1erl(1) User Commands erl(1)
2
6 erl - The Erlang emulator.
7
9 The erl program starts an Erlang runtime system. The exact details (for
10 example, whether erl is a script or a program and which other programs
11 it calls) are system-dependent.
12 Windows users probably want to use the werl program instead, which runs
14 in its own window with scrollbars and supports command-line editing.
15 The erl program on Windows provides no line editing in its shell, and
16 on Windows 95 there is no way to scroll back to text that has scrolled
17 off the screen. The erl program must be used, however, in pipelines or
18 if you want to redirect standard input or output.
19 Note:
21 As from ERTS 5.9 (Erlang/OTP R15B) the runtime system does by default
22 not bind schedulers to logical processors. For more information, see
23 system flag +sbt.
24
27 erl <arguments>
28 Starts an Erlang runtime system.
30 The arguments can be divided into emulator flags, flags, and
32 plain arguments:
33 * Any argument starting with character + is interpreted as an
35 emulator flag.
36 As indicated by the name, emulator flags control the behav‐
38 ior of the emulator.
39 * Any argument starting with character - (hyphen) is inter‐
41 preted as a flag, which is to be passed to the Erlang part
42 of the runtime system, more specifically to the init system
43 process, see init(3).
44 The init process itself interprets some of these flags, the
46 init flags. It also stores any remaining flags, the user
47 flags. The latter can be retrieved by calling init:get_argu‐
48 ment/1.
49 A small number of "-" flags exist, which now actually are
51 emulator flags, see the description below.
52 * Plain arguments are not interpreted in any way. They are
54 also stored by the init process and can be retrieved by
55 calling init:get_plain_arguments/0. Plain arguments can oc‐
56 cur before the first flag, or after a -- flag. Also, the
57 -extra flag causes everything that follows to become plain
58 arguments.
59 Examples:
61 % erl +W w -sname arnie +R 9 -s my_init -extra +bertie
63 (arnie@host)1> init:get_argument(sname).
64 {ok,[["arnie"]]}
65 (arnie@host)2> init:get_plain_arguments().
66 ["+bertie"]
67 Here +W w and +R 9 are emulator flags. -s my_init is an init
69 flag, interpreted by init. -sname arnie is a user flag, stored
70 by init. It is read by Kernel and causes the Erlang runtime sys‐
71 tem to become distributed. Finally, everything after -extra
72 (that is, +bertie) is considered as plain arguments.
73 % erl -myflag 1
75 1> init:get_argument(myflag).
76 {ok,[["1"]]}
77 2> init:get_plain_arguments().
78 []
79 Here the user flag -myflag 1 is passed to and stored by the init
81 process. It is a user-defined flag, presumably used by some
82 user-defined application.
83
85 In the following list, init flags are marked "(init flag)". Unless oth‐
86 erwise specified, all other flags are user flags, for which the values
87 can be retrieved by calling init:get_argument/1. Notice that the list
88 of user flags is not exhaustive, there can be more application-specific
89 flags that instead are described in the corresponding application docu‐
90 mentation.
91 -- (init flag):
93 Everything following -- up to the next flag (-flag or +flag) is
94 considered plain arguments and can be retrieved using
95 init:get_plain_arguments/0.
96 -Application Par Val:
98 Sets the application configuration parameter Par to the value Val
99 for the application Application; see app(4) and application(3).
100 -args_file FileName:
102 Command-line arguments are read from the file FileName. The argu‐
103 ments read from the file replace flag '-args_file FileName' on the
104 resulting command line.
105 The file FileName is to be a plain text file and can contain com‐
107 ments and command-line arguments. A comment begins with a # charac‐
108 ter and continues until the next end of line character. Backslash
109 (\\) is used as quoting character. All command-line arguments ac‐
110 cepted by erl are allowed, also flag -args_file FileName. Be care‐
111 ful not to cause circular dependencies between files containing
112 flag -args_file, though.
113 The flag -extra is treated in special way. Its scope ends at the
115 end of the file. Arguments following an -extra flag are moved on
116 the command line into the -extra section, that is, the end of the
117 command line following after an -extra flag.
118 -async_shell_start:
120 The initial Erlang shell does not read user input until the system
121 boot procedure has been completed (Erlang/OTP 5.4 and later). This
122 flag disables the start synchronization feature and lets the shell
123 start in parallel with the rest of the system.
124 -boot File:
126 Specifies the name of the boot file, File.boot, which is used to
127 start the system; see init(3). Unless File contains an absolute
128 path, the system searches for File.boot in the current and
129 $ROOT/bin directories.
130 Defaults to $ROOT/bin/start.boot.
132 -boot_var Var Dir:
134 If the boot script contains a path variable Var other than $ROOT,
135 this variable is expanded to Dir. Used when applications are in‐
136 stalled in another directory than $ROOT/lib; see sys‐
137 tools:make_script/1,2 in SASL.
138 -code_path_cache:
140 Enables the code path cache of the code server; see code(3).
141 -compile Mod1 Mod2 ...:
143 Compiles the specified modules and then terminates (with non-zero
144 exit code if the compilation of some file did not succeed). Implies
145 -noinput.
146 Not recommended; use erlc instead.
148 -config Config [Config]:
150 Specifies the name of one or more configuration files, Config.con‐
151 fig, which is used to configure applications; see app(4) and appli‐
152 cation(3).
153 -connect_all false:
155 If this flag is present, global does not maintain a fully connected
156 network of distributed Erlang nodes, and then global name registra‐
157 tion cannot be used; see global(3).
158 -cookie Cookie:
160 Obsolete flag without any effect and common misspelling for -set‐
161 cookie. Use -setcookie instead.
162 -detached:
164 Starts the Erlang runtime system detached from the system console.
165 Useful for running daemons and backgrounds processes. Implies
166 -noinput.
167 -emu_args:
169 Useful for debugging. Prints the arguments sent to the emulator.
170 -emu_type Type:
172 Start an emulator of a different type. For example, to start the
173 lock-counter emualator, use -emu_type lcnt. (The emulator must al‐
174 ready be built. Use the configure option --enable-lock-counter to
175 build the lock-counter emulator.)
176 -env Variable Value:
178 Sets the host OS environment variable Variable to the value Value
179 for the Erlang runtime system. Example:
180 % erl -env DISPLAY gin:0
182 In this example, an Erlang runtime system is started with environ‐
184 ment variable DISPLAY set to gin:0.
185 -epmd_module Module (init flag):
187 Configures the module responsible to communicate to epmd. Defaults
188 to erl_epmd.
189 -erl_epmd_port Port (init flag):
191 Configures the port used by erl_epmd to listen for connection and
192 connect to other nodes. See erl_epmd for more details. Defaults to
193 0.
194 -eval Expr (init flag):
196 Makes init evaluate the expression Expr; see init(3).
197 -extra (init flag):
199 Everything following -extra is considered plain arguments and can
200 be retrieved using init:get_plain_arguments/0.
201 -heart:
203 Starts heartbeat monitoring of the Erlang runtime system; see
204 heart(3).
