type) { + case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER32: + printf("SET '%s' = %u\n", sn->dimension, sn->sensor_reading.uint32_value + ); + break; + + case IPMI_MONITORING_SENSOR_READING_TYPE_DOUBLE: + printf("SET '%s' = %lld\n", sn->dimension, + (long long int) (sn->sensor_reading.double_value * sn->multiplier) + ); + break; + + case IPMI_MONITORING_SENSOR_READING_TYPE_UNSIGNED_INTEGER8_BOOL: + printf("SET '%s' = %u\n", sn->dimension, sn->sensor_reading.bool_value + ); + break; + + default: + case IPMI_MONITORING_SENSOR_READING_TYPE_UNKNOWN: + // this should never happen because we also do the same check at netdata_get_sensor() + sn->do_metric = false; + break; + } - // so, we assume it needed 2x the time - // we find the average in microseconds - // and we round-up to the closest second + printf("END\n"); + did_metric = true; + } + } - return (int)(( total * 2 / checks / 1000000 ) + 1); -} + if(likely(sn->do_state)) { + if(unlikely(!is_sensor_updated(sn->last_collected_state_ut, state->updates.now_ut, state->sensors.freq_ut))) { + if (unlikely(state->debug)) + fprintf(stderr, "%s: %s() sensor '%s' state is not UPDATED (last updated %llu, now %llu, freq %llu\n", + program_name, __FUNCTION__, sn->sensor_name, sn->last_collected_state_ut, state->updates.now_ut, state->sensors.freq_ut); + } + else { + if (unlikely(!sn->state_chart_sent)) { + sn->state_chart_sent = true; + + printf("CHART 'ipmi.sensor_state_%s' '' 'IPMI Sensor State' 'state' 'states' 'ipmi.sensor_state' 'line' %d %d '' '%s' '%s'\n", + sn_dfe.name, sn->priority, update_every, program_name, "sensors"); + + printf("CLABEL 'sensor' '%s' 1\n", sn->sensor_name); + printf("CLABEL 'type' '%s' 1\n", sn->type); + printf("CLABEL 'component' '%s' 1\n", sn->component); + printf("CLABEL_COMMIT\n"); + + printf("DIMENSION 'nominal' '' absolute 1 1\n"); + printf("DIMENSION 'warning' '' absolute 1 1\n"); + printf("DIMENSION 'critical' '' absolute 1 1\n"); + printf("DIMENSION 'unknown' '' absolute 1 1\n"); + } -int parse_inband_driver_type (const char *str) -{ - fatal_assert(str); + printf("BEGIN 'ipmi.sensor_state_%s'\n", sn_dfe.name); + printf("SET 'nominal' = %lld\n", sn->sensor_state == IPMI_MONITORING_STATE_NOMINAL ? 1LL : 0LL); + printf("SET 'warning' = %lld\n", sn->sensor_state == IPMI_MONITORING_STATE_WARNING ? 1LL : 0LL); + printf("SET 'critical' = %lld\n", sn->sensor_state == IPMI_MONITORING_STATE_CRITICAL ? 1LL : 0LL); + printf("SET 'unknown' = %lld\n", sn->sensor_state == IPMI_MONITORING_STATE_UNKNOWN ? 1LL : 0LL); + printf("END\n"); + did_state = true; + } + } - if (strcasecmp (str, IPMI_PARSE_DEVICE_KCS_STR) == 0) - return (IPMI_MONITORING_DRIVER_TYPE_KCS); - else if (strcasecmp (str, IPMI_PARSE_DEVICE_SSIF_STR) == 0) - return (IPMI_MONITORING_DRIVER_TYPE_SSIF); - /* support "open" for those that might be used to - * ipmitool. - */ - else if (strcasecmp (str, IPMI_PARSE_DEVICE_OPENIPMI_STR) == 0 - || strcasecmp (str, IPMI_PARSE_DEVICE_OPENIPMI_STR2) == 0) - return (IPMI_MONITORING_DRIVER_TYPE_OPENIPMI); - /* support "bmc" for those that might be used to - * ipmitool. - */ - else if (strcasecmp (str, IPMI_PARSE_DEVICE_SUNBMC_STR) == 0 - || strcasecmp (str, IPMI_PARSE_DEVICE_SUNBMC_STR2) == 0) - return (IPMI_MONITORING_DRIVER_TYPE_SUNBMC); + if(likely(did_metric || did_state)) + total_sensors_sent++; + } + dfe_done(sn); - return (-1); + return total_sensors_sent; } -int parse_outofband_driver_type (const char *str) -{ - fatal_assert(str); +static size_t send_ipmi_sel_metrics_to_netdata(struct netdata_ipmi_state *state) { + static bool sel_chart_generated = false; + + if(likely(state->sel.status == ICS_RUNNING)) { + if(unlikely(!sel_chart_generated)) { + sel_chart_generated = true; + printf("CHART ipmi.events '' 'IPMI Events' 'events' 'events' ipmi.sel area %d %d '' '%s' '%s'\n" + , state->sel.priority + 2 + , (int)(state->sel.freq_ut / USEC_PER_SEC) + , program_name + , "sel" + ); + printf("DIMENSION events '' absolute 1 1\n"); + } - if (strcasecmp (str, IPMI_PARSE_DEVICE_LAN_STR) == 0) - return (IPMI_MONITORING_PROTOCOL_VERSION_1_5); - /* support "lanplus" for those that might be used to ipmitool. - * support typo variants to ease. - */ - else if (strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR) == 0 - || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR2) == 0 - || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR3) == 0 - || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR4) == 0 - || strcasecmp (str, IPMI_PARSE_DEVICE_LAN_2_0_STR5) == 0) - return (IPMI_MONITORING_PROTOCOL_VERSION_2_0); + printf( + "BEGIN ipmi.events\n" + "SET events = %zu\n" + "END\n" + , state->sel.events + ); + } - return (-1); + return state->sel.events; } -int host_is_local(const char *host) -{ - if (host && (!strcmp(host, "localhost") || !strcmp(host, "127.0.0.1") || !strcmp(host, "::1"))) - return (1); - - return (0); -} +// ---------------------------------------------------------------------------- +// main, command line arguments parsing int main (int argc, char **argv) { + bool netdata_do_sel = IPMI_ENABLE_SEL_BY_DEFAULT; + stderror = stderr; clocks_init(); + int update_every = IPMI_SENSORS_MIN_UPDATE_EVERY; // this is the minimum update frequency + int update_every_sel = IPMI_SEL_MIN_UPDATE_EVERY; // this is the minimum update frequency for SEL events + bool debug = false; + // ------------------------------------------------------------------------ // initialization of netdata plugin @@ -1610,40 +1450,85 @@ int main (int argc, char **argv) { error_log_errors_per_period = 100; error_log_throttle_period = 3600; - // ------------------------------------------------------------------------ // parse command line parameters - int i, freq = 0; + int i, freq_s = 0; for(i = 1; i < argc ; i++) { - if(isdigit(*argv[i]) && !freq) { + if(isdigit(*argv[i]) && !freq_s) { int n = str2i(argv[i]); if(n > 0 && n < 86400) { - freq = n; + freq_s = n; continue; } } else if(strcmp("version", argv[i]) == 0 || strcmp("-version", argv[i]) == 0 || strcmp("--version", argv[i]) == 0 || strcmp("-v", argv[i]) == 0 || strcmp("-V", argv[i]) == 0) { - printf("freeipmi.plugin %s\n", VERSION); + printf("%s %s\n", program_name, VERSION); exit(0); } else if(strcmp("debug", argv[i]) == 0) { - debug = 1; + debug = true; continue; } else if(strcmp("sel", argv[i]) == 0) { - netdata_do_sel = 1; + netdata_do_sel = true; continue; } else if(strcmp("no-sel", argv[i]) == 0) { - netdata_do_sel = 0; + netdata_do_sel = false; continue; } + else if(strcmp("reread-sdr-cache", argv[i]) == 0) { + global_sel_flags |= IPMI_MONITORING_SEL_FLAGS_REREAD_SDR_CACHE; + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_REREAD_SDR_CACHE; + remove_reread_sdr_after_first_use = false; + if (debug) fprintf(stderr, "%s: reread-sdr-cache enabled for both sensors and SEL\n", program_name); + } + else if(strcmp("interpret-oem-data", argv[i]) == 0) { + global_sel_flags |= IPMI_MONITORING_SEL_FLAGS_INTERPRET_OEM_DATA; + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_INTERPRET_OEM_DATA; + if (debug) fprintf(stderr, "%s: interpret-oem-data enabled for both sensors and SEL\n", program_name); + } + else if(strcmp("assume-system-event-record", argv[i]) == 0) { + global_sel_flags |= IPMI_MONITORING_SEL_FLAGS_ASSUME_SYSTEM_EVENT_RECORD; + if (debug) fprintf(stderr, "%s: assume-system-event-record enabled\n", program_name); + } + else if(strcmp("ignore-non-interpretable-sensors", argv[i]) == 0) { + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_IGNORE_NON_INTERPRETABLE_SENSORS; + if (debug) fprintf(stderr, "%s: ignore-non-interpretable-sensors enabled\n", program_name); + } + else