1.2.65

• A run that has ended keeps its outcome until the daemon records it. A
  concurrencyPolicy: Replace launch leaves that run as it is, a Forbid
  launch starts its job, and a pooled run that finishes while the pool
  scheduler renews its lease keeps its result. POST /jobs/{name}/cancel
  and the MCP cron_cancel_job tool ask the runs that have not ended to
  stop. When every run has ended, the request returns 409 and the tool
  reports an error.
• When the daemon cancels a run, it waits up to killTimeout for the job's
  own process and stops waiting when that process exits. On every platform
  except Windows, the forced kill of the rest of the process group follows
  within about 0.25 seconds. A workload that needs all of killTimeout to
  shut down must be the job's own process: an argv list, or exec as the
  last step of a shell command.
• The daemon records a canceled or timed-out run within a time limit when a
  process that the run started holds its output open. A canceled run has
  30 seconds. A run that its executionTimeout ended, or a verification
  check that reached verify.timeout, has 30 seconds for each stream. The
  daemon then logs a warning, keeps the output captured so far, and records
  the run.
• executionTimeout bounds a run until its captured output closes. When the
  job's own process exits inside the deadline and a process that it started
  still holds a captured stream open at the deadline, the daemon times the
  run out and records exit code -100.
• When the operating system refuses the daemon's signal to a job's process,
  such as EPERM for a command that sudo runs, the daemon logs a warning
  and completes the cancel sequence. The process keeps running until it
  exits by itself.
• A recorded run's finished_at is the instant the run ended, and ranAt
  and the archived output carry the same instant. A canceled or timed-out
  run that is recorded after a later run of the same job leaves the later
  run as the job's last run. onlyIfLastSucceeded, maxTimeSinceSuccess,
  and missed-run catch-up go by the instant each run ended.
• When two runs of a job overlap, the success of the earlier run leaves the
  retries of the later scheduled run in place until that run has scheduled
  a retry. Once a retry is pending or its attempt is running, a success of
  any run of the job ends the retry sequence. A run canceled through
  POST /jobs/{name}/cancel or cron_cancel_job follows the same rule.
• When concurrencyPolicy: Forbid drops a scheduled fire of a job that has
  no pool, the job's retry sequence stays as it is. A scheduled fire of a
  pooled job ends the job's retry sequence before it queues its own entry.
• When the store write that records a pooled entry's completion fails, the
  pool scheduler writes it again on its next pass and on each lease
  heartbeat, every five seconds.
• GET /pools, POST /pools/{name}/queue/{key}/cancel, and
  POST /jobs/{name}/start for a pooled job answer 503 with
  pool state is unavailable whenever pool state cannot be read or
  written. GET /jobs then answers 200, and the pool object of each
  pooled job carries queueUnavailable: true. The MCP cron_list_pools,
  cron_cancel_queued, and cron_run_job tools report the 503 as a tool
  error with the same text. The daemon logs a warning and drops a scheduled
  fire of a pooled job that finds pool state unavailable.
• Once a run is recorded and its reports are sent, the daemon frees the
  run's process handles and the copy of its output that reports read.
• A workflow whose configuration lists a task before its upstreams settles a
  failure in one scheduler pass.
• The workflow retention pass lists a workflow's runs once and deletes them
  in batches of up to 32, each run under its lease. It keeps a run that a
  recovery in preparation or an open recovery batch references. Under
  state.maxOpsPerSecond of about 2 or lower, or on a slow store, a busy
  workflow can hold more than retainRuns finished runs, and a scheduled
  workflow run can start late.
• A date-range recovery takes the lease of each source run before it writes
  its batch. While retention or another recovery holds the lease of a
  source run, the request answers 409 with
  source run is busy; retry shortly. When a source run is gone or its key
  holds another run, the message is
  recovery preview is stale; preview again.
• POST /dags/{name}/recover, POST /dags/{name}/runs/{run_key}/recover,
  and POST /dags/{name}/backfill answer 503 whenever the state store
  cannot answer: on a store I/O error or a timeout, on a document that
  cannot be read, and while the configured store has not started. The
  message is recovery state is unavailable or
  backfill state is unavailable. After a fault during a backfill, the
  same request creates the rest of the runs. The MCP
  cron_preview_recovery, cron_recover_dag, and cron_backfill_dag
  tools report these answers as tool errors.
• Publishing an artifact or an XCom value costs the same at any number of
  names in the scope. Workflow recovery's limit of 10,000 artifacts counts
  names.