205 -hidden:
207 Starts the Erlang runtime system as a hidden node, if it is run as
208 a distributed node. Hidden nodes always establish hidden connec‐
209 tions to all other nodes except for nodes in the same global group.
210 Hidden connections are not published on any of the connected nodes,
211 that is, none of the connected nodes are part of the result from
212 nodes/0 on the other node. See also hidden global groups;
213 global_group(3).
214 -hosts Hosts:
216 Specifies the IP addresses for the hosts on which Erlang boot
217 servers are running, see erl_boot_server(3). This flag is mandatory
218 if flag -loader inet is present.
219 The IP addresses must be specified in the standard form (four deci‐
221 mal numbers separated by periods, for example, "150.236.20.74".
222 Hosts names are not acceptable, but a broadcast address (preferably
223 limited to the local network) is.
224 -id Id:
226 Specifies the identity of the Erlang runtime system. If it is run
227 as a distributed node, Id must be identical to the name supplied
228 together with flag -sname or -name.
229 -init_debug:
231 Makes init write some debug information while interpreting the boot
232 script.
233 -instr (emulator flag):
235 Selects an instrumented Erlang runtime system (virtual machine) to
236 run, instead of the ordinary one. When running an instrumented run‐
237 time system, some resource usage data can be obtained and analyzed
238 using the instrument module. Functionally, it behaves exactly like
239 an ordinary Erlang runtime system.
240 -loader Loader:
242 Specifies the method used by erl_prim_loader to load Erlang modules
243 into the system; see erl_prim_loader(3). Two Loader methods are
244 supported:
245 * efile, which means use the local file system, this is the de‐
247 fault.
248 * inet, which means use a boot server on another machine. The flags
250 -id, -hosts and -setcookie must also be specified.
251 If Loader is something else, the user-supplied Loader port program
253 is started.
254 -make:
256 Makes the Erlang runtime system invoke make:all() in the current
257 working directory and then terminate; see make(3). Implies -noin‐
258 put.
259 -man Module:
261 Displays the manual page for the Erlang module Module. Only sup‐
262 ported on Unix.
263 -mode interactive | embedded:
265 Modules are auto loaded when they are first referenced if the run‐
266 time system runs in interactive mode, which is the default. In em‐
267 bedded mode modules are not auto loaded. The latter is recommended
268 when the boot script preloads all modules, as conventionally hap‐
269 pens in OTP releases. See code(3).
270 -name Name:
272 Makes the Erlang runtime system into a distributed node. This flag
273 invokes all network servers necessary for a node to become distrib‐
274 uted; see net_kernel(3). It is also ensured that epmd runs on the
275 current host before Erlang is started; see epmd(1).and the
276 -start_epmd option.
277 The node name will be Name@Host, where Host is the fully qualified
279 host name of the current host. For short names, use flag -sname in‐
280 stead.
281 If Name is set to undefined the node will be started in a special
283 mode optimized to be the temporary client of another node. When en‐
284 abled the node will request a dynamic node name from the first node
285 it connects to. In addition these distribution settings will be
286 set:
287 -dist_listen false -hidden -dist_auto_connect never
289 Because -dist_auto_connect is set to never, the system will have to
291 manually call net_kernel:connect_node/1 in order to start the dis‐
292 tribution. If the distribution channel is closed, when a node uses
293 a dynamic node name, the node will stop the distribution and a new
294 call to net_kernel:connect_node/1 has to be made. Note that the
295 node name may change if the distribution is dropped and then set up
296 again.
297 Note:
299 The dynamic node name feature is supported from OTP 23. Both the tem‐
300 porary client node and the first connected peer node (supplying the
301 dynamic node name) must be at least OTP 23 for it to work.
302 Warning:
305 Starting a distributed node without also specifying -proto_dist
306 inet_tls will expose the node to attacks that may give the attacker
307 complete access to the node and in extension the cluster. When using
308 un-secure distributed nodes, make sure that the network is configured
309 to keep potential attackers out.
310 -noinput:
313 Ensures that the Erlang runtime system never tries to read any in‐
314 put. Implies -noshell.
315 -noshell:
317 Starts an Erlang runtime system with no shell. This flag makes it
318 possible to have the Erlang runtime system as a component in a se‐
319 ries of Unix pipes.
320 -nostick:
322 Disables the sticky directory facility of the Erlang code server;
323 see code(3).
324 -oldshell:
326 Invokes the old Erlang shell from Erlang/OTP 3.3. The old shell can
327 still be used.
328 -pa Dir1 Dir2 ...:
330 Adds the specified directories to the beginning of the code path,
331 similar to code:add_pathsa/1. Note that the order of the given di‐
332 rectories will be reversed in the resulting path.
333 As an alternative to -pa, if several directories are to be
335 prepended to the code path and the directories have a common parent
336 directory, that parent directory can be specified in environment
337 variable ERL_LIBS; see code(3).
338 -pz Dir1 Dir2 ...:
340 Adds the specified directories to the end of the code path, similar
341 to code:add_pathsz/1; see code(3).
342 -path Dir1 Dir2 ...:
344 Replaces the path specified in the boot script; see script(4).
345 -proto_dist Proto:
347 Specifies a protocol for Erlang distribution:
350 inet_tcp:
352 TCP over IPv4 (the default)
353 inet_tls:
355 Distribution over TLS/SSL, See the Using SSL for Erlang Distri‐
356 bution User's Guide for details on how to setup a secure distrib‐
357 uted node.
358 inet6_tcp:
360 TCP over IPv6
361 For example, to start up IPv6 distributed nodes:
363 % erl -name test@ipv6node.example.com -proto_dist inet6_tcp
365 -remsh Node:
367 Starts Erlang with a remote shell connected to Node.
368 If no -name or -sname is given the node will be started using
370 -sname undefined. If Node does not contain a hostname, one is auto‐
371 matically taken from -name or -sname
372 Note:
374 Before OTP-23 the user needed to supply a valid -sname or -name for
375 -remsh to work. This is still the case if the target node is not run‐
376 ning OTP-23 or later.
377 -rsh Program:
380 Specifies an alternative to ssh for starting a slave node on a re‐
381 mote host; see slave(3).
382 -run Mod [Func [Arg1, Arg2, ...]] (init flag):
384 Makes init call the specified function. Func defaults to start. If
385 no arguments are provided, the function is assumed to be of arity
386 0. Otherwise it is assumed to be of arity 1, taking the list
387 [Arg1,Arg2,...] as argument. All arguments are passed as strings.
388 See init(3).
389 -s Mod [Func [Arg1, Arg2, ...]] (init flag):
391 Makes init call the specified function. Func defaults to start. If
392 no arguments are provided, the function is assumed to be of arity
393 0. Otherwise it is assumed to be of arity 1, taking the list
394 [Arg1,Arg2,...] as argument. All arguments are passed as atoms. See
395 init(3).
396 -setcookie Cookie:
398 Sets the magic cookie of the node to Cookie; see er‐
399 lang:set_cookie/2.
400 -shutdown_time Time:
402 Specifies how long time (in milliseconds) the init process is al‐
403 lowed to spend shutting down the system. If Time milliseconds have
404 elapsed, all processes still existing are killed. Defaults to in‐
405 finity.
