Source file src/runtime/metrics/doc.go
1 // Copyright 2020 The Go Authors. All rights reserved. 2 // Use of this source code is governed by a BSD-style 3 // license that can be found in the LICENSE file. 4 5 // Note: run 'go generate' (which will run 'go test -generate') to update the "Supported metrics" list. 6 //go:generate go test -run=Docs -generate 7 8 /* 9 Package metrics provides a stable interface to access implementation-defined 10 metrics exported by the Go runtime. This package is similar to existing functions 11 like [runtime.ReadMemStats] and [runtime/debug.ReadGCStats], but significantly more general. 12 13 The set of metrics defined by this package may evolve as the runtime itself 14 evolves, and also enables variation across Go implementations, whose relevant 15 metric sets may not intersect. 16 17 # Interface 18 19 Metrics are designated by a string key, rather than, for example, a field name in 20 a struct. The full list of supported metrics is always available in the slice of 21 Descriptions returned by [All]. Each [Description] also includes useful information 22 about the metric. 23 24 Thus, users of this API are encouraged to sample supported metrics defined by the 25 slice returned by All to remain compatible across Go versions. Of course, situations 26 arise where reading specific metrics is critical. For these cases, users are 27 encouraged to use build tags, and although metrics may be deprecated and removed, 28 users should consider this to be an exceptional and rare event, coinciding with a 29 very large change in a particular Go implementation. 30 31 Each metric key also has a "kind" (see [ValueKind]) that describes the format of the 32 metric's value. 33 In the interest of not breaking users of this package, the "kind" for a given metric 34 is guaranteed not to change. If it must change, then a new metric will be introduced 35 with a new key and a new "kind." 36 37 # Metric key format 38 39 As mentioned earlier, metric keys are strings. Their format is simple and well-defined, 40 designed to be both human and machine readable. It is split into two components, 41 separated by a colon: a rooted path and a unit. The choice to include the unit in 42 the key is motivated by compatibility: if a metric's unit changes, its semantics likely 43 did also, and a new key should be introduced. 44 45 For more details on the precise definition of the metric key's path and unit formats, see 46 the documentation of the Name field of the Description struct. 47 48 # A note about floats 49 50 This package supports metrics whose values have a floating-point representation. In 51 order to improve ease-of-use, this package promises to never produce the following 52 classes of floating-point values: NaN, infinity. 53 54 # Supported metrics 55 56 Below is the full list of supported metrics, ordered lexicographically. 57 58 /cgo/go-to-c-calls:calls 59 Count of calls made from Go to C by the current process. 60 61 /cpu/classes/gc/mark/assist:cpu-seconds 62 Estimated total CPU time goroutines spent performing GC 63 tasks to assist the GC and prevent it from falling behind the 64 application. This metric is an overestimate, and not directly 65 comparable to system CPU time measurements. Compare only with 66 other /cpu/classes metrics. 67 68 /cpu/classes/gc/mark/dedicated:cpu-seconds 69 Estimated total CPU time spent performing GC tasks on processors 70 (as defined by GOMAXPROCS) dedicated to those tasks. This metric 71 is an overestimate, and not directly comparable to system CPU 72 time measurements. Compare only with other /cpu/classes metrics. 73 74 /cpu/classes/gc/mark/idle:cpu-seconds 75 Estimated total CPU time spent performing GC tasks on spare CPU 76 resources that the Go scheduler could not otherwise find a use 77 for. This should be subtracted from the total GC CPU time to 78 obtain a measure of compulsory GC CPU time. This metric is an 79 overestimate, and not directly comparable to system CPU time 80 measurements. Compare only with other /cpu/classes metrics. 