• Nodes that share a state store need synchronized clocks, and they should
  move to this release together. Until they do, a node on the previous
  release can answer not found for an artifact or an XCom value while a
  node on this release republishes it.
• Planning a workflow rerun from a task follows each dependency once.
• A configuration directory of more than 1,024 files reloads by reparsing
  only the files that changed. An include tree of that size reparses each
  changed file, each file that includes it, and the entry file.
• A configuration file, an included file, or an env_file that changes
  while the daemon loads the configuration is parsed again. So is a
  configuration directory when a file joins or leaves it during the parse.
• A source that cannot be read twice (a pipe, a FIFO, a terminal,
  /dev/stdin, or /dev/fd/N) is read again only when the file that names
  it is parsed again. Jobs keep the environment read from a piped
  env_file across a reload request and across an edit to another file.
• When the configuration named by -c is a pipe, a FIFO, or a terminal,
  only a reload request (SIGHUP or cronstable service reload) reads it
  again. When another source changes, the once-a-minute check logs a
  warning and keeps the loaded configuration. A reload that reads zero
  bytes from such a configuration does the same. In both cases,
  cronstable_config_last_reload_successful reads 0.
• A file that two files of one include tree both include is loaded under
  each of them and is no cycle. Only a file that holds defaults alone can
  be shared: a shared file that defines a job, a pool, or a web or
  logging section fails the load. A file that includes itself, directly
  or through other files, fails with include cycle detected at <path>.
• A configuration file, a file of a configuration directory, or an included
  file that is not valid UTF-8 is a configuration error. The message is
  Could not load config file '<path>': followed by the decoding error.
  Windows PowerShell 5.1 writes little-endian UTF-16 when it redirects
  output to a file, which is the usual cause.
• The file that cluster.etcd.password.fromFile names is a configuration
  source, as an env_file is: the once-a-minute check parses the
  configuration again when the file changes. Change the password on the
  etcd server before you replace the file.
• Configuration parse time grows in proportion to the number of keys in a
  mapping. A pools or web.headers block with 2,000 entries loads in
  under a second.
• The jobs and workflow tasks of one configuration file share one read of an
  env_file that they name or inherit from defaults.
• cronstable --validate-config and cronstable --job-set-id exit without
  loading psutil. They load asyncio only for a configuration that has a
  push section or enables web.bonjour.
• GET /activity takes two optional query parameters that cap the jobs in
  the response. jobs=N returns at most N jobs: the first N in
  configuration order, or the first N by name with sort=name. Each
  dashboard draws at most 80 heatmap rows and sends that cap.
• The daemon compresses the dashboard page with zlib at level 9 for the
  first browser that accepts gzip and serves the stored result after that.
  Such a browser downloads about 180 KB. The page answers
  304 Not Modified to an If-None-Match header that carries its ETag
  in weak form or in a list.
• The MCP cron_list_jobs tool builds full job rows only for the page that
  it returns, so the cost of a call follows the page size.
• Some terminals send Alt with a navigation key as an Esc ahead of the
  key's own sequence. On every platform except Windows, the terminal
  dashboard reads that as the plain key, and the panel or drawer that is on
  top stays open.
• Except on Windows, the terminal dashboard asks the terminal to mark each
  paste (bracketed paste), and marked text is text only. The text field
  with the focus takes it without its line breaks and other control
  characters, and no pasted character acts as a shortcut. The Windows
  console marks no paste, so a pasted line break, tab, or escape arrives
  as Enter, Tab, or Esc.
• On Windows, the terminal dashboard reads key events from the console. A
  character outside the Basic Multilingual Plane, such as an emoji, arrives
  as one character, and a character composed with Alt and the numeric
  keypad is delivered. Shift-Tab steps the tabs of the job drawer and the
  state inspector backward. The dashboard restores the console's input mode
  when it exits, including after an error.
• In the web dashboard, a job's countdown and its Next at time stay tied
  to the /jobs response that delivered them. When the daemon answers a
  poll with 304 Not Modified, the dashboard keeps the jobs it holds, and
  their countdowns run on. After the computer sleeps or its clock is set,
  the next poll takes a full response and restarts the countdowns from it.
  The Next at label names its day from the browser's local date.

Performance compared with 1.2.64

Regression check: passed for 189 of 191 metrics with regression limits. The remaining metrics could not be compared.