406 -sname Name:
408 Makes the Erlang runtime system into a distributed node, similar to
409 -name, but the host name portion of the node name Name@Host will be
410 the short name, not fully qualified.
411 This is sometimes the only way to run distributed Erlang if the Do‐
413 main Name System (DNS) is not running. No communication can exist
414 between nodes running with flag -sname and those running with flag
415 -name, as node names must be unique in distributed Erlang systems.
416 Warning:
418 Starting a distributed node without also specifying -proto_dist
419 inet_tls will expose the node to attacks that may give the attacker
420 complete access to the node and in extension the cluster. When using
421 un-secure distributed nodes, make sure that the network is configured
422 to keep potential attackers out.
423 -start_epmd true | false:
426 Specifies whether Erlang should start epmd on startup. By default
427 this is true, but if you prefer to start epmd manually, set this to
428 false.
429 This only applies if Erlang is started as a distributed node, i.e.
431 if -name or -sname is specified. Otherwise, epmd is not started
432 even if -start_epmd true is given.
433 Note that a distributed node will fail to start if epmd is not run‐
435 ning.
436 -version (emulator flag):
438 Makes the emulator print its version number. The same as erl +V.
439
441 erl invokes the code for the Erlang emulator (virtual machine), which
442 supports the following flags:
443 +a size:
445 Suggested stack size, in kilowords, for threads in the async thread
446 pool. Valid range is 16-8192 kilowords. The default suggested stack
447 size is 16 kilowords, that is, 64 kilobyte on 32-bit architectures.
448 This small default size has been chosen because the number of async
449 threads can be large. The default size is enough for drivers deliv‐
450 ered with Erlang/OTP, but might not be large enough for other dy‐
451 namically linked-in drivers that use the driver_async() functional‐
452 ity. Notice that the value passed is only a suggestion, and it can
453 even be ignored on some platforms.
454 +A size:
456 Sets the number of threads in async thread pool. Valid range is
457 1-1024. The async thread pool is used by linked-in drivers to han‐
458 dle work that may take a very long time. Since OTP 21 there are
459 very few linked-in drivers in the default Erlang/OTP distribution
460 that uses the async thread pool. Most of them have been migrated to
461 dirty IO schedulers. Defaults to 1.
462 +B [c | d | i]:
464 Option c makes Ctrl-C interrupt the current shell instead of invok‐
465 ing the emulator break handler. Option d (same as specifying +B
466 without an extra option) disables the break handler. Option i makes
467 the emulator ignore any break signal.
468 If option c is used with oldshell on Unix, Ctrl-C will restart the
470 shell process rather than interrupt it.
471 Notice that on Windows, this flag is only applicable for werl, not
473 erl (oldshell). Notice also that Ctrl-Break is used instead of
474 Ctrl-C on Windows.
475 +c true | false:
477 Enables or disables time correction:
478 true:
480 Enables time correction. This is the default if time correction
481 is supported on the specific platform.
482 false:
484 Disables time correction.
485 For backward compatibility, the boolean value can be omitted. This
487 is interpreted as +c false.
488 +C no_time_warp | single_time_warp | multi_time_warp:
490 Sets time warp mode:
491 no_time_warp:
493 No time warp mode (the default)
494 single_time_warp:
496 Single time warp mode
497 multi_time_warp:
499 Multi-time warp mode
500 +d:
502 If the emulator detects an internal error (or runs out of memory),
503 it, by default, generates both a crash dump and a core dump. The
504 core dump is, however, not very useful as the content of process
505 heaps is destroyed by the crash dump generation.
506 Option +d instructs the emulator to produce only a core dump and no
508 crash dump if an internal error is detected.
509 Calling erlang:halt/1 with a string argument still produces a crash
511 dump. On Unix systems, sending an emulator process a SIGUSR1 signal
512 also forces a crash dump.
513 +dcg DecentralizedCounterGroupsLimit:
515 Limits the number of decentralized counter groups used by decen‐
516 tralized counters optimized for update operations in the Erlang
517 runtime system. By default, the limit is 256.
518 When the number of schedulers is less than or equal to the limit,
520 each scheduler has its own group. When the number of schedulers is
521 larger than the groups limit, schedulers share groups. Shared
522 groups degrade the performance for updating counters while many
523 reader groups degrade the performance for reading counters. So, the
524 limit is a tradeoff between performance for update operations and
525 performance for read operations. Each group consumes 64 bytes in
526 each counter.
527 Notice that a runtime system using decentralized counter groups
529 benefits from binding schedulers to logical processors, as the
530 groups are distributed better between schedulers with this option.
531 This option only affects decentralized counters used for the coun‐
533 ters that are keeping track of the memory consumption and the num‐
534 ber of terms in ETS tables of type ordered_set with the write_con‐
535 currency option activated.
536 +e Number:
538 Sets the maximum number of ETS tables. This limit is partially ob‐
539 solete.
540 +ec:
542 Forces option compressed on all ETS tables. Only intended for test
543 and evaluation.
544 +fnl:
546 The virtual machine works with filenames as if they are encoded us‐
547 ing the ISO Latin-1 encoding, disallowing Unicode characters with
548 code points > 255.
549 For more information about Unicode filenames, see section Unicode
551 Filenames in the STDLIB User's Guide. Notice that this value also
552 applies to command-line parameters and environment variables (see
553 section Unicode in Environment and Parameters in the STDLIB User's
554 Guide).
555 +fnu[{w|i|e}]:
557 The virtual machine works with filenames as if they are encoded us‐
558 ing UTF-8 (or some other system-specific Unicode encoding). This is
559 the default on operating systems that enforce Unicode encoding,
560 that is, Windows MacOS X and Android.
561 The +fnu switch can be followed by w, i, or e to control how
563 wrongly encoded filenames are to be reported:
564 * w means that a warning is sent to the error_logger whenever a
566 wrongly encoded filename is "skipped" in directory listings. This
567 is the default.
568 * i means that those wrongly encoded filenames are silently ig‐
570 nored.
571 * e means that the API function returns an error whenever a wrongly
573 encoded filename (or directory name) is encountered.
574 Notice that file:read_link/1 always returns an error if the link
576 points to an invalid filename.
577 For more information about Unicode filenames, see section Unicode
579 Filenames in the STDLIB User's Guide. Notice that this value also
580 applies to command-line parameters and environment variables (see
581 section Unicode in Environment and Parameters in the STDLIB User's
582 Guide).
583 +fna[{w|i|e}]:
585 Selection between +fnl and +fnu is done based on the current locale
586 settings in the OS. This means that if you have set your terminal
587 for UTF-8 encoding, the filesystem is expected to use the same en‐
588 coding for filenames. This is the default on all operating systems,
589 except Android, MacOS X and Windows.
590 The +fna switch can be followed by w, i, or e. This has effect if
592 the locale settings cause the behavior of +fnu to be selected; see
593 the description of +fnu above. If the locale settings cause the be‐
594 havior of +fnl to be selected, then w, i, or e have no effect.
