v1.36.1

• Revert a change that caused expensive logging in handling resource offers

v1.36.0

• Added support for multiple Kubernetes compute clusters
• Optimized quota reading

v1.35.0

• Added max ports to task constraints
• Added leader URL to /info
• Changed max priority to 16,000,000

v1.34.2

• Integration test improvements

v1.34.1

• Adds pool name to matching logs

v1.34.0

• Adds COOK_INSTANCE_NUM environment variable
• Adds metrics on instance fetch rates
• Captures the time it takes to list jobs

v1.26.0

• Added global launch rate limit
• Made per-user rate limit more gradual
• Sped up /unscheduled endpoint with new query and truncating long lists

v1.25.1

Only contains refactorings.

v1.21.0

• Added support for contacting a data local service to obtain cost data for scheduling
• Fixed bug in quota-checking when running without pools
• Fixed bug in the rebalancer's retrieval of DRU divisors when running with pools
• Fixed integer overflows in timer tasks when the scheduler runs for a long time

v1.17.2

• Fixed issue where task reconciliation was failing