81 82 /cpu/classes/gc/pause:cpu-seconds 83 Estimated total CPU time spent with the application paused by 84 the GC. Even if only one thread is running during the pause, 85 this is computed as GOMAXPROCS times the pause latency because 86 nothing else can be executing. This is the exact sum of samples 87 in /sched/pauses/total/gc:seconds if each sample is multiplied 88 by GOMAXPROCS at the time it is taken. This metric is an 89 overestimate, and not directly comparable to system CPU time 90 measurements. Compare only with other /cpu/classes metrics. 91 92 /cpu/classes/gc/total:cpu-seconds 93 Estimated total CPU time spent performing GC tasks. This metric 94 is an overestimate, and not directly comparable to system CPU 95 time measurements. Compare only with other /cpu/classes metrics. 96 Sum of all metrics in /cpu/classes/gc. 97 98 /cpu/classes/idle:cpu-seconds 99 Estimated total available CPU time not spent executing 100 any Go or Go runtime code. In other words, the part of 101 /cpu/classes/total:cpu-seconds that was unused. This metric is 102 an overestimate, and not directly comparable to system CPU time 103 measurements. Compare only with other /cpu/classes metrics. 104 105 /cpu/classes/scavenge/assist:cpu-seconds 106 Estimated total CPU time spent returning unused memory to the 107 underlying platform in response eagerly in response to memory 108 pressure. This metric is an overestimate, and not directly 109 comparable to system CPU time measurements. Compare only with 110 other /cpu/classes metrics. 111 112 /cpu/classes/scavenge/background:cpu-seconds 113 Estimated total CPU time spent performing background tasks to 114 return unused memory to the underlying platform. This metric is 115 an overestimate, and not directly comparable to system CPU time 116 measurements. Compare only with other /cpu/classes metrics. 117 118 /cpu/classes/scavenge/total:cpu-seconds 119 Estimated total CPU time spent performing tasks that return 120 unused memory to the underlying platform. This metric is an 121 overestimate, and not directly comparable to system CPU time 122 measurements. Compare only with other /cpu/classes metrics. 123 Sum of all metrics in /cpu/classes/scavenge. 124 125 /cpu/classes/total:cpu-seconds 126 Estimated total available CPU time for user Go code or the Go 127 runtime, as defined by GOMAXPROCS. In other words, GOMAXPROCS 128 integrated over the wall-clock duration this process has been 129 executing for. This metric is an overestimate, and not directly 130 comparable to system CPU time measurements. Compare only with 131 other /cpu/classes metrics. Sum of all metrics in /cpu/classes. 132 133 /cpu/classes/user:cpu-seconds 134 Estimated total CPU time spent running user Go code. This may 135 also include some small amount of time spent in the Go runtime. 136 This metric is an overestimate, and not directly comparable 137 to system CPU time measurements. Compare only with other 138 /cpu/classes metrics. 139 140 /gc/cleanups/executed:cleanups 141 Approximate total count of cleanup functions (created 142 by runtime.AddCleanup) executed by the runtime. Subtract 143 /gc/cleanups/queued:cleanups to approximate cleanup queue 144 length. Useful for detecting slow cleanups holding up the queue. 145 146 /gc/cleanups/queued:cleanups 147 Approximate total count of cleanup functions (created by 148 runtime.AddCleanup) queued by the runtime for execution. 149 Subtract from /gc/cleanups/executed:cleanups to approximate 150 cleanup queue length. Useful for detecting slow cleanups holding 151 up the queue. 152 153 /gc/cycles/automatic:gc-cycles 154 Count of completed GC cycles generated by the Go runtime. 155 156 /gc/cycles/forced:gc-cycles 157 Count of completed GC cycles forced by the application. 158 159 /gc/cycles/total:gc-cycles 160 Count of all completed GC cycles. 161 162 /gc/finalizers/executed:finalizers 163 Total count of finalizer functions (created by 164 runtime.SetFinalizer) executed by the runtime. Subtract 165 /gc/finalizers/queued:finalizers to approximate finalizer queue 166 length. Useful for detecting finalizers overwhelming the queue, 167 either by being too slow, or by there being too many of them. 168 169 /gc/finalizers/queued:finalizers 170 Total count of finalizer functions (created by 171 runtime.SetFinalizer) and queued by the runtime for execution. 