595 For more information about Unicode filenames, see section Unicode
597 Filenames in the STDLIB User's Guide. Notice that this value also
598 applies to command-line parameters and environment variables (see
599 section Unicode in Environment and Parameters in the STDLIB User's
600 Guide).
601 +hms Size:
603 Sets the default heap size of processes to the size Size.
604 +hmbs Size:
606 Sets the default binary virtual heap size of processes to the size
607 Size.
608 +hmax Size:
610 Sets the default maximum heap size of processes to the size Size.
611 Defaults to 0, which means that no maximum heap size is used. For
612 more information, see process_flag(max_heap_size, MaxHeapSize).
613 +hmaxel true|false:
615 Sets whether to send an error logger message or not for processes
616 reaching the maximum heap size. Defaults to true. For more informa‐
617 tion, see process_flag(max_heap_size, MaxHeapSize).
618 +hmaxk true|false:
620 Sets whether to kill processes reaching the maximum heap size or
621 not. Default to true. For more information, see
622 process_flag(max_heap_size, MaxHeapSize).
623 +hpds Size:
625 Sets the initial process dictionary size of processes to the size
626 Size.
627 +hmqd off_heap|on_heap:
629 Sets the default value of the message_queue_data process flag. De‐
630 faults to on_heap. If +hmqd is not passed, on_heap will be the de‐
631 fault. For more information, see process_flag(message_queue_data,
632 MQD).
633 +IOp PollSets:
635 Sets the number of IO pollsets to use when polling for I/O. This
636 option is only used on platforms that support concurrent updates of
637 a pollset, otherwise the same number of pollsets are used as IO
638 poll threads. The default is 1.
639 +IOt PollThreads:
641 Sets the number of IO poll threads to use when polling for I/O. The
642 maximum number of poll threads allowed is 1024. The default is 1.
643 A good way to check if more IO poll threads are needed is to use
645 microstate accounting and see what the load of the IO poll thread
646 is. If it is high it could be a good idea to add more threads.
647 +IOPp PollSetsPercentage:
649 Similar to +IOp but uses percentages to set the number of IO
650 pollsets to create, based on the number of poll threads configured.
651 If both +IOPp and +IOp are used, +IOPp is ignored.
652 +IOPt PollThreadsPercentage:
654 Similar to +IOt but uses percentages to set the number of IO poll
655 threads to create, based on the number of schedulers configured. If
656 both +IOPt and +IOt are used, +IOPt is ignored.
657 +L:
659 Prevents loading information about source filenames and line num‐
660 bers. This saves some memory, but exceptions do not contain infor‐
661 mation about the filenames and line numbers.
662 +MFlag Value:
664 Memory allocator-specific flags. For more information, see erts_al‐
665 loc(3).
666 +pc Range:
668 Sets the range of characters that the system considers printable in
669 heuristic detection of strings. This typically affects the shell,
670 debugger, and io:format functions (when ~tp is used in the format
671 string).
672 Two values are supported for Range:
674 latin1:
676 The default. Only characters in the ISO Latin-1 range can be con‐
677 sidered printable. This means that a character with a code point
678 > 255 is never considered printable and that lists containing
679 such characters are displayed as lists of integers rather than
680 text strings by tools.
681 unicode:
683 All printable Unicode characters are considered when determining
684 if a list of integers is to be displayed in string syntax. This
685 can give unexpected results if, for example, your font does not
686 cover all Unicode characters.
687 See also io:printable_range/0 in STDLIB.
689 +P Number:
691 Sets the maximum number of simultaneously existing processes for
692 this system if a Number is passed as value. Valid range for Number
693 is [1024-134217727]
694 NOTE: The actual maximum chosen may be much larger than the Number
696 passed. Currently the runtime system often, but not always, chooses
697 a value that is a power of 2. This might, however, be changed in
698 the future. The actual value chosen can be checked by calling er‐
699 lang:system_info(process_limit).
700 The default value is 262144
702 +Q Number:
704 Sets the maximum number of simultaneously existing ports for this
705 system if a Number is passed as value. Valid range for Number is
706 [1024-134217727]
707 NOTE: The actual maximum chosen may be much larger than the actual
709 Number passed. Currently the runtime system often, but not always,
710 chooses a value that is a power of 2. This might, however, be
711 changed in the future. The actual value chosen can be checked by
712 calling erlang:system_info(port_limit).
713 The default value used is normally 65536. However, if the runtime
715 system is able to determine maximum amount of file descriptors that
716 it is allowed to open and this value is larger than 65536, the cho‐
717 sen value will increased to a value larger or equal to the maximum
718 amount of file descriptors that can be opened.
719 On Windows the default value is set to 8196 because the normal OS
721 limitations are set higher than most machines can handle.
722 +R ReleaseNumber:
724 Sets the compatibility mode.
725 The distribution mechanism is not backward compatible by default.
727 This flag sets the emulator in compatibility mode with an earlier
728 Erlang/OTP release ReleaseNumber. The release number must be in the
729 range <current release>-2..<current release>. This limits the emu‐
730 lator, making it possible for it to communicate with Erlang nodes
731 (as well as C- and Java nodes) running that earlier release.
732 Note:
734 Ensure that all nodes (Erlang-, C-, and Java nodes) of a distributed
735 Erlang system is of the same Erlang/OTP release, or from two differ‐
736 ent Erlang/OTP releases X and Y, where all Y nodes have compatibility
737 mode X.
738 +r:
741 Forces ETS memory block to be moved on realloc.
742 +rg ReaderGroupsLimit:
744 Limits the number of reader groups used by read/write locks opti‐
745 mized for read operations in the Erlang runtime system. By default
746 the reader groups limit is 64.
747 When the number of schedulers is less than or equal to the reader
749 groups limit, each scheduler has its own reader group. When the
750 number of schedulers is larger than the reader groups limit, sched‐
751 ulers share reader groups. Shared reader groups degrade read lock
752 and read unlock performance while many reader groups degrade write
753 lock performance. So, the limit is a tradeoff between performance
754 for read operations and performance for write operations. Each
755 reader group consumes 64 byte in each read/write lock.
756 Notice that a runtime system using shared reader groups benefits
758 from binding schedulers to logical processors, as the reader groups
759 are distributed better between schedulers.
760 +S Schedulers:SchedulerOnline:
762 Sets the number of scheduler threads to create and scheduler
763 threads to set online. The maximum for both values is 1024. If the
764 Erlang runtime system is able to determine the number of logical
765 processors configured and logical processors available, Schedulers
766 defaults to logical processors configured, and SchedulersOnline de‐
767 faults to logical processors available; otherwise the default val‐
768 ues are 1. If the emulator detects that it is subject to a CPU
769 quota, the default value for SchedulersOnline will be limited ac‐
770 cordingly.
771 Schedulers can be omitted if :SchedulerOnline is not and con‐
773 versely. The number of schedulers online can be changed at runtime
774 through erlang:system_flag(schedulers_online, SchedulersOnline).
775 If Schedulers or SchedulersOnline is specified as a negative num‐
777 ber, the value is subtracted from the default number of logical
778 processors configured or logical processors available, respec‐
779 tively.