if(strcmp("bridge-sensors", argv[i]) == 0) { + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_BRIDGE_SENSORS; + if (debug) fprintf(stderr, "%s: bridge-sensors enabled\n", program_name); + } + else if(strcmp("shared-sensors", argv[i]) == 0) { + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_SHARED_SENSORS; + if (debug) fprintf(stderr, "%s: shared-sensors enabled\n", program_name); + } + else if(strcmp("no-discrete-reading", argv[i]) == 0) { + global_sensor_reading_flags &= ~(IPMI_MONITORING_SENSOR_READING_FLAGS_DISCRETE_READING); + if (debug) fprintf(stderr, "%s: discrete-reading disabled\n", program_name); + } + else if(strcmp("ignore-scanning-disabled", argv[i]) == 0) { + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_IGNORE_SCANNING_DISABLED; + if (debug) fprintf(stderr, "%s: ignore-scanning-disabled enabled\n", program_name); + } + else if(strcmp("assume-bmc-owner", argv[i]) == 0) { + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_ASSUME_BMC_OWNER; + if (debug) fprintf(stderr, "%s: assume-bmc-owner enabled\n", program_name); + } +#if defined(IPMI_MONITORING_SEL_FLAGS_ENTITY_SENSOR_NAMES) && defined(IPMI_MONITORING_SENSOR_READING_FLAGS_ENTITY_SENSOR_NAMES) + else if(strcmp("entity-sensor-names", argv[i]) == 0) { + global_sel_flags |= IPMI_MONITORING_SEL_FLAGS_ENTITY_SENSOR_NAMES; + global_sensor_reading_flags |= IPMI_MONITORING_SENSOR_READING_FLAGS_ENTITY_SENSOR_NAMES; + if (debug) fprintf(stderr, "%s: entity-sensor-names enabled for both sensors and SEL\n", program_name); + } +#endif else if(strcmp("-h", argv[i]) == 0 || strcmp("--help", argv[i]) == 0) { fprintf(stderr, "\n" - " netdata freeipmi.plugin %s\n" - " Copyright (C) 2016-2017 Costa Tsaousis <costa@tsaousis.gr>\n" + " netdata %s %s\n" + " Copyright (C) 2023 Netdata Inc.\n" " Released under GNU General Public License v3 or later.\n" " All rights reserved.\n" "\n" @@ -1661,16 +1546,53 @@ int main (int argc, char **argv) { " no-sel enable/disable SEL collection\n" " default: %s\n" "\n" + " reread-sdr-cache re-read SDR cache on every iteration\n" + " default: disabled\n" + "\n" + " interpret-oem-data attempt to parse OEM data\n" + " default: disabled\n" + "\n" + " assume-system-event-record \n" + " tread illegal SEL events records as normal\n" + " default: disabled\n" + "\n" + " ignore-non-interpretable-sensors \n" + " do not read sensors that cannot be interpreted\n" + " default: disabled\n" + "\n" + " bridge-sensors bridge sensors not owned by the BMC\n" + " default: disabled\n" + "\n" + " shared-sensors enable shared sensors, if found\n" + " default: disabled\n" + "\n" + " no-discrete-reading do not read sensors that their event/reading type code is invalid\n" + " default: enabled\n" + "\n" + " ignore-scanning-disabled \n" + " Ignore the scanning bit and read sensors no matter what\n" + " default: disabled\n" + "\n" + " assume-bmc-owner assume the BMC is the sensor owner no matter what\n" + " (usually bridging is required too)\n" + " default: disabled\n" + "\n" +#if defined(IPMI_MONITORING_SEL_FLAGS_ENTITY_SENSOR_NAMES) && defined(IPMI_MONITORING_SENSOR_READING_FLAGS_ENTITY_SENSOR_NAMES) + " entity-sensor-names sensor names prefixed with entity id and instance\n" + " default: disabled\n" + "\n" +#endif " hostname HOST\n" " username USER\n" " password PASS connect to remote IPMI host\n" " default: local IPMI processor\n" "\n" + " no-auth-code-check\n" " noauthcodecheck don't check the authentication codes returned\n" "\n" " driver-type IPMIDRIVER\n" " Specify the driver type to use instead of doing an auto selection. \n" - " The currently available outofband drivers are LAN and LAN_2_0,\n" + " The currently available outofband drivers are LAN and LAN_2_0,\n" " which perform IPMI 1.5 and IPMI 2.0 respectively. \n" " The currently available inband drivers are KCS, SSIF, OPENIPMI and SUNBMC.