172 Subtract from /gc/finalizers/executed:finalizers to approximate 173 finalizer queue length. Useful for detecting slow finalizers 174 holding up the queue. 175 176 /gc/gogc:percent 177 Heap size target percentage configured by the user, otherwise 178 100. This value is set by the GOGC environment variable, and the 179 runtime/debug.SetGCPercent function. 180 181 /gc/gomemlimit:bytes 182 Go runtime memory limit configured by the user, otherwise 183 math.MaxInt64. This value is set by the GOMEMLIMIT environment 184 variable, and the runtime/debug.SetMemoryLimit function. 185 186 /gc/heap/allocs-by-size:bytes 187 Distribution of heap allocations by approximate size. 188 Bucket counts increase monotonically. Note that this does not 189 include tiny objects as defined by /gc/heap/tiny/allocs:objects, 190 only tiny blocks. 191 192 /gc/heap/allocs:bytes 193 Cumulative sum of memory allocated to the heap by the 194 application. 195 196 /gc/heap/allocs:objects 197 Cumulative count of heap allocations triggered by the 198 application. Note that this does not include tiny objects as 199 defined by /gc/heap/tiny/allocs:objects, only tiny blocks. 200 201 /gc/heap/frees-by-size:bytes 202 Distribution of freed heap allocations by approximate size. 203 Bucket counts increase monotonically. Note that this does not 204 include tiny objects as defined by /gc/heap/tiny/allocs:objects, 205 only tiny blocks. 206 207 /gc/heap/frees:bytes 208 Cumulative sum of heap memory freed by the garbage collector. 209 210 /gc/heap/frees:objects 211 Cumulative count of heap allocations whose storage was freed 212 by the garbage collector. Note that this does not include tiny 213 objects as defined by /gc/heap/tiny/allocs:objects, only tiny 214 blocks. 215 216 /gc/heap/goal:bytes 217 Heap size target for the end of the GC cycle. 218 219 /gc/heap/live:bytes 220 Heap memory occupied by live objects that were marked by the 221 previous GC. 222 223 /gc/heap/objects:objects 224 Number of objects, live or unswept, occupying heap memory. 225 226 /gc/heap/tiny/allocs:objects 227 Count of small allocations that are packed together into blocks. 228 These allocations are counted separately from other allocations 229 because each individual allocation is not tracked by the 230 runtime, only their block. Each block is already accounted for 231 in allocs-by-size and frees-by-size. 232 233 /gc/limiter/last-enabled:gc-cycle 234 GC cycle the last time the GC CPU limiter was enabled. 235 This metric is useful for diagnosing the root cause of an 236 out-of-memory error, because the limiter trades memory for CPU 237 time when the GC's CPU time gets too high. This is most likely 238 to occur with use of SetMemoryLimit. The first GC cycle is cycle 239 1, so a value of 0 indicates that it was never enabled. 240 241 /gc/pauses:seconds 242 Deprecated. Prefer the identical /sched/pauses/total/gc:seconds. 243 244 /gc/scan/globals:bytes 245 The total amount of global variable space that is scannable. 246 247 /gc/scan/heap:bytes 248 The total amount of heap space that is scannable. 249 250 /gc/scan/stack:bytes 251 The number of bytes of stack that were scanned last GC cycle. 252 253 /gc/scan/total:bytes 254 The total amount space that is scannable. Sum of all metrics in 255 /gc/scan. 256 257 /gc/stack/starting-size:bytes 258 The stack size of new goroutines. 259 260 /godebug/non-default-behavior/allowmultiplevcs:events 261 The number of non-default behaviors executed by the cmd/go 262 package due to a non-default GODEBUG=allowmultiplevcs=... 263 setting. 264 265 /godebug/non-default-behavior/asynctimerchan:events 266 The number of non-default behaviors executed by the time package 267 due to a non-default GODEBUG=asynctimerchan=... setting. 268 269 /godebug/non-default-behavior/containermaxprocs:events 270 The number of non-default behaviors executed by the runtime 271 package due to a non-default GODEBUG=containermaxprocs=... 272 setting. 273 274 /godebug/non-default-behavior/embedfollowsymlinks:events 275 The number of non-default behaviors executed by the cmd/go 276 package due to a non-default GODEBUG=embedfollowsymlinks=... 