mcelog plugin¶


Plugin name	`mcelog`
Type	read
Callbacks	config, init, read, shutdown
Status	supported
Since	5.8
Copyright	2016–2017 Intel Corporation
Configuration	collectd.conf(5)

The purpose of mcelog plugin is to send notifications and stats relevant to Machine Check Exceptions (MCE) when they occur. The plugin leverages the mcelog Linux utility to detect that an exception has occurred. mcelog supports a client server model and does the logging and accounting of exceptions when they occur. The plugin simply leverages the client protocol of mcelog to detect when an exception has occurred. The goal of this equivalence feature is to expose Reliability, Availability and Serviceability (RAS) features metrics and events provided by the platform to higher level fault management applications. The plugin does the following:

Checks mcelog server liveliness, reports a failure if it’s not running or if it fails.
Retrieve aggregated Memory Corrected and Uncorrected Errors from the client protocol (Submit event/stat).

Mcelog must be configured to run on the platform in daemon mode and logging capabilities must be enabled. For a full description of available options please refer to the collectd.conf(5) manual page.

Synopsis¶

  <Plugin mcelog>
    <Memory>
      McelogClientSocket "/var/run/mcelog-client"
      PersistentNotification false
    </Memory>
  </Plugin>

Will be changed after branch "feat_mcelog_mem_notification_level" is merged (default if all commented for now is socket):

 # <Plugin mcelog>
 #   <Memory>
 #     McelogClientSocket "/var/run/mcelog-client"
 #     PersistentNotification false
 #   </Memory>
 #   McelogLogfile "/var/log/mcelog"
 # </Plugin>

Parameters¶

None yet

Metrics¶

Metric/Feature Name	Date Type	Format Example	Internal Collectd Version	Description	Comments
Memory corrected errors	Int	51522	None	Number of Corrected memory errors since the system boot	gets metrics from mcelog daemon.
Memory corrected errors in 24 Hours	Int	51522	None	Number of Corrected memory errors since previous 24 hours	gets metrics from mcelog daemon.
Memory Uncorrected errors	Int	51522	None	Number of Corrected memory errors since the system boot	gets metrics from mcelog daemon.
Memory Uncorrected errors in 24 Hours	Int	51522	None	Number of Corrected memory errors since previous 24 hours	gets metrics from mcelog daemon.
Socket	Int	0	None	Socker number error occurred on	gets metrics from mcelog daemon.
Channel	Char	0	None	Memory channel each channel represents a DIMM module	gets metrics from mcelog daemon.
Memory DIMM	Char	B1	None	Memory DIMM corresponding the memory used by the cores errors occurred on	gets metrics from mcelog daemon.
Memory Slot	Char	1	None	Memory slot corresponding the memory used by the cores errors occurred on	gets metrics from mcelog daemon.
CPU ID	Int	0	Future	CPU ID of the cores errors occurred on. Will be added to new EDAC plugin
Memory Page	Hex	0x12345	Future	Memory page corresponding the memory used by the cores errors occurred on. Will be added to new EDAC plugin	Not part of Collectd. Currently available with kernel EDAC logs
Memory Offset	Hex	0x0	Future	Memory offset in the page. Will be added to new EDAC plugin	Not part of Collectd. Currently available with kernel EDAC logs
Memory Row	Hex	0x12345			Not part of Collectd. Currently available with kernel EDAC logs
Memory Grain	Int	8	Future	The byte granularity or the error grain. Will be added to new EDAC plugin	Not part of Collectd. Currently available with kernel EDAC logs
Error Syndrome	Hex	0x6ce3	Future	Memory syndrome corresponding the memory used by the cores errors occurred on. Will be added to new EDAC plugin	Not part of Collectd. Currently available with kernel EDAC logs
Error Type	Text		Future	Error type. Will be added to new EDAC plugin	Not part of Collectd. Currently available with kernel EDAC logs
Error code	Integer	0101:0090	Future	Error code put out by EDAC. Will be added to new EDAC plugin	Not part of Collectd. Currently available with kernel EDAC logs
Logging	Log path			Configurable logging path	Not part of Collectd. Currently available with kernel EDAC logs
dimmX or rankX directory info	Varying		Future	Expose interface files provided by sysfs through mcX/dimmX or rankX directories	Not part of Collectd. Currently available with kernel EDAC logs
csrowX directory info	Varying		Future	Expose interface files provided by sysfs through mcX/csrowX directories	Not part of Collectd. Currently available with kernel EDAC logs
RAS interrupts	Count on each core	[CoreID]:[InterruptCont]	Future	Expose the RAS related interrupts on cores of interest via Collectd	Discussion open to see if this info can be exposed through the plugin.

Example Graph¶

None yet. Add one now!

Dependencies¶

mcelog

Also See¶

RAS/mcelog Plugin High Level Design Tests Executed

Configuration¶

Configuration options · from collectd.conf(5)

See this section in the full configuration reference

The mcelog plugin uses mcelog to retrieve machine check exceptions.

By default the plugin connects to "/var/run/mcelog-client" to check if the mcelog server is running. When the server is running, the plugin will tail the specified logfile to retrieve machine check exception information and send a notification with the details from the logfile. The plugin will use the mcelog client protocol to retrieve memory related machine check exceptions. Note that for memory exceptions, notifications are only sent when there is a change in the number of corrected/uncorrected memory errors.

The Memory block¶

Note: these options cannot be used in conjunction with the logfile options, they are mutually exclusive.

McelogClientSocket Path Connect to the mcelog client socket using the UNIX domain socket at Path. Defaults to "/var/run/mcelog-client".
PersistentNotification true|false Override default configuration to only send notifications when sent when there is a change in the number of corrected/uncorrected memory errors. When set to true notifications will be sent for every read cycle. Default is false. Does not affect the stats being dispatched.
McelogLogfile Path: The mcelog file to parse. Defaults to "/var/log/mcelog". Note: this option cannot be used in conjunction with the memory block options, they are mutually exclusive.