780 Specifying value 0 for Schedulers or SchedulersOnline resets the
782 number of scheduler threads or scheduler threads online, respec‐
783 tively, to its default value.
784 +SP SchedulersPercentage:SchedulersOnlinePercentage:
786 Similar to +S but uses percentages to set the number of scheduler
787 threads to create, based on logical processors configured, and
788 scheduler threads to set online, based on logical processors avail‐
789 able. Specified values must be > 0. For example, +SP 50:25 sets the
790 number of scheduler threads to 50% of the logical processors con‐
791 figured, and the number of scheduler threads online to 25% of the
792 logical processors available. SchedulersPercentage can be omitted
793 if :SchedulersOnlinePercentage is not and conversely. The number of
794 schedulers online can be changed at runtime through erlang:sys‐
795 tem_flag(schedulers_online, SchedulersOnline).
796 This option interacts with +S settings. For example, on a system
798 with 8 logical cores configured and 8 logical cores available, the
799 combination of the options +S 4:4 +SP 50:25 (in either order) re‐
800 sults in 2 scheduler threads (50% of 4) and 1 scheduler thread on‐
801 line (25% of 4).
802 +SDcpu DirtyCPUSchedulers:DirtyCPUSchedulersOnline:
804 Sets the number of dirty CPU scheduler threads to create and dirty
805 CPU scheduler threads to set online. The maximum for both values is
806 1024, and each value is further limited by the settings for normal
807 schedulers:
808 * The number of dirty CPU scheduler threads created cannot exceed
810 the number of normal scheduler threads created.
811 * The number of dirty CPU scheduler threads online cannot exceed
813 the number of normal scheduler threads online.
814 For details, see the +S and +SP. By default, the number of dirty
816 CPU scheduler threads created equals the number of normal scheduler
817 threads created, and the number of dirty CPU scheduler threads on‐
818 line equals the number of normal scheduler threads online. DirtyC‐
819 PUSchedulers can be omitted if :DirtyCPUSchedulersOnline is not and
820 conversely. The number of dirty CPU schedulers online can be
821 changed at runtime through erlang:system_flag(dirty_cpu_sched‐
822 ulers_online, DirtyCPUSchedulersOnline).
823 The amount of dirty CPU schedulers is limited by the amount of nor‐
825 mal schedulers in order to limit the effect on processes executing
826 on ordinary schedulers. If the amount of dirty CPU schedulers was
827 allowed to be unlimited, dirty CPU bound jobs would potentially
828 starve normal jobs.
829 Typical users of the dirty CPU schedulers are large garbage collec‐
831 tions, json protocol encode/decoders written as nifs and matrix ma‐
832 nipulation libraries.
833 You can use msacc(3) in order to see the current load of the dirty
835 CPU schedulers threads and adjust the number used accordingly.
836 +SDPcpu DirtyCPUSchedulersPercentage:DirtyCPUSchedulersOnlinePercent‐
838 age:
839 Similar to +SDcpu but uses percentages to set the number of dirty
840 CPU scheduler threads to create and the number of dirty CPU sched‐
841 uler threads to set online. Specified values must be > 0. For exam‐
842 ple, +SDPcpu 50:25 sets the number of dirty CPU scheduler threads
843 to 50% of the logical processors configured and the number of dirty
844 CPU scheduler threads online to 25% of the logical processors
845 available. DirtyCPUSchedulersPercentage can be omitted if :DirtyC‐
846 PUSchedulersOnlinePercentage is not and conversely. The number of
847 dirty CPU schedulers online can be changed at runtime through er‐
848 lang:system_flag(dirty_cpu_schedulers_online, DirtyCPUSchedulersOn‐
849 line).
850 This option interacts with +SDcpu settings. For example, on a sys‐
852 tem with 8 logical cores configured and 8 logical cores available,
853 the combination of the options +SDcpu 4:4 +SDPcpu 50:25 (in either
854 order) results in 2 dirty CPU scheduler threads (50% of 4) and 1
855 dirty CPU scheduler thread online (25% of 4).
856 +SDio DirtyIOSchedulers:
858 Sets the number of dirty I/O scheduler threads to create. Valid
859 range is 1-1024. By default, the number of dirty I/O scheduler
860 threads created is 10.
861 The amount of dirty IO schedulers is not limited by the amount of
863 normal schedulers like the amount of dirty CPU schedulers. This
864 since only I/O bound work is expected to execute on dirty I/O
865 schedulers. If the user should schedule CPU bound jobs on dirty I/O
866 schedulers, these jobs might starve ordinary jobs executing on or‐
867 dinary schedulers.
868 Typical users of the dirty IO schedulers are reading and writing to
870 files.
871 You can use msacc(3) in order to see the current load of the dirty
873 IO schedulers threads and adjust the number used accordingly.
874 +sFlag Value:
876 Scheduling specific flags.
877 +sbt BindType:
879 Sets scheduler bind type.
880 Schedulers can also be bound using flag +stbt. The only differ‐
882 ence between these two flags is how the following errors are han‐
883 dled:
884 * Binding of schedulers is not supported on the specific plat‐
886 form.
887 * No available CPU topology. That is, the runtime system was not
889 able to detect the CPU topology automatically, and no user-de‐
890 fined CPU topology was set.
891 If any of these errors occur when +sbt has been passed, the run‐
893 time system prints an error message, and refuses to start. If any
894 of these errors occur when +stbt has been passed, the runtime
895 system silently ignores the error, and start up using unbound
896 schedulers.
897 Valid BindTypes:
899 u:
901 unbound - Schedulers are not bound to logical processors, that
902 is, the operating system decides where the scheduler threads
903 execute, and when to migrate them. This is the default.
904 ns:
906 no_spread - Schedulers with close scheduler identifiers are
907 bound as close as possible in hardware.
908 ts:
910 thread_spread - Thread refers to hardware threads (such as In‐
911 tel's hyper-threads). Schedulers with low scheduler identi‐
912 fiers, are bound to the first hardware thread of each core,
913 then schedulers with higher scheduler identifiers are bound to
914 the second hardware thread of each core,and so on.
915 ps:
917 processor_spread - Schedulers are spread like thread_spread,
918 but also over physical processor chips.
919 s:
921 spread - Schedulers are spread as much as possible.
922 nnts:
924 no_node_thread_spread - Like thread_spread, but if multiple
925 Non-Uniform Memory Access (NUMA) nodes exist, schedulers are
926 spread over one NUMA node at a time, that is, all logical pro‐
927 cessors of one NUMA node are bound to schedulers in sequence.
928 nnps:
930 no_node_processor_spread - Like processor_spread, but if multi‐
931 ple NUMA nodes exist, schedulers are spread over one NUMA node
932 at a time, that is, all logical processors of one NUMA node are
933 bound to schedulers in sequence.
934 tnnps:
936 thread_no_node_processor_spread - A combination of
937 thread_spread, and no_node_processor_spread. Schedulers are
938 spread over hardware threads across NUMA nodes, but schedulers
939 are only spread over processors internally in one NUMA node at
940 a time.