\n" "\n" @@ -1680,6 +1602,9 @@ int main (int argc, char **argv) { " sensor-config-file FILE filename to read sensor configuration\n" " default: %s\n" "\n" + " sel-config-file FILE filename to read sel configuration\n" + " default: %s\n" + "\n" " ignore N1,N2,N3,... sensor IDs to ignore\n" " default: none\n" "\n" @@ -1700,11 +1625,12 @@ int main (int argc, char **argv) { " For more information:\n" " https://github.com/netdata/netdata/tree/master/collectors/freeipmi.plugin\n" "\n" - , VERSION - , netdata_update_every + , program_name, VERSION + , update_every , netdata_do_sel?"enabled":"disabled" , sdr_cache_directory?sdr_cache_directory:"system default" , sensor_config_file?sensor_config_file:"system default" + , sel_config_file?sel_config_file:"system default" ); exit(1); } @@ -1713,7 +1639,7 @@ int main (int argc, char **argv) { char *s = argv[i]; // mask it be hidden from the process tree while(*s) *s++ = 'x'; - if(debug) fprintf(stderr, "freeipmi.plugin: hostname set to '%s'\n", hostname); + if(debug) fprintf(stderr, "%s: hostname set to '%s'\n", program_name, hostname); continue; } else if(i < argc && strcmp("username", argv[i]) == 0) { @@ -1721,7 +1647,7 @@ int main (int argc, char **argv) { char *s = argv[i]; // mask it be hidden from the process tree while(*s) *s++ = 'x'; - if(debug) fprintf(stderr, "freeipmi.plugin: username set to '%s'\n", username); + if(debug) fprintf(stderr, "%s: username set to '%s'\n", program_name, username); continue; } else if(i < argc && strcmp("password", argv[i]) == 0) { @@ -1729,149 +1655,247 @@ int main (int argc, char **argv) { char *s = argv[i]; // mask it be hidden from the process tree while(*s) *s++ = 'x'; - if(debug) fprintf(stderr, "freeipmi.plugin: password set to '%s'\n", password); + if(debug) fprintf(stderr, "%s: password set to '%s'\n", program_name, password); continue; } else if(strcmp("driver-type", argv[i]) == 0) { if (hostname) { - protocol_version=parse_outofband_driver_type(argv[++i]); - if(debug) fprintf(stderr, "freeipmi.plugin: outband protocol version set to '%d'\n", protocol_version); + protocol_version = netdata_parse_outofband_driver_type(argv[++i]); + if(debug) fprintf(stderr, "%s: outband protocol version set to '%d'\n", + program_name, protocol_version); } else { - driver_type=parse_inband_driver_type(argv[++i]); - if(debug) fprintf(stderr, "freeipmi.plugin: inband driver type set to '%d'\n", driver_type); + driver_type = netdata_parse_inband_driver_type(argv[++i]); + if(debug) fprintf(stderr, "%s: inband driver type set to '%d'\n", + program_name, driver_type); } continue; - } else if (i < argc && strcmp("noauthcodecheck", argv[i]) == 0) { - if (!hostname || host_is_local(hostname)) { + } else if (i < argc && (strcmp("noauthcodecheck", argv[i]) == 0 || strcmp("no-auth-code-check", argv[i]) == 0)) { + if (!hostname || netdata_host_is_localhost(hostname)) { if (debug) - fprintf( - stderr, - "freeipmi.plugin: noauthcodecheck workaround flag is ignored for inband configuration\n"); - } else if (protocol_version < 0 || protocol_version == IPMI_MONITORING_PROTOCOL_VERSION_1_5) { + fprintf(stderr, "%s: noauthcodecheck workaround flag is ignored for inband configuration\n", + program_name); + + } + else if (protocol_version < 0 || protocol_version == IPMI_MONITORING_PROTOCOL_VERSION_1_5) { workaround_flags |= IPMI_MONITORING_WORKAROUND_FLAGS_PROTOCOL_VERSION_1_5_NO_AUTH_CODE_CHECK; + if (debug) - fprintf(stderr, "freeipmi.plugin: noauthcodecheck workaround flag enabled\n"); - } else { + fprintf(stderr, "%s: noauthcodecheck workaround flag enabled\n", program_name); + } + else { if (debug) - fprintf( - stderr, - "freeipmi.plugin: noauthcodecheck workaround flag is ignored for protocol version 2.0\n"); + fprintf(stderr, "%s: noauthcodecheck