277 setting. 278 279 /godebug/non-default-behavior/execerrdot:events 280 The number of non-default behaviors executed by the os/exec 281 package due to a non-default GODEBUG=execerrdot=... setting. 282 283 /godebug/non-default-behavior/gocachehash:events 284 The number of non-default behaviors executed by the cmd/go 285 package due to a non-default GODEBUG=gocachehash=... setting. 286 287 /godebug/non-default-behavior/gocachetest:events 288 The number of non-default behaviors executed by the cmd/go 289 package due to a non-default GODEBUG=gocachetest=... setting. 290 291 /godebug/non-default-behavior/gocacheverify:events 292 The number of non-default behaviors executed by the cmd/go 293 package due to a non-default GODEBUG=gocacheverify=... setting. 294 295 /godebug/non-default-behavior/gotestjsonbuildtext:events 296 The number of non-default behaviors executed by the cmd/go 297 package due to a non-default GODEBUG=gotestjsonbuildtext=... 298 setting. 299 300 /godebug/non-default-behavior/gotypesalias:events 301 The number of non-default behaviors executed by the go/types 302 package due to a non-default GODEBUG=gotypesalias=... setting. 303 304 /godebug/non-default-behavior/http2client:events 305 The number of non-default behaviors executed by the net/http 306 package due to a non-default GODEBUG=http2client=... setting. 307 308 /godebug/non-default-behavior/http2server:events 309 The number of non-default behaviors executed by the net/http 310 package due to a non-default GODEBUG=http2server=... setting. 311 312 /godebug/non-default-behavior/httplaxcontentlength:events 313 The number of non-default behaviors executed by the net/http 314 package due to a non-default GODEBUG=httplaxcontentlength=... 315 setting. 316 317 /godebug/non-default-behavior/httpmuxgo121:events 318 The number of non-default behaviors executed by the net/http 319 package due to a non-default GODEBUG=httpmuxgo121=... setting. 320 321 /godebug/non-default-behavior/httpservecontentkeepheaders:events 322 The number of non-default behaviors executed 323 by the net/http package due to a non-default 324 GODEBUG=httpservecontentkeepheaders=... setting. 325 326 /godebug/non-default-behavior/installgoroot:events 327 The number of non-default behaviors executed by the go/build 328 package due to a non-default GODEBUG=installgoroot=... setting. 329 330 /godebug/non-default-behavior/multipartmaxheaders:events 331 The number of non-default behaviors executed by 332 the mime/multipart package due to a non-default 333 GODEBUG=multipartmaxheaders=... setting. 334 335 /godebug/non-default-behavior/multipartmaxparts:events 336 The number of non-default behaviors executed by 337 the mime/multipart package due to a non-default 338 GODEBUG=multipartmaxparts=... setting. 339 340 /godebug/non-default-behavior/multipathtcp:events 341 The number of non-default behaviors executed by the net package 342 due to a non-default GODEBUG=multipathtcp=... setting. 343 344 /godebug/non-default-behavior/netedns0:events 345 The number of non-default behaviors executed by the net package 346 due to a non-default GODEBUG=netedns0=... setting. 347 348 /godebug/non-default-behavior/panicnil:events 349 The number of non-default behaviors executed by the runtime 350 package due to a non-default GODEBUG=panicnil=... setting. 351 352 /godebug/non-default-behavior/randautoseed:events 353 The number of non-default behaviors executed by the math/rand 354 package due to a non-default GODEBUG=randautoseed=... setting. 355 356 /godebug/non-default-behavior/randseednop:events 357 The number of non-default behaviors executed by the math/rand 358 package due to a non-default GODEBUG=randseednop=... setting. 359 360 /godebug/non-default-behavior/rsa1024min:events 361 The number of non-default behaviors executed by the crypto/rsa 362 package due to a non-default GODEBUG=rsa1024min=... setting. 363 364 /godebug/non-default-behavior/tarinsecurepath:events 365 The number of non-default behaviors executed by the archive/tar 366 package due to a non-default GODEBUG=tarinsecurepath=... 367 setting. 