This adds some new time.time() logs around _process_broker that allow us
to see the length of time processing a db took. The information gets
added to the in_progress_shard range for the corresponding db.
Also included are some changes to the CleavingContext to enable tracking
the number of replicate calls + time spent running those replicate
calls.
UpgradeImpact: below new recon metrics have been added into
``sharding_in_progress`` section of sharder recon.
total_replicate_time: Cumulative time spent on replication during
cleaving after it completes.
processing_time: Time spent to process a single sharding broker.
total_sharding_time: Total time sharder spent on sharding from
start to finish.
tombstones: number of tombstones container has in total.
Change-Id: I4ad1e22fd011652d80269a76353e5fc14db249f5
Co-Authored-By: Jianjian Huo <jhuo@nvidia.com>
Signed-off-by: Christian Ohanaja <cohanaja@nvidia.com>
9babfefade