workaround flag is ignored for protocol version 2.0\n", + program_name); } continue; } else if(i < argc && strcmp("sdr-cache-dir", argv[i]) == 0) { sdr_cache_directory = argv[++i]; - if(debug) fprintf(stderr, "freeipmi.plugin: SDR cache directory set to '%s'\n", sdr_cache_directory); + + if(debug) + fprintf(stderr, "%s: SDR cache directory set to '%s'\n", program_name, sdr_cache_directory); + continue; } else if(i < argc && strcmp("sensor-config-file", argv[i]) == 0) { sensor_config_file = argv[++i]; - if(debug) fprintf(stderr, "freeipmi.plugin: sensor config file set to '%s'\n", sensor_config_file); + if(debug) fprintf(stderr, "%s: sensor config file set to '%s'\n", program_name, sensor_config_file); + continue; + } + else if(i < argc && strcmp("sel-config-file", argv[i]) == 0) { + sel_config_file = argv[++i]; + if(debug) fprintf(stderr, "%s: sel config file set to '%s'\n", program_name, sel_config_file); continue; } else if(i < argc && strcmp("ignore", argv[i]) == 0) { - excluded_record_ids_parse(argv[++i]); + excluded_record_ids_parse(argv[++i], debug); continue; } else if(i < argc && strcmp("ignore-status", argv[i]) == 0) { - excluded_status_record_ids_parse(argv[++i]); + excluded_status_record_ids_parse(argv[++i], debug); continue; } - collector_error("freeipmi.plugin: ignoring parameter '%s'", argv[i]); + collector_error("%s(): ignoring parameter '%s'", __FUNCTION__, argv[i]); } errno = 0; - if(freq >= netdata_update_every) - netdata_update_every = freq; - - else if(freq) - collector_error("update frequency %d seconds is too small for IPMI. Using %d.", freq, netdata_update_every); + if(freq_s && freq_s < update_every) + collector_error("%s(): update frequency %d seconds is too small for IPMI. Using %d.", + __FUNCTION__, freq_s, update_every); + update_every = freq_s = MAX(freq_s, update_every); + update_every_sel = MAX(update_every, update_every_sel); // ------------------------------------------------------------------------ // initialize IPMI - struct ipmi_monitoring_ipmi_config ipmi_config; - - if(debug) fprintf(stderr, "freeipmi.plugin: calling _init_ipmi_config()\n"); - - _init_ipmi_config(&ipmi_config); - if(debug) { - fprintf(stderr, "freeipmi.plugin: calling ipmi_monitoring_init()\n"); - ipmimonitoring_init_flags|=IPMI_MONITORING_FLAGS_DEBUG|IPMI_MONITORING_FLAGS_DEBUG_IPMI_PACKETS; + fprintf(stderr, "%s: calling ipmi_monitoring_init()\n", program_name); + ipmimonitoring_init_flags |= IPMI_MONITORING_FLAGS_DEBUG|IPMI_MONITORING_FLAGS_DEBUG_IPMI_PACKETS; } - if(ipmi_monitoring_init(ipmimonitoring_init_flags, &errnum) < 0) - fatal("ipmi_monitoring_init: %s", ipmi_monitoring_ctx_strerror(errnum)); - - if(debug) fprintf(stderr, "freeipmi.plugin: detecting IPMI minimum update frequency...\n"); - freq = ipmi_detect_speed_secs(&ipmi_config); - if(debug) fprintf(stderr, "freeipmi.plugin: IPMI minimum update frequency was calculated to %d seconds.\n", freq); - - if(freq > netdata_update_every) { - collector_info("enforcing minimum data collection frequency, calculated to %d seconds.", freq); - netdata_update_every = freq; - } + int rc; + if(ipmi_monitoring_init(ipmimonitoring_init_flags, &rc) < 0) + fatal("ipmi_monitoring_init: %s", ipmi_monitoring_ctx_strerror(rc)); + // ------------------------------------------------------------------------ + // create the data collection threads + + struct ipmi_collection_thread sensors_data = { + .type = IPMI_COLLECT_TYPE_SENSORS, + .freq_s = update_every, + .spinlock = NETDATA_SPINLOCK_INITIALIZER, + .debug = debug, + .state = { + .debug = debug, + .sensors = { + .status = ICS_INIT, + .last_iteration_ut = now_monotonic_usec(), + .freq_ut = update_every * USEC_PER_SEC, + .priority = IPMI_SENSORS_DASHBOARD_PRIORITY, + .dict = dictionary_create_advanced(DICT_OPTION_DONT_OVERWRITE_VALUE|DICT_OPTION_FIXED_SIZE, NULL, sizeof(struct