*For a pool with partitionable slots*
By default, when the condor negotiator matches a partitionable slot to a job, it
-gives the whole partitionable slot to the schedd to use for as many jobs as possible. This creates a depth-first kind of pool filling, one where jobs tend to be packed as much as possible on the fewest number of machines. The best way to turn this off is to disable the schedd-side partitionable slot splitting. The downside of this change is that each machine can only match one new job per neegotiation cycle. Also, if ConsumptionPolicy has been enabled, this will also cause a depth-first filling of machines.
+gives the whole partitionable slot to the schedd to use for as many jobs as possible. This creates a depth-first kind of pool filling, one where jobs tend to be packed as much as possible on the fewest number of machines. The best way to turn this off is to disable the schedd-side partitionable slot splitting. The downside of this change is that each machine can only match one new job per negotiation cycle. Also, if ConsumptionPolicy has been enabled, this will also cause a depth-first filling of machines.
-One the schedd, make the following configuration change. This requires a condor_reconfig of the schedd to take effect.
+On the schedd, make the following configuration change. This requires a condor_reconfig of the schedd to take effect.
{code}
CLAIM_PARTITIONABLE_LEFTOVERS = false