Both versions were benchmarked in alternating rounds on the same machine. Results use the lowest runtime and median memory usage across rounds. Negative changes mean less time or memory; positive changes mean more.

Benchmark comparisons (197 metrics)

Noise (±) estimates run-to-run variation across both versions. A result fails the regression check only if the increase exceeds its percentage limit, its minimum absolute change, and 2 times the noise estimate. n/a means too few rounds to estimate noise; the other limits still apply.

The Regression limit column accounts for the minimum absolute change, expressed as a percentage of the baseline. Not enforced marks informational metrics.

Startup times include Python startup overhead. Their percentage changes exclude that overhead when it can be measured separately and leaves a positive runtime in both versions.

Benchmark	1.2.64	1.2.65	Change	Noise (±)	Regression limit
dag.gc_excess_60	1.192 s	59.62 ms	-95.0%	2.8%	25%
state.artifact_put_distinct_2k	496.71 ms	25.31 ms	-94.9%	1.8%	25%
dag.recover_plan_from_1k	1.314 s	98.09 ms	-92.5%	1.3%	15%
pools.admit_cycle_1k	1.247 s	126.68 ms	-89.8%	0.7%	25%
webui.dirty_frame_rows_500	10.24 ms	1.22 ms	-88.1%	6.4%	Not enforced
state.artifact_get_oldest_2k	544.58 ms	70.92 ms	-87.0%	0.8%	25%
mcp.list_jobs_page_500	728.51 ms	118.17 ms	-83.8%	1.5%	15%
webui.boot_rows_500	731.70 ms	469.30 ms	-35.9%	4.9%	Not enforced
mcp.list_jobs_500	178.44 ms	114.57 ms	-35.8%	1.2%	15%
config.parse_yaml_rich_20	300.65 ms	213.79 ms	-28.9%	0.5%	15%
startup.validate_config_100	202.64 ms	163.02 ms	-21.2%	0.7%	25%
startup.job_set_id_100	206.54 ms	168.70 ms	-19.8%	0.7%	25%
mem.launch_reap_steady_20x55	0.44 MB	0.37 MB	-15.2%	0.0%	15%
webapi.index_gzip_bytes	213.63 KB	181.36 KB	-15.1%	0.0%	15%
state.artifact_get_newest	11.23 ms	9.69 ms	-13.7%	2.0%	44.51% (declared 25%)
loop.stall_launch_herd_8k	54.38 ms	61.53 ms	+13.1%	2.1%	Not enforced
config.parse_yaml_3k	1.580 s	1.435 s	-9.2%	0.6%	15%
startup.import_cronexpr	24.41 ms	25.01 ms	+8.8%	1.0%	25%
config.parse_yaml_300	157.55 ms	143.97 ms	-8.6%	0.4%	15%
config.reload_warm_include_50	79.83 ms	74.06 ms	-7.2%	0.4%	25%
state.derive_max_cold	12.77 ms	13.68 ms	+7.1%	1.0%	78.33% (declared 15%)
config.reload_warm_50	36.61 ms	34.13 ms	-6.8%	1.1%	27.32% (declared 25%)
state.list_records_2k	14.49 ms	15.45 ms	+6.6%	4.1%	69.01% (declared 15%)
job.launch_reap_2k	108.50 ms	115.08 ms	+6.1%	1.2%	15%
startup.daemon_first_pass_2k	1.423 s	1.341 s	-5.9%	0.3%	25%
job.launch_fail_retry_2k	125.69 ms	131.86 ms	+4.9%	1.3%	15%
loop.sla_pass_5k_x10	126.66 ms	132.75 ms	+4.8%	3.3%	15%