sensor)), + }, + }, + }, sel_data = { + .type = IPMI_COLLECT_TYPE_SEL, + .freq_s = update_every_sel, + .spinlock = NETDATA_SPINLOCK_INITIALIZER, + .debug = debug, + .state = { + .debug = debug, + .sel = { + .status = ICS_INIT, + .last_iteration_ut = now_monotonic_usec(), + .freq_ut = update_every_sel * USEC_PER_SEC, + .priority = IPMI_SEL_DASHBOARD_PRIORITY, + }, + }, + }; + + netdata_thread_t sensors_thread = 0, sel_thread = 0; + + netdata_thread_create(&sensors_thread, "IPMI[sensors]", NETDATA_THREAD_OPTION_DONT_LOG, netdata_ipmi_collection_thread, &sensors_data); + + if(netdata_do_sel) + netdata_thread_create(&sel_thread, "IPMI[sel]", NETDATA_THREAD_OPTION_DONT_LOG, netdata_ipmi_collection_thread, &sel_data); // ------------------------------------------------------------------------ // the main loop - if(debug) fprintf(stderr, "freeipmi.plugin: starting data collection\n"); + if(debug) fprintf(stderr, "%s: starting data collection\n", program_name); time_t started_t = now_monotonic_sec(); size_t iteration = 0; - usec_t step = netdata_update_every * USEC_PER_SEC; + usec_t step = 100 * USEC_PER_MS; + bool global_chart_created = false; + bool tty = isatty(fileno(stderr)) == 1; heartbeat_t hb; heartbeat_init(&hb); for(iteration = 0; 1 ; iteration++) { usec_t dt = heartbeat_next(&hb, step); - if (iteration) { - if (iteration == 1) { - fprintf( - stdout, - "CHART netdata.freeipmi_availability_status '' 'Plugin availability status' 'status' plugins netdata.plugin_availability_status line 146000 %d\n" - "DIMENSION available '' absolute 1 1\n", - netdata_update_every); + if(!tty) + fprintf(stdout, "\n"); // keepalive to avoid parser read timeout (2 minutes) during ipmi_detect_speed_secs() + + struct netdata_ipmi_state state = {0 }; + + spinlock_lock(&sensors_data.spinlock); + state.sensors = sensors_data.state.sensors; + spinlock_unlock(&sensors_data.spinlock); + + spinlock_lock(&sel_data.spinlock); + state.sel = sel_data.state.sel; + spinlock_unlock(&sel_data.spinlock); + + switch(state.sensors.status) { + case ICS_RUNNING: + step = update_every * USEC_PER_SEC; + if(state.sensors.last_iteration_ut < now_monotonic_usec() - IPMI_RESTART_IF_SENSORS_DONT_ITERATE_EVERY_SECONDS * USEC_PER_SEC) { + collector_error("%s(): sensors have not be collected for %zu seconds. Exiting to restart.", + __FUNCTION__, (size_t)((now_monotonic_usec() - state.sensors.last_iteration_ut) / USEC_PER_SEC)); + + fprintf(stdout, "EXIT\n"); + fflush(stdout); + exit(0); + } + break; + + case ICS_INIT: + continue; + + case ICS_INIT_FAILED: + collector_error("%s(): sensors failed to initialize. Calling DISABLE.", __FUNCTION__); + fprintf(stdout, "DISABLE\n"); + fflush(stdout); + exit(0); + + case ICS_FAILED: + collector_error("%s(): sensors fails repeatedly to collect metrics. Exiting to restart.", __FUNCTION__); + fprintf(stdout, "EXIT\n"); + fflush(stdout); + exit(0); + } + + if(netdata_do_sel) { + switch (state.sensors.status) { + case ICS_RUNNING: + case ICS_INIT: + break; + + case ICS_INIT_FAILED: + case ICS_FAILED: + collector_error("%s(): SEL fails to collect events. Disabling SEL collection.", __FUNCTION__); + netdata_do_sel = false; + break; } - fprintf( - stdout, - "BEGIN netdata.freeipmi_availability_status\n" - "SET available = 1\n" - "END\n"); } - if(debug && iteration) - fprintf(stderr, "freeipmi.plugin: iteration %zu, dt %llu usec, sensors collected %zu, sensors sent to netdata %zu \n" + if(unlikely(debug)) + fprintf(stderr, "%s: calling send_ipmi_sensor_metrics_to_netdata()\n", program_name); + + state.updates.now_ut = now_monotonic_usec(); + send_ipmi_sensor_metrics_to_netdata(&state); + + if(netdata_do_sel) + send_ipmi_sel_metrics_to_netdata(&state); + + if(unlikely(debug)) + fprintf(stderr, "%s: iteration %zu, dt %llu usec, sensors ever