[LU-18975] Nodemap-specific tunable parameter configuration

• Type: Improvement
• Resolution: Unresolved
• Priority: Minor
• Fix Version/s: Lustre 2.17.0
• Labels: easy
• Assignee: Fred Dilger
• Reporter: Andreas Dilger

Description

With nodemaps, the MGS should be able to configure individual nodemaps to have different parameters, like JOBID_VAR if one cluster is scheduled by SLURM and one is not, or for different MTFL tenants.

Adding a "-N|--name NODEMAP" argument to the "lctl set_param -P" command line would be one possible interface, and store the parameters into an params:NODEMAP file on the MGS that is sent to clients that are part of NODEMAP.

One potential issue with having a separate params:NODEMAP file is that the client would have to explicitly request this configuration file from the MGS at mount time, but the client itself does not even know it is part of NODEMAP. It seems impractical to "append" the params:NODEMAP records to the end of params, since this would mess up config llog processing on the client if some other parameter was added to params at runtime, since llog files are sequential and the client tracks its last-processed record index locally so that it will process only new records on each update.

It seems reasonable for clients to request some "extra" config file like params:nodemap (i.e. literally ":nodemap"), and it is up to the MGS to do the remapping to params:NODEMAP internally for each client). For many clients, that would be nodemap=default, so possibly params:default if it exists (or maybe that is just the regular params file and is ignored), and for clients in nodemap=NODEMAP they would get params:NODEMAP if it exists.

Another option might also be that the nodemap-specific params:NODEMAP file is supplied to the client instead of params if it exists, and this replaces the regular params file. That might make a bit more work for the admin since parameters would need to be copied over to each of the params:NODEMAP files, but it would allow for disjoint parameters settings (i.e. not everything in params is inherited by clients into params:NODEMAP).

Issue Links

• Clones: LU-11077 Client-specific tunable parameter configuration
• is related to: LU-14639 confusion of lru_size=0 if lru-resize disabled
• is related to: LU-16931 add lctl mechanism (list_param?) to report changes to tunable parameters
• is related to: LU-16749 apply a filter to the configuration log.
• is related to: LU-10360 use Imperative Recovery logs for client->MDT/OST connections
• is related to: LU-15246 Add per device adaptive timeout parameters