When cAdvisor starts up, it would read the `vendor` files in
`/sys/bus/pci/devices/*` to see if any NVIDIA devices (vendor ID: 0x10de) are
attached to the node. If no NVIDIA devices are found, this code path would
become dormant for the rest of cAdvisor lifetime. If NVIDIA devices are found,
we would start a goroutine that would check for the presence of NVML by trying
to dynamically load it at regular intervals. We need to do this regular
checking instead of doing it just once because it may happen that cAdvisor is
started before the NVIDIA drivers and NVML are installed. Once the NVML
dynamic loading succeeds, we would use NVML’s query methods to find out how
many devices exist on the node and create a map from their minor numbers to
their handles and cache that map. The goroutine would exit at this point.
If we detected the presence of NVML in the previous step, whenever a new
container is detected by cAdvisor, cAdvisor would read the `devices.list` file
from the container's devices cgroup. The `devices.list` file lists the
major:minor number of all the devices that the container is allowed to access.
If we find any device with major number 195 (which is the major number assigned
to NVIDIA devices), we would cache the list of corresponding minor numbers for
that container.
During every housekeeping operation, in addition to collecting all the existing
metrics, we will use the cached NVIDIA device minor numbers and the map from
minor numbers to device handles to get metrics for GPU devices attached to the
container.
This ensures each goroutine is given its own Netlink connection, and
presumably avoids having a message destined for one goroutine read by
another goroutine.
The stats are only populated when cAdvisor is running outside network namespaces.
We'll add a different backend to retrieve the same data from within namespaces.
This adds an Exists() interface to detect when the container is dead.
Before reporting an update error we check is Exists() is true.
Some documentation was added as well.
This allows a ContainerHandlerFactory to register a CanHandle() function
which is called to determine whether the factory can handle a particular
container.
This commit disables being able to run cAdvisor without lmctfy. This
should be enabled again with a "no-op" global factory which I would like
to do in a separate PR.