941 db:
943 default_bind - Binds schedulers the default way. Defaults to
944 thread_no_node_processor_spread (which can change in the fu‐
945 ture).
946 Binding of schedulers is only supported on newer Linux, Solaris,
948 FreeBSD, and Windows systems.
949 If no CPU topology is available when flag +sbt is processed and
951 BindType is any other type than u, the runtime system fails to
952 start. CPU topology can be defined using flag +sct. Notice that
953 flag +sct can have to be passed before flag +sbt on the command
954 line (if no CPU topology has been automatically detected).
955 The runtime system does by default not bind schedulers to logical
957 processors.
958 Note:
960 If the Erlang runtime system is the only operating system process
961 that binds threads to logical processors, this improves the perfor‐
962 mance of the runtime system. However, if other operating system
963 processes (for example another Erlang runtime system) also bind
964 threads to logical processors, there can be a performance penalty
965 instead. This performance penalty can sometimes be severe. If so,
966 you are advised not to bind the schedulers.
967 How schedulers are bound matters. For example, in situations when
970 there are fewer running processes than schedulers online, the
971 runtime system tries to migrate processes to schedulers with low
972 scheduler identifiers. The more the schedulers are spread over
973 the hardware, the more resources are available to the runtime
974 system in such situations.
975 Note:
977 If a scheduler fails to bind, this is often silently ignored, as it
978 is not always possible to verify valid logical processor identi‐
979 fiers. If an error is reported, it is reported to the error_logger.
980 If you want to verify that the schedulers have bound as requested,
981 call erlang:system_info(scheduler_bindings).
982 +sbwt none|very_short|short|medium|long|very_long:
985 Sets scheduler busy wait threshold. Defaults to medium. The
986 threshold determines how long schedulers are to busy wait when
987 running out of work before going to sleep.
988 Note:
990 This flag can be removed or changed at any time without prior no‐
991 tice.
992 +sbwtdcpu none|very_short|short|medium|long|very_long:
995 As +sbwt but affects dirty CPU schedulers. Defaults to short.
996 Note:
998 This flag can be removed or changed at any time without prior no‐
999 tice.
1000 +sbwtdio none|very_short|short|medium|long|very_long:
1003 As +sbwt but affects dirty IO schedulers. Defaults to short.
1004 Note:
1006 This flag can be removed or changed at any time without prior no‐
1007 tice.
1008 +scl true|false:
1011 Enables or disables scheduler compaction of load. By default
1012 scheduler compaction of load is enabled. When enabled, load bal‐
1013 ancing strives for a load distribution, which causes as many
1014 scheduler threads as possible to be fully loaded (that is, not
1015 run out of work). This is accomplished by migrating load (for ex‐
1016 ample, runnable processes) into a smaller set of schedulers when
1017 schedulers frequently run out of work. When disabled, the fre‐
1018 quency with which schedulers run out of work is not taken into
1019 account by the load balancing logic.
1020 +scl false is similar to +sub true, but +sub true also balances
1022 scheduler utilization between schedulers.
1023 +sct CpuTopology:
1025 * <Id> = integer(); when 0 =< <Id> =< 65535
1028 * <IdRange> = <Id>-<Id>
1030 * <IdOrIdRange> = <Id> | <IdRange>
1032 * <IdList> = <IdOrIdRange>,<IdOrIdRange> | <IdOrIdRange>
1034 * <LogicalIds> = L<IdList>
1036 * <ThreadIds> = T<IdList> | t<IdList>
1038 * <CoreIds> = C<IdList> | c<IdList>
1040 * <ProcessorIds> = P<IdList> | p<IdList>
1042 * <NodeIds> = N<IdList> | n<IdList>
1044 * <IdDefs> = <LogicalIds><ThreadIds><CoreIds><Proces‐
1046 sorIds><NodeIds> | <LogicalIds><ThreadIds><Cor‐
1047 eIds><NodeIds><ProcessorIds>
1048 * CpuTopology = <IdDefs>:<IdDefs> | <IdDefs>
1050 Sets a user-defined CPU topology. The user-defined CPU topology
1052 overrides any automatically detected CPU topology. The CPU topol‐
1053 ogy is used when binding schedulers to logical processors.
1054 Uppercase letters signify real identifiers and lowercase letters
1056 signify fake identifiers only used for description of the topol‐
1057 ogy. Identifiers passed as real identifiers can be used by the
1058 runtime system when trying to access specific hardware; if they
1059 are incorrect the behavior is undefined. Faked logical CPU iden‐
1060 tifiers are not accepted, as there is no point in defining the
1061 CPU topology without real logical CPU identifiers. Thread, core,
1062 processor, and node identifiers can be omitted. If omitted, the
1063 thread ID defaults to t0, the core ID defaults to c0, the proces‐
1064 sor ID defaults to p0, and the node ID is left undefined. Either
1065 each logical processor must belong to only one NUMA node, or no
1066 logical processors must belong to any NUMA nodes.
1067 Both increasing and decreasing <IdRange>s are allowed.
1069 NUMA node identifiers are system wide. That is, each NUMA node on
1071 the system must have a unique identifier. Processor identifiers
1072 are also system wide. Core identifiers are processor wide. Thread
1073 identifiers are core wide.
1074 The order of the identifier types implies the hierarchy of the
1076 CPU topology. The valid orders are as follows:
1077 * <LogicalIds><ThreadIds><CoreIds><ProcessorIds><NodeIds>, that
1079 is, thread is part of a core that is part of a processor, which
1080 is part of a NUMA node.
1081 * <LogicalIds><ThreadIds><CoreIds><NodeIds><ProcessorIds>, that
1083 is, thread is part of a core that is part of a NUMA node, which
1084 is part of a processor.
1085 A CPU topology can consist of both processor external, and pro‐
1087 cessor internal NUMA nodes as long as each logical processor be‐
1088 longs to only one NUMA node. If <ProcessorIds> is omitted, its
1089 default position is before <NodeIds>. That is, the default is
1090 processor external NUMA nodes.
1091 If a list of identifiers is used in an <IdDefs>:
1093 * <LogicalIds> must be a list of identifiers.
1095 * At least one other identifier type besides <LogicalIds> must
1097 also have a list of identifiers.
1098 * All lists of identifiers must produce the same number of iden‐
1100 tifiers.
1101 A simple example. A single quad core processor can be described
1103 as follows:
1104 % erl +sct L0-3c0-3
1106 1> erlang:system_info(cpu_topology).
1107 [{processor,[{core,{logical,0}},
1108 {core,{logical,1}},
1109 {core,{logical,2}},
1110 {core,{logical,3}}]}]
1111 A more complicated example with two quad core processors, each
1113 processor in its own NUMA node. The ordering of logical proces‐
1114 sors is a bit weird. This to give a better example of identifier
1115 lists:
1116 % erl +sct L0-1,3-2c0-3p0N0:L7,4,6-5c0-3p1N1
1118 1> erlang:system_info(cpu_topology).