state.boot_rehydrate_populated	26.05 ms	27.14 ms	+4.2%	1.3%	38.38% (declared 15%)
config.reload_warm_dir_1k	182.37 ms	188.77 ms	+3.5%	0.3%	25%
webui.rows_layout_200	57.40 ms	55.40 ms	-3.5%	3.1%	25%
state.fanout_gather_100	4.70 ms	4.85 ms	+3.1%	3.3%	212.8% (declared 25%)
webui.tick_500	81.50 ms	79.30 ms	-2.7%	5.0%	25%
config.reload_gc_100k	1.990 s	1.941 s	-2.5%	1.6%	25%
job.report_noop_100k	41.34 ms	42.36 ms	+2.4%	0.7%	15%
tui.poll_absorb_5k	87.73 ms	85.62 ms	-2.4%	1.7%	15%
tui.wallboard_paint_5k	161.21 ms	157.36 ms	-2.4%	1.8%	15%
loop.stall_mcp_jobs_500	6.69 ms	6.84 ms	+2.4%	1.8%	Not enforced
webapi.jobs_payload_500	41.92 ms	42.90 ms	+2.3%	2.2%	23.85% (declared 15%)
loop.stall_metrics_2000	24.10 ms	24.66 ms	+2.3%	2.4%	Not enforced
prometheus.first_scrape_after_reload_500	162.41 ms	166.10 ms	+2.3%	1.9%	15%
loop.idle_wake_rate	12.23 ms	12.49 ms	+2.1%	2.3%	40.88% (declared 25%)
webui.radar_walk_500x150	52.10 ms	51.00 ms	-2.1%	1.8%	25%
config.jobconfig_3k	50.18 ms	49.23 ms	-1.9%	1.7%	19.93% (declared 15%)
state.append_pruned_1k	38.91 ms	39.64 ms	+1.9%	2.3%	25%
state.gc_sweep_2k_streams	20.81 ms	20.42 ms	-1.9%	2.5%	72.07% (declared 25%)
config.parse_dir_cold_200	243.32 ms	238.96 ms	-1.8%	1.1%	25%
webui.filter_keys_500	57.80 ms	56.80 ms	-1.7%	2.7%	25%
schedule.pressure_20k_48h	13.98 ms	14.22 ms	+1.7%	2.5%	71.52% (declared 15%)
webui.select_move_500	64.10 ms	63.00 ms	-1.7%	6.3%	25%
dag.adopt_scan_500	50.45 ms	51.24 ms	+1.6%	2.0%	39.64% (declared 25%)
cronexpr.parse_complex	18.18 ms	17.91 ms	-1.5%	2.9%	55.02% (declared 15%)
dag.mapped_drain_256	15.57 ms	15.35 ms	-1.4%	2.4%	64.23% (declared 25%)
webui.wallboard_poll_500	55.80 ms	55.00 ms	-1.4%	3.7%	25%
tui.palette_type_5k	152.98 ms	150.83 ms	-1.4%	1.2%	15%
dag.full_sweep_50x200	106.30 ms	104.81 ms	-1.4%	4.4%	25%
statsd.emit_2k	7.14 ms	7.24 ms	+1.4%	0.5%	140% (declared 25%)
redact.adversarial_10k	29.63 ms	30.03 ms	+1.4%	0.6%	33.75% (declared 15%)
webui.sort_500	52.70 ms	52.00 ms	-1.3%	3.2%	25%
mcp.query_metrics_500	115.95 ms	114.50 ms	-1.3%	1.5%	25%
schedule.due_pass_100k	16.32 ms	16.11 ms	-1.3%	2.7%	61.28% (declared 25%)
dag.finish_fanin_20x1k	57.08 ms	57.77 ms	+1.2%	1.8%	25%
config.parse_crontab_1k	9.83 ms	9.71 ms	-1.2%	2.2%	101.8% (declared 15%)
config.interp_2k	52.77 ms	52.16 ms	-1.2%	0.9%	18.95% (declared 15%)
state.list_records_warm	89.88 ms	90.90 ms	+1.1%	2.6%	15%
mem.gc_pause_100k	142.12 ms	143.70 ms	+1.1%	0.4%	25%
webui.render_fleet_15x400	22.64 ms	22.40 ms	-1.1%	2.8%	25%
webapi.resources_poll_20x240	120.93 ms	122.18 ms	+1.0%	1.3%	15%
state.list_documents_600	33.68 ms	34.03 ms	+1.0%	1.7%	35.63% (declared 15%)
job.stream_passthrough_240k	117.56 ms	118.74 ms	+1.0%	1.2%	15%
