Hardware PluginsΒΆ
Hardware plugins are similar to status plugins but are called less frequently since they are assumed to be more static kinds of information, e.g. the motherboard serial number or CPU vendor information. They are stored as a sub-field of the node's information and can be viewed with:
scyld-nodectl --all ls -L
As with status plugins, scripts-available/hardware
is populated with Penguin-provided scripts covering a range of
hardware options.
Default frequency is data collection every 300 seconds but this
may be overridden on a node-by-node basis with attribute _hardware_secs
.
For larger-scale management and control, one can set the
_hardware_plugins
and/or _hardware_secs
attributes
inside an attribute group and then join nodes to that group.
Building Hardware Plugins into an Image
As with status plugins, admins can sym-link or add scripts to the
scripts-enabled/hardware
directory inside a disk image.
On-The-Fly Plugins
The _hardware_plugins
attribute is used for enabling/disabling
hardware plugins on-the-fly.
If _hardware_plugins=infiniband
, then the system will look for the
script in scripts-available/hardware/infiniband.sh
.
Available Hardware Plugins
corehardware
provides basic hardware information about the server: total RAM, CPU count, CPU architecture, CPU model, BIOS/UEFI mode, boot style, vendor and product information, BIOS version/date/vendor
Note that
ram_total
is reported in KiB, not bytes; soram_total=1000
indicates that 1024000 bytes of total memory is available
infiniband
provides basic information about any Infiniband network devices found on the server
storage
provides basic information about any storage devices found, including NVME
nvidia
provides basic information about an NVIDIA GPUs or accelerators
Note
Changes to _hardware_plugins
will likely be detected on
the next status cycle, usually every 10 seconds, but will not be
processed until the next hardware-update cycle, usually every
300 seconds unless changed by the admin.