368 369 /godebug/non-default-behavior/tls10server:events 370 The number of non-default behaviors executed by the crypto/tls 371 package due to a non-default GODEBUG=tls10server=... setting. 372 373 /godebug/non-default-behavior/tls3des:events 374 The number of non-default behaviors executed by the crypto/tls 375 package due to a non-default GODEBUG=tls3des=... setting. 376 377 /godebug/non-default-behavior/tlsmaxrsasize:events 378 The number of non-default behaviors executed by the crypto/tls 379 package due to a non-default GODEBUG=tlsmaxrsasize=... setting. 380 381 /godebug/non-default-behavior/tlsrsakex:events 382 The number of non-default behaviors executed by the crypto/tls 383 package due to a non-default GODEBUG=tlsrsakex=... setting. 384 385 /godebug/non-default-behavior/tlssha1:events 386 The number of non-default behaviors executed by the crypto/tls 387 package due to a non-default GODEBUG=tlssha1=... setting. 388 389 /godebug/non-default-behavior/tlsunsafeekm:events 390 The number of non-default behaviors executed by the crypto/tls 391 package due to a non-default GODEBUG=tlsunsafeekm=... setting. 392 393 /godebug/non-default-behavior/updatemaxprocs:events 394 The number of non-default behaviors executed by the runtime 395 package due to a non-default GODEBUG=updatemaxprocs=... setting. 396 397 /godebug/non-default-behavior/winreadlinkvolume:events 398 The number of non-default behaviors executed by the os package 399 due to a non-default GODEBUG=winreadlinkvolume=... setting. 400 401 /godebug/non-default-behavior/winsymlink:events 402 The number of non-default behaviors executed by the os package 403 due to a non-default GODEBUG=winsymlink=... setting. 404 405 /godebug/non-default-behavior/x509keypairleaf:events 406 The number of non-default behaviors executed by the crypto/tls 407 package due to a non-default GODEBUG=x509keypairleaf=... 408 setting. 409 410 /godebug/non-default-behavior/x509negativeserial:events 411 The number of non-default behaviors executed by the crypto/x509 412 package due to a non-default GODEBUG=x509negativeserial=... 413 setting. 414 415 /godebug/non-default-behavior/x509rsacrt:events 416 The number of non-default behaviors executed by the crypto/x509 417 package due to a non-default GODEBUG=x509rsacrt=... setting. 418 419 /godebug/non-default-behavior/x509sha256skid:events 420 The number of non-default behaviors executed by the crypto/x509 421 package due to a non-default GODEBUG=x509sha256skid=... setting. 422 423 /godebug/non-default-behavior/x509usefallbackroots:events 424 The number of non-default behaviors executed by the crypto/x509 425 package due to a non-default GODEBUG=x509usefallbackroots=... 426 setting. 427 428 /godebug/non-default-behavior/x509usepolicies:events 429 The number of non-default behaviors executed by the crypto/x509 430 package due to a non-default GODEBUG=x509usepolicies=... 431 setting. 432 433 /godebug/non-default-behavior/zipinsecurepath:events 434 The number of non-default behaviors executed by the archive/zip 435 package due to a non-default GODEBUG=zipinsecurepath=... 436 setting. 437 438 /memory/classes/heap/free:bytes 439 Memory that is completely free and eligible to be returned to 440 the underlying system, but has not been. This metric is the 441 runtime's estimate of free address space that is backed by 442 physical memory. 443 444 /memory/classes/heap/objects:bytes 445 Memory occupied by live objects and dead objects that have not 446 yet been marked free by the garbage collector. 447 448 /memory/classes/heap/released:bytes 449 Memory that is completely free and has been returned to the 450 underlying system. This metric is the runtime's estimate of free 451 address space that is still mapped into the process, but is not 452 backed by physical memory. 453 454 /memory/classes/heap/stacks:bytes 455 Memory allocated from the heap that is reserved for stack space, 456 whether or not it is currently in-use. Currently, this 457 represents all stack memory for goroutines. It also includes all 458 OS thread stacks in non-cgo programs. Note that stacks may be 459 allocated differently in the future, and this may change. 460 461 /memory/classes/heap/unused:bytes 462 Memory that is reserved for heap objects but is not currently 463 used to hold heap objects. 