cluster.derive_cold_50	130.40 ms	131.69 ms	+1.0%	1.1%	15%
mcp.fleet_15x400	146.67 ms	148.09 ms	+1.0%	1.4%	15%
state.lease_renew_200	21.62 ms	21.41 ms	-1.0%	1.2%	231.3% (declared 25%)
loop.stall_completions_500	6.64 ms	6.70 ms	+1.0%	2.6%	75.29% (declared 25%)
webui.rows_diff_500	53.20 ms	52.70 ms	-0.9%	2.0%	25%
tui.dag_graph_paint_2k	183.78 ms	182.07 ms	-0.9%	1.4%	15%
pools.enqueue_burst_200	103.10 ms	102.15 ms	-0.9%	2.1%	25%
jobapi.sem_denied_64x10	39.68 ms	39.32 ms	-0.9%	0.9%	25%
state.artifact_list_churn_x500	35.69 ms	35.37 ms	-0.9%	0.6%	28.02% (declared 25%)
webapi.dags_poll_1k	111.43 ms	112.42 ms	+0.9%	0.5%	15%
schedule.lint_2500_zoned	52.43 ms	51.99 ms	-0.8%	3.1%	15%
redact.secrets_5k	17.55 ms	17.69 ms	+0.8%	0.8%	56.99% (declared 15%)
state.mutate_document_1k	112.43 ms	113.35 ms	+0.8%	1.3%	44.47% (declared 25%)
mem.rss_daemon_import	37.66 MB	37.96 MB	+0.8%	0.1%	25%
mem.rss_daemon_web_100	52.46 MB	52.88 MB	+0.8%	0.1%	25%
schedule.cold_build_100k	392.88 ms	395.97 ms	+0.8%	3.0%	15%
cronexpr.next_simple	32.39 ms	32.14 ms	-0.8%	1.4%	30.87% (declared 15%)
ical.render_500x7d	42.61 ms	42.29 ms	-0.8%	2.1%	23.47% (declared 15%)
tui.log_restyle_5k	25.34 ms	25.52 ms	+0.7%	0.8%	39.47% (declared 15%)
cronexpr.next_dst_2k	32.89 ms	32.65 ms	-0.7%	4.1%	30.41% (declared 15%)
startup.python_baseline	16.31 ms	16.20 ms	-0.7%	1.9%	Not enforced
webapi.sse_burst_20k	39.89 ms	40.16 ms	+0.7%	1.3%	15%
webapi.index_bytes	631.03 KB	635.35 KB	+0.7%	0.0%	15%
dag.plan_claim_10k	12.23 ms	12.31 ms	+0.7%	3.2%	81.79% (declared 15%)
webapi.sse_tail_pump_4x20k	90.74 ms	91.35 ms	+0.7%	1.7%	15%
tui.week_rows_500	110.54 ms	109.80 ms	-0.7%	0.5%	15%
startup.mcp_bridge	100.28 ms	99.61 ms	-0.7%	1.2%	25%
json.loads_wide_int_2k	203.59 ms	202.25 ms	-0.7%	0.9%	15%
cluster.parse_summaries_6k	71.73 ms	72.19 ms	+0.6%	1.9%	15%
schedule.reseed_100k	167.93 ms	169.00 ms	+0.6%	1.8%	15%
fingerprint.job_set_id_10k	259.86 ms	261.46 ms	+0.6%	0.6%	15%
mcp.handle_200	123.06 ms	123.81 ms	+0.6%	0.4%	15%
webui.palette_keys_500	49.80 ms	50.10 ms	+0.6%	2.1%	25%
webapi.dag_run_poll_1k	80.06 ms	80.53 ms	+0.6%	0.8%	25%
resources.monitor_stop_100	148.46 ms	147.61 ms	-0.6%	2.0%	25%
startup.jobcli_state_get	102.75 ms	102.14 ms	-0.6%	1.3%	25%
json.roundtrip_3k	504.58 ms	501.84 ms	-0.5%	1.3%	15%
webui.timeline_500	55.70 ms	56.00 ms	+0.5%	1.1%	25%
state.depends_on_past_gate	1.77 ms	1.78 ms	+0.5%	3.9%	113.1% (declared 25%)
dag.build_layered_10k	44.84 ms	45.07 ms	+0.5%	2.1%	22.3% (declared 15%)
prometheus.render_500	15.20 ms	15.28 ms	+0.5%	1.8%	65.79% (declared 15%)
resources.usage_from_dict_500	77.94 ms	77.56 ms	-0.5%	2.3%	15%