collected %zu, sensors last collected %zu \n" + , program_name , iteration , dt - , netdata_sensors_collected - , netdata_sensors_updated + , dictionary_entries(state.sensors.dict) + , state.sensors.collected ); - netdata_mark_as_not_updated(); + if (!global_chart_created) { + global_chart_created = true; - if(debug) fprintf(stderr, "freeipmi.plugin: calling ipmi_collect_data()\n"); - if(ipmi_collect_data(&ipmi_config) < 0) - fatal("data collection failed."); + fprintf(stdout, + "CHART netdata.freeipmi_availability_status '' 'Plugin availability status' 'status' " + "plugins netdata.plugin_availability_status line 146000 %d '' '%s' '%s'\n" + "DIMENSION available '' absolute 1 1\n", + update_every, program_name, ""); + } - if(debug) fprintf(stderr, "freeipmi.plugin: calling send_metrics_to_netdata()\n"); - send_metrics_to_netdata(); - fflush(stdout); + fprintf(stdout, + "BEGIN netdata.freeipmi_availability_status\n" + "SET available = 1\n" + "END\n"); // restart check (14400 seconds) - if (now_monotonic_sec() - started_t > 14400) { + if (now_monotonic_sec() - started_t > IPMI_RESTART_EVERY_SECONDS) { + collector_error("%s(): reached my lifetime expectancy. Exiting to restart.", __FUNCTION__); fprintf(stdout, "EXIT\n"); fflush(stdout); exit(0); } + + fflush(stdout); } } - diff --git a/collectors/freeipmi.plugin/metrics.csv b/collectors/freeipmi.plugin/metrics.csv index 9d493a531..4c90d5c1b 100644 --- a/collectors/freeipmi.plugin/metrics.csv +++ b/collectors/freeipmi.plugin/metrics.csv @@ -1,10 +1,10 @@ metric,scope,dimensions,unit,description,chart_type,labels,plugin,module -ipmi.sel,,events,events,"IPMI Events",area,,freeipmi.plugin, -ipmi.sensors_states,,"nominal, critical, warning",sensors,"IPMI Sensors State",line,,freeipmi.plugin, -ipmi.temperatures_c,,a dimension per sensor,Celsius,"System Celsius Temperatures read by IPMI",line,,freeipmi.plugin, -ipmi.temperatures_f,,a dimension per sensor,Fahrenheit,"System Celsius Temperatures read by IPMI",line,,freeipmi.plugin, -ipmi.voltages,,a dimension per sensor,Volts,"System Voltages read by IPMI",line,,freeipmi.plugin, -ipmi.amps,,a dimension per sensor,Amps,"System Current read by IPMI",line,,freeipmi.plugin, -ipmi.rpm,,a dimension per sensor,RPM,"System Fans read by IPMI",line,,freeipmi.plugin, -ipmi.watts,,a dimension per sensor,Watts,"System Power read by IPMI",line,,freeipmi.plugin, -ipmi.percent,,a dimension per sensor,%,"System Metrics read by IPMI",line,,freeipmi.plugin,
\ No newline at end of file +ipmi.sel,,events,events,"IPMI Events",area,,freeipmi.plugin,sel +ipmi.sensor_state,sensor,"nominal, critical, warning, unknown",state,"IPMI Sensors State",line,"sensor, type, component",freeipmi.plugin,sensors +ipmi.sensor_temperature_c,sensor,temperature,Celsius,"IPMI Sensor Temperature Celsius",line,"sensor, type, component",freeipmi.plugin,sensors +ipmi.sensor_temperature_f,sensor,temperature,Fahrenheit,"IPMI Sensor Temperature Fahrenheit",line,"sensor, type, component",freeipmi.plugin,sensors +ipmi.sensor_voltage,sensor,voltage,Volts,"IPMI Sensor Voltage",line,"sensor, type, component",freeipmi.plugin,sensors +ipmi.sensor_ampere,sensor,ampere,Amps,"IPMI Sensor Current",line,"sensor, type, component",freeipmi.plugin,sensors +ipmi.sensor_fan_speed,sensor,rotations,RPM,"IPMI Sensor Fans Speed",line,"sensor, type, component",freeipmi.plugin,sensors +ipmi.sensor_power,sensor,power,Watts,"IPMI Sensor Power",line,"sensor, type, component",freeipmi.plugin,sensors +ipmi.sensor_reading_percent,sensor,percentage,%,"IPMI Sensor Reading Percentage",line,"sensor, type, component",freeipmi.plugin,sensors diff --git a/collectors/freeipmi.plugin/multi_metadata.yaml b/collectors/freeipmi.plugin/multi_metadata.yaml new file mode 100644 index 000000000..21333278a --- /dev/null +++ b/collectors/freeipmi.plugin/multi_metadata.yaml @@ -0,0 +1,199 @@ +name: freeipmi.plugin +modules: + - meta: + plugin_name: freeipmi.plugin + module_name: sel + monitored_instance: + name: freeipmi sel + link: '' + categories: [] + icon_filename: '' + related_resources: + integrations: + list: [] + info_provided_to_referring_integrations: + description: '' + keywords: [] + most_popular: false + overview: + data_collection: + metrics_description: '' + method_description: '' + supported_platforms: + include: [] + exclude: [] + multi-instance: true + additional_permissions: + description: '' + default_behavior: + auto_detection: + description: '' + limits: + description: '' + performance_impact: + description: '' + setup: + prerequisites: + list: [] + configuration: + file: + name: '' + description: '' + options: + description: '' + folding: + title: '' + enabled: true + list: [] + examples: + folding: + enabled: true + title: '' + list: [] + troubleshooting: + problems: + list: [] + alerts: [] + metrics: + folding: + title: Metrics + enabled: false + description: "" + availability: [] + scopes: + - name: global + description: "" + labels: [] + metrics: + - name: ipmi.sel + description: IPMI Events + unit: "events" + chart_type: area + dimensions: + - name: events + - meta: + plugin_name: freeipmi.plugin + module_name: sensors + monitored_instance: + name: freeipmi sensors + link: '' + categories: [] + icon_filename: '' + related_resources: + integrations: + list: [] + info_provided_to_referring_integrations: + description: '' + keywords: [] + most_popular: false + overview: + data_collection: + metrics_description: '' + method_description: '' + supported_platforms: + include: [] + exclude: [] + multi-instance: true + additional_permissions: + description: '' + default_behavior: + auto_detection: + description: '' + limits: + description: '' + performance_impact: + description: '' + setup: + prerequisites: + list: [] + configuration: + file: + name: '' + description: '' + options: + description: '' + folding: + title: '' + enabled: true + list: [] + examples: + folding: + enabled: true + title: '' + list: [] + troubleshooting: + problems: + list: [] + alerts: + - name: ipmi_sensor_state + link: https://github.com/netdata/netdata/blob/master/health/health.d/ipmi.conf + metric: ipmi.sensor_state + info: IPMI sensor ${label:sensor} (${label:component}) state + metrics: + folding: + title: Metrics + enabled: false + description: "" + availability: [] + scopes: + - name: sensor + description: "" + labels: + - name: sensor + description: TBD + - name: type + description: TBD + - name: component + description: TBD + metrics: + - name: ipmi.sensor_state + description: IPMI Sensors State + unit: "state" + chart_type: line + dimensions: + - name: nominal + - name: critical + - name: warning + - name: unknown + - name: ipmi.sensor_temperature_c + description: IPMI Sensor Temperature Celsius + unit: "Celsius" + chart_type: line + dimensions: + - name: temperature + - name: ipmi.sensor_temperature_f + description: IPMI Sensor Temperature Fahrenheit + unit: "Fahrenheit" + chart_type: line + dimensions: + - name: temperature + - name: ipmi.sensor_voltage + description: IPMI Sensor Voltage + unit: "Volts" + chart_type: line + dimensions: + - name: voltage + - name: ipmi.sensor_ampere + description: IPMI Sensor Current + unit: "Amps" + chart_type: line + dimensions: + - name: ampere + - name: ipmi.sensor_fan_speed + description: IPMI Sensor Fans Speed + unit: "RPM" + chart_type: line + dimensions: + - name: rotations + - name: ipmi.sensor_power + description: IPMI Sensor Power + unit: "Watts" + chart_type: line + dimensions: + - name: power + - name: ipmi.sensor_reading_percent + description: IPMI Sensor Reading Percentage + unit: "%" + chart_type: line + dimensions: + - name: percentage |