1119 [{node,[{processor,[{core,{logical,0}},
1120 {core,{logical,1}},
1121 {core,{logical,3}},
1122 {core,{logical,2}}]}]},
1123 {node,[{processor,[{core,{logical,7}},
1124 {core,{logical,4}},
1125 {core,{logical,6}},
1126 {core,{logical,5}}]}]}]
1127 As long as real identifiers are correct, it is OK to pass a CPU
1129 topology that is not a correct description of the CPU topology.
1130 When used with care this can be very useful. This to trick the
1131 emulator to bind its schedulers as you want. For example, if you
1132 want to run multiple Erlang runtime systems on the same machine,
1133 you want to reduce the number of schedulers used and manipulate
1134 the CPU topology so that they bind to different logical CPUs. An
1135 example, with two Erlang runtime systems on a quad core machine:
1136 % erl +sct L0-3c0-3 +sbt db +S3:2 -detached -noinput -noshell -sname one
1138 % erl +sct L3-0c0-3 +sbt db +S3:2 -detached -noinput -noshell -sname two
1139 In this example, each runtime system have two schedulers each on‐
1141 line, and all schedulers online will run on different cores. If
1142 we change to one scheduler online on one runtime system, and
1143 three schedulers online on the other, all schedulers online will
1144 still run on different cores.
1145 Notice that a faked CPU topology that does not reflect how the
1147 real CPU topology looks like is likely to decrease the perfor‐
1148 mance of the runtime system.
1149 For more information, see erlang:system_info(cpu_topology).
1151 +sfwi Interval:
1153 Sets scheduler-forced wakeup interval. All run queues are scanned
1154 each Interval milliseconds. While there are sleeping schedulers
1155 in the system, one scheduler is woken for each non-empty run
1156 queue found. Interval default to 0, meaning this feature is dis‐
1157 abled.
1158 Note:
1160 This feature has been introduced as a temporary workaround for
1161 long-executing native code, and native code that does not bump re‐
1162 ductions properly in OTP. When these bugs have been fixed, this
1163 flag will be removed.
1164 +spp Bool:
1167 Sets default scheduler hint for port parallelism. If set to true,
1168 the virtual machine schedules port tasks when it improves paral‐
1169 lelism in the system. If set to false, the virtual machine tries
1170 to perform port tasks immediately, improving latency at the ex‐
1171 pense of parallelism. Default to false. The default used can be
1172 inspected in runtime by calling erlang:system_info(port_parallel‐
1173 ism). The default can be overridden on port creation by passing
1174 option parallelism to erlang:open_port/2.
1175 +sss size:
1177 Suggested stack size, in kilowords, for scheduler threads. Valid
1178 range is 20-8192 kilowords. The default suggested stack size is
1179 128 kilowords.
1180 +sssdcpu size:
1182 Suggested stack size, in kilowords, for dirty CPU scheduler
1183 threads. Valid range is 20-8192 kilowords. The default suggested
1184 stack size is 40 kilowords.
1185 +sssdio size:
1187 Suggested stack size, in kilowords, for dirty IO scheduler
1188 threads. Valid range is 20-8192 kilowords. The default suggested
1189 stack size is 40 kilowords.
1190 +stbt BindType:
1192 Tries to set the scheduler bind type. The same as flag +sbt ex‐
1193 cept how some errors are handled. For more information, see +sbt.
1194 +sub true|false:
1196 Enables or disables scheduler utilization balancing of load. By
1197 default scheduler utilization balancing is disabled and instead
1198 scheduler compaction of load is enabled, which strives for a load
1199 distribution that causes as many scheduler threads as possible to
1200 be fully loaded (that is, not run out of work). When scheduler
1201 utilization balancing is enabled, the system instead tries to
1202 balance scheduler utilization between schedulers. That is, strive
1203 for equal scheduler utilization on all schedulers.
1204 +sub true is only supported on systems where the runtime system
1206 detects and uses a monotonically increasing high-resolution
1207 clock. On other systems, the runtime system fails to start.
1208 +sub true implies +scl false. The difference between +sub true
1210 and +scl false is that +scl false does not try to balance the
1211 scheduler utilization.
1212 +swct very_eager|eager|medium|lazy|very_lazy:
1214 Sets scheduler wake cleanup threshold. Defaults to medium. Con‐
1215 trols how eager schedulers are to be requesting wakeup because of
1216 certain cleanup operations. When a lazy setting is used, more
1217 outstanding cleanup operations can be left undone while a sched‐
1218 uler is idling. When an eager setting is used, schedulers are
1219 more frequently woken, potentially increasing CPU-utilization.
1220 Note:
1222 This flag can be removed or changed at any time without prior no‐
1223 tice.
1224 +sws default|legacy:
1227 Sets scheduler wakeup strategy. Default strategy changed in ERTS
1228 5.10 (Erlang/OTP R16A). This strategy was known as proposal in
1229 Erlang/OTP R15. The legacy strategy was used as default from R13
1230 up to and including R15.
1231 Note:
1233 This flag can be removed or changed at any time without prior no‐
1234 tice.
1235 +swt very_low|low|medium|high|very_high:
1238 Sets scheduler wakeup threshold. Defaults to medium. The thresh‐
1239 old determines when to wake up sleeping schedulers when more work
1240 than can be handled by currently awake schedulers exists. A low
1241 threshold causes earlier wakeups, and a high threshold causes
1242 later wakeups. Early wakeups distribute work over multiple sched‐
1243 ulers faster, but work does more easily bounce between sched‐
1244 ulers.
1245 Note:
1247 This flag can be removed or changed at any time without prior no‐
1248 tice.
1249 +swtdcpu very_low|low|medium|high|very_high:
1252 As +swt but affects dirty CPU schedulers. Defaults to medium.
1253 Note:
1255 This flag can be removed or changed at any time without prior no‐
1256 tice.
1257 +swtdio very_low|low|medium|high|very_high:
1260 As +swt but affects dirty IO schedulers. Defaults to medium.
1261 Note:
1263 This flag can be removed or changed at any time without prior no‐
1264 tice.
1265 +t size:
1268 Sets the maximum number of atoms the virtual machine can handle.
1269 Defaults to 1,048,576.
1270 +T Level:
1272 Enables modified timing and sets the modified timing level. Valid
1273 range is 0-9. The timing of the runtime system is changed. A high
1274 level usually means a greater change than a low level. Changing the
1275 timing can be very useful for finding timing-related bugs.
1276 Modified timing affects the following:
1278 Process spawning:
1280 A process calling spawn, spawn_link, spawn_monitor, or spawn_opt
1281 is scheduled out immediately after completing the call. When
1282 higher modified timing levels are used, the caller also sleeps
1283 for a while after it is scheduled out.
1284 Context reductions:
1286 The number of reductions a process is allowed to use before it is
1287 scheduled out is increased or reduced.
1288 Input reductions:
1290 The number of reductions performed before checking I/O is in‐
1291 creased or reduced.
1292 Note:
1294 Performance suffers when modified timing is enabled. This flag is
1295 only intended for testing and debugging.
1296 return_to and return_from trace messages are lost when tracing on the
1298 spawn BIFs.