startup.import_config	94.28 ms	93.79 ms	-0.5%	1.7%	25%
cluster.peer_serve_400	133.83 ms	134.46 ms	+0.5%	0.6%	15%
schedule.next_fires_2k	20.04 ms	19.95 ms	-0.5%	2.0%	49.9% (declared 15%)
dag.advance_quiescent_1k	63.25 ms	63.55 ms	+0.5%	0.4%	25%
startup.version	51.54 ms	51.58 ms	+0.5%	1.3%	25%
state.append_1k	52.51 ms	52.28 ms	-0.4%	1.0%	95.22% (declared 15%)
webui.render_rows_500	16.20 ms	16.13 ms	-0.4%	2.6%	25%
job.stream_capture_120k	29.75 ms	29.62 ms	-0.4%	1.4%	33.61% (declared 15%)
dag.advance_quiescent_chain	62.56 ms	62.81 ms	+0.4%	0.8%	25%
tui.drawer_paint_5k	89.97 ms	89.61 ms	-0.4%	1.2%	15%
tui.view_sort_5k	105.95 ms	106.37 ms	+0.4%	0.6%	15%
push.seal_500	86.82 ms	87.15 ms	+0.4%	0.8%	15%
cronexpr.test_match_200k	21.50 ms	21.58 ms	+0.4%	1.2%	46.51% (declared 15%)
webui.week_walk_500x20	55.80 ms	55.60 ms	-0.4%	0.7%	25%
cluster.poll_round_15x400	137.07 ms	136.61 ms	-0.3%	0.7%	15%
state.derive_max_warm	121.30 ms	121.71 ms	+0.3%	2.2%	15%
schedule.reseed_local_20k	74.72 ms	74.47 ms	-0.3%	1.2%	15%
dag.run_wide_500	91.13 ms	91.43 ms	+0.3%	1.4%	25%
tui.tail_ingest_30k	144.15 ms	143.69 ms	-0.3%	0.9%	15%
prometheus.counter_snapshot_2k	106.36 ms	106.69 ms	+0.3%	1.5%	15%
schedule.suggest_slot_5k_x20	63.64 ms	63.83 ms	+0.3%	1.3%	15%
cronexpr.occurrences_1k	11.37 ms	11.40 ms	+0.3%	2.4%	87.98% (declared 15%)
webui.render_term_5k	98.10 ms	98.40 ms	+0.3%	1.4%	25%
dag.list_dags_warm_x300	38.40 ms	38.29 ms	-0.3%	1.2%	25%
tui.table_paint_5k	142.00 ms	142.42 ms	+0.3%	1.5%	15%
tui.mark_idle_300	226.59 ms	227.26 ms	+0.3%	1.6%	15%
cluster.fleet_view_15x400	81.93 ms	82.17 ms	+0.3%	0.7%	25%
dag.advance_waiting_mapped_1k	106.38 ms	106.08 ms	-0.3%	0.8%	25%
cronexpr.parse_simple	36.91 ms	36.81 ms	-0.3%	1.1%	27.09% (declared 15%)
push.seal_xwing_500	112.27 ms	112.56 ms	+0.3%	0.4%	15%
push.fanout_8x100	143.39 ms	143.76 ms	+0.3%	1.2%	15%
cronexpr.next_complex	12.62 ms	12.65 ms	+0.2%	1.6%	79.23% (declared 15%)
startup.import_daemon	180.26 ms	180.53 ms	+0.2%	0.9%	25%
dag.run_chain_100	127.08 ms	127.38 ms	+0.2%	0.6%	25%
webapi.activity_build_500	161.60 ms	161.97 ms	+0.2%	1.0%	15%
json.roundtrip_orjson_3k	173.71 ms	173.32 ms	-0.2%	1.9%	15%
redact.clean_20k	18.22 ms	18.18 ms	-0.2%	2.3%	54.88% (declared 15%)
mem.rss_version	14.41 MB	14.44 MB	+0.2%	0.6%	25%
webui.dag_tasks_2k	49.50 ms	49.40 ms	-0.2%	2.5%	25%
mem.rss_daemon_10k	142.52 MB	142.79 MB	+0.2%	0.1%	25%
jobapi.read_path_10k	108.94 ms	108.74 ms	-0.2%	1.8%	15%
schedule.duplicates_20k	16.83 ms	16.86 ms	+0.2%	2.9%	59.41% (declared 15%)
webui.tail_term_2k	55.20 ms	55.10 ms	-0.2%	1.6%	25%
fingerprint.job_set_id_rich_2k	213.04 ms	213.42 ms	+0.2%	0.7%	15%