464 465 /memory/classes/metadata/mcache/free:bytes 466 Memory that is reserved for runtime mcache structures, but not 467 in-use. 468 469 /memory/classes/metadata/mcache/inuse:bytes 470 Memory that is occupied by runtime mcache structures that are 471 currently being used. 472 473 /memory/classes/metadata/mspan/free:bytes 474 Memory that is reserved for runtime mspan structures, but not 475 in-use. 476 477 /memory/classes/metadata/mspan/inuse:bytes 478 Memory that is occupied by runtime mspan structures that are 479 currently being used. 480 481 /memory/classes/metadata/other:bytes 482 Memory that is reserved for or used to hold runtime metadata. 483 484 /memory/classes/os-stacks:bytes 485 Stack memory allocated by the underlying operating system. 486 In non-cgo programs this metric is currently zero. This may 487 change in the future.In cgo programs this metric includes 488 OS thread stacks allocated directly from the OS. Currently, 489 this only accounts for one stack in c-shared and c-archive build 490 modes, and other sources of stacks from the OS are not measured. 491 This too may change in the future. 492 493 /memory/classes/other:bytes 494 Memory used by execution trace buffers, structures for debugging 495 the runtime, finalizer and profiler specials, and more. 496 497 /memory/classes/profiling/buckets:bytes 498 Memory that is used by the stack trace hash map used for 499 profiling. 500 501 /memory/classes/total:bytes 502 All memory mapped by the Go runtime into the current process 503 as read-write. Note that this does not include memory mapped 504 by code called via cgo or via the syscall package. Sum of all 505 metrics in /memory/classes. 506 507 /sched/gomaxprocs:threads 508 The current runtime.GOMAXPROCS setting, or the number of 509 operating system threads that can execute user-level Go code 510 simultaneously. 511 512 /sched/goroutines:goroutines 513 Count of live goroutines. 514 515 /sched/latencies:seconds 516 Distribution of the time goroutines have spent in the scheduler 517 in a runnable state before actually running. Bucket counts 518 increase monotonically. 519 520 /sched/pauses/stopping/gc:seconds 521 Distribution of individual GC-related stop-the-world stopping 522 latencies. This is the time it takes from deciding to stop the 523 world until all Ps are stopped. This is a subset of the total 524 GC-related stop-the-world time (/sched/pauses/total/gc:seconds). 525 During this time, some threads may be executing. Bucket counts 526 increase monotonically. 527 528 /sched/pauses/stopping/other:seconds 529 Distribution of individual non-GC-related stop-the-world 530 stopping latencies. This is the time it takes from deciding 531 to stop the world until all Ps are stopped. This is a 532 subset of the total non-GC-related stop-the-world time 533 (/sched/pauses/total/other:seconds). During this time, some 534 threads may be executing. Bucket counts increase monotonically. 535 536 /sched/pauses/total/gc:seconds 537 Distribution of individual GC-related stop-the-world pause 538 latencies. This is the time from deciding to stop the world 539 until the world is started again. Some of this time is spent 540 getting all threads to stop (this is measured directly in 541 /sched/pauses/stopping/gc:seconds), during which some threads 542 may still be running. Bucket counts increase monotonically. 543 544 /sched/pauses/total/other:seconds 545 Distribution of individual non-GC-related stop-the-world 546 pause latencies. This is the time from deciding to stop the 547 world until the world is started again. Some of this time 548 is spent getting all threads to stop (measured directly in 549 /sched/pauses/stopping/other:seconds). Bucket counts increase 550 monotonically. 551 552 /sync/mutex/wait/total:seconds 553 Approximate cumulative time goroutines have spent blocked on a 554 sync.Mutex, sync.RWMutex, or runtime-internal lock. This metric 555 is useful for identifying global changes in lock contention. 556 Collect a mutex or block profile using the runtime/pprof package 557 for more detailed contention data. 558 */ 559 package metrics 560