1299 This flag can be removed or changed at any time without prior notice.
1301 +v:
1304 Verbose.
1305 +V:
1307 Makes the emulator print its version number.
1308 +W w | i | e:
1310 Sets the mapping of warning messages for error_logger. Messages
1311 sent to the error logger using one of the warning routines can be
1312 mapped to errors (+W e), warnings (+W w), or information reports
1313 (+W i). Defaults to warnings. The current mapping can be retrieved
1314 using error_logger:warning_map/0. For more information, see er‐
1315 ror_logger:warning_map/0 in Kernel.
1316 +zFlag Value:
1318 Miscellaneous flags:
1319 +zdbbl size:
1321 Sets the distribution buffer busy limit (dist_buf_busy_limit) in
1322 kilobytes. Valid range is 1-2097151. Defaults to 1024.
1323 A larger buffer limit allows processes to buffer more outgoing
1325 messages over the distribution. When the buffer limit has been
1326 reached, sending processes will be suspended until the buffer
1327 size has shrunk. The buffer limit is per distribution channel. A
1328 higher limit gives lower latency and higher throughput at the ex‐
1329 pense of higher memory use.
1330 +zdntgc time:
1332 Sets the delayed node table garbage collection time (de‐
1333 layed_node_table_gc) in seconds. Valid values are either infinity
1334 or an integer in the range 0-100000000. Defaults to 60.
1335 Node table entries that are not referred linger in the table for
1337 at least the amount of time that this parameter determines. The
1338 lingering prevents repeated deletions and insertions in the ta‐
1339 bles from occurring.
1340
1342 ERL_CRASH_DUMP:
1343 If the emulator needs to write a crash dump, the value of this
1344 variable is the filename of the crash dump file. If the variable is
1345 not set, the name of the crash dump file is erl_crash.dump in the
1346 current directory.
1347 ERL_CRASH_DUMP_NICE:
1349 Unix systems: If the emulator needs to write a crash dump, it uses
1350 the value of this variable to set the nice value for the process,
1351 thus lowering its priority. Valid range is 1-39 (higher values are
1352 replaced with 39). The highest value, 39, gives the process the
1353 lowest priority.
1354 ERL_CRASH_DUMP_SECONDS:
1356 Unix systems: This variable gives the number of seconds that the
1357 emulator is allowed to spend writing a crash dump. When the given
1358 number of seconds have elapsed, the emulator is terminated.
1359 ERL_CRASH_DUMP_SECONDS=0:
1361 If the variable is set to 0 seconds, the runtime system does not
1362 even attempt to write the crash dump file. It only terminates.
1363 This is the default if option -heart is passed to erl and
1364 ERL_CRASH_DUMP_SECONDS is not set.
1365 ERL_CRASH_DUMP_SECONDS=S:
1367 If the variable is set to a positive value S, wait for S seconds
1368 to complete the crash dump file and then terminates the runtime
1369 system with a SIGALRM signal.
1370 ERL_CRASH_DUMP_SECONDS=-1:
1372 A negative value causes the termination of the runtime system to
1373 wait indefinitely until the crash dump file has been completly
1374 written. This is the default if option -heart is not passed to
1375 erl and ERL_CRASH_DUMP_SECONDS is not set.
1376 See also heart(3).
1378 ERL_CRASH_DUMP_BYTES:
1380 This variable sets the maximum size of a crash dump file in bytes.
1381 The crash dump will be truncated if this limit is exceeded. If the
1382 variable is not set, no size limit is enforced by default. If the
1383 variable is set to 0, the runtime system does not even attempt to
1384 write a crash dump file.
1385 Introduced in ERTS 8.1.2 (Erlang/OTP 19.2).
1387 ERL_AFLAGS:
1389 The content of this variable is added to the beginning of the com‐
1390 mand line for erl.
1391 Flag -extra is treated in a special way. Its scope ends at the end
1393 of the environment variable content. Arguments following an -extra
1394 flag are moved on the command line into section -extra, that is,
1395 the end of the command line following an -extra flag.
1396 ERL_ZFLAGS and ERL_FLAGS:
1398 The content of these variables are added to the end of the command
1399 line for erl.
1400 Flag -extra is treated in a special way. Its scope ends at the end
1402 of the environment variable content. Arguments following an -extra
1403 flag are moved on the command line into section -extra, that is,
1404 the end of the command line following an -extra flag.
1405 ERL_LIBS:
1407 Contains a list of additional library directories that the code
1408 server searches for applications and adds to the code path; see
1409 code(3).
1410 ERL_EPMD_ADDRESS:
1412 Can be set to a comma-separated list of IP addresses, in which case
1413 the epmd daemon listens only on the specified address(es) and on
1414 the loopback address (which is implicitly added to the list if it
1415 has not been specified).
1416 ERL_EPMD_PORT:
1418 Can contain the port number to use when communicating with epmd.
1419 The default port works fine in most cases. A different port can be
1420 specified to allow nodes of independent clusters to co-exist on the
1421 same host. All nodes in a cluster must use the same epmd port num‐
1422 ber.
1423
1425 On Unix systems, the Erlang runtime will interpret two types of sig‐
1426 nals.
1427 SIGUSR1:
1429 A SIGUSR1 signal forces a crash dump.
1430 SIGTERM:
1432 A SIGTERM will produce a stop message to the init process. This is
1433 equivalent to a init:stop/0 call.
1434 Introduced in ERTS 8.3 (Erlang/OTP 19.3)
1436 The signal SIGUSR2 is reserved for internal usage. No other signals are
1438 handled.
1439
1441 The standard Erlang/OTP system can be reconfigured to change the de‐
1442 fault behavior on startup.
1443 The .erlang startup file:
1445 When Erlang/OTP is started, the system searches for a file named
1446 .erlang in the user's home directory.
1447 If an .erlang file is found, it is assumed to contain valid Erlang
1449 expressions. These expressions are evaluated as if they were input
1450 to the shell.
1451 A typical .erlang file contains a set of search paths, for example:
1453 io:format("executing user profile in HOME/.erlang\n",[]).
1455 code:add_path("/home/calvin/test/ebin").
1456 code:add_path("/home/hobbes/bigappl-1.2/ebin").
1457 io:format(".erlang rc finished\n",[]).
1458 user_default and shell_default:
1460 Functions in the shell that are not prefixed by a module name are
1461 assumed to be functional objects (funs), built-in functions (BIFs),
1462 or belong to the module user_default or shell_default.
1463 To include private shell commands, define them in a module user_de‐
1465 fault and add the following argument as the first line in the .er‐
1466 lang file:
1467 code:load_abs("..../user_default").
1469 erl:
1471 If the contents of .erlang are changed and a private version of
1472 user_default is defined, the Erlang/OTP environment can be custom‐
1473 ized. More powerful changes can be made by supplying command-line
1474 arguments in the startup script erl. For more information, see
1475 init(3).
1476
1478 epmd(1), erl_prim_loader(3), erts_alloc(3), init(3), application(3),
1479 auth(3), code(3), erl_boot_server(3), heart(3), net_kernel(3), make(3)
1480Ericsson AB erts 11.2.2.2 erl(1)