webapi.summary_status_500	133.19 ms	132.96 ms	-0.2%	1.8%	15%
state.inventory_2k_streams	55.45 ms	55.36 ms	-0.2%	1.5%	Not enforced
dag.list_runs_warm	7.96 ms	7.98 ms	+0.2%	1.4%	25%
webapi.gzip_body_500	51.84 ms	51.92 ms	+0.2%	0.8%	19.29% (declared 15%)
dag.fanin_barrier_1k	164.63 ms	164.87 ms	+0.1%	0.6%	25%
mem.jobconfig_2k	1.62 MB	1.62 MB	+0.1%	0.6%	30.94% (declared 15%)
startup.import_tui	94.83 ms	94.81 ms	+0.1%	1.1%	25%
pair.qr_encode_14	223.64 ms	223.91 ms	+0.1%	1.8%	15%
dag.build_chain_10k	25.64 ms	25.61 ms	-0.1%	1.9%	39.01% (declared 15%)
webui.logo_recovery	85.20 ms	85.10 ms	-0.1%	1.3%	25%
cluster.job_owner_2k	51.51 ms	51.56 ms	+0.1%	1.1%	19.41% (declared 15%)
tui.log_search_20k	13.27 ms	13.26 ms	-0.1%	1.6%	75.37% (declared 15%)
tui.fleet_paint_15x400	121.01 ms	121.08 ms	+0.1%	0.7%	15%
webapi.trends_build_5k	123.24 ms	123.19 ms	-0.0%	2.4%	15%
state.kv_roundtrip_200	24.97 ms	24.96 ms	-0.0%	1.8%	40.05% (declared 15%)
webapi.pools_poll_4x130	92.73 ms	92.76 ms	+0.0%	0.7%	25%
loop.stall_jobs_500	18.08 ms	18.08 ms	+0.0%	3.0%	Not enforced
mem.jobconfig_envfile_2k	16.37 MB	16.37 MB	+0.0%	0.1%	15%
mem.gc_manifest_scan_128x2k	14.32 MB	14.32 MB	+0.0%	0.0%	15%
mem.parse_peak_rich_20	4.21 MB	4.21 MB	-0.0%	0.0%	15%
mem.tui_state_5k	2.23 MB	2.23 MB	+0.0%	0.0%	15%
webapi.auth_scope_20k	69.97 ms	69.97 ms	-0.0%	1.7%	15%
prometheus.exposition_bytes_500	821.15 KB	821.17 KB	+0.0%	0.0%	15%
webui.append_line_5k	12.0 us	12.0 us	-0.0%	0.0%	25%
dag.run_doc_bytes_1k	228.53 KB	228.53 KB	+0.0%	0.0%	5%
dag.mapped_doc_bytes_1k	725.81 KB	725.81 KB	+0.0%	0.0%	5%
tui.frame_bytes_5k	60.22 KB	60.22 KB	+0.0%	0.0%	15%
webui.log_count_5k	80.0 us	80.0 us	+0.0%	0.0%	2500% (declared 25%)
webui.ledger_analyze_500x600	75.90 ms	75.90 ms	+0.0%	1.2%	25%
webapi.jobs_bytes_500	181.42 KB	181.42 KB	+0.0%	0.0%	15%
webapi.jobs_gzip_500	5.19 KB	5.19 KB	+0.0%	0.0%	15%
webapi.activity_bytes_500	2569.83 KB	2569.83 KB	+0.0%	0.0%	15%
webapi.fleet_bytes_15x400	963.62 KB	963.62 KB	+0.0%	0.0%	15%
cluster.peer_bytes_512	83.55 KB	83.55 KB	+0.0%	0.0%	15%
prometheus.counter_snapshot_bytes_2k	620.59 KB	620.59 KB	+0.0%	0.0%	15%
mcp.list_jobs_bytes_500	155.54 KB	155.54 KB	+0.0%	0.0%	15%
mcp.fleet_bytes_15x400	2052.76 KB	2052.76 KB	+0.0%	0.0%	15%
mcp.tools_list_bytes	19.95 KB	19.95 KB	+0.0%	0.0%	15%
mem.crontab_10k	2.80 MB	2.80 MB	+0.0%	0.0%	17.85% (declared 15%)
mem.retired_output_1k	0.12 MB	0.12 MB	+0.0%	0.0%	15%
mem.run_history_series_5	9.57 MB	9.57 MB	+0.0%	0.0%	15%
mem.cluster_view_50x400	11.64 MB	11.64 MB	+0.0%	0.0%	15%

No percentage comparison available: webapi.activity_capped_bytes_500, webui.swim_400x49.

Full Changelog: 1.2.64...1.2.65