Original author(s) | Paul Kranenburg |
---|---|
Developer(s) | Dmitry Levin |
Stable release | |
Repository | |
Written in | C [2] |
Operating system | Linux |
Platform | AArch64, DEC Alpha, ARC, ARM EABI/OABI, AVR32, Blackfin, C-SKY, HP PA-RISC, IA-32, IA-64, LoongArch, Motorola 68k, Imagination META, MicroBlaze, MIPS, Nios II, OpenRISC, Power ISA 32/64 bit, RISC-V, System/390/z/Architecture, SuperH 32/64 bit, SPARC 32/64 bit, TILE, TILEPro, TILE-Gx, x86-64, x32 ABI, Xtensa |
Available in | English [note 1] |
Type | Debugging |
License | LGPL v2.1+ [note 2] [4] |
Website | strace |
strace is a diagnostic, debugging and instructional userspace utility for Linux. It is used to monitor and tamper with interactions between processes and the Linux kernel, which include system calls, signal deliveries, and changes of process state. The operation of strace is made possible by the kernel feature known as ptrace.
Some Unix-like systems provide other diagnostic tools similar to strace, such as truss.
Strace was originally written for SunOS by Paul Kranenburg in 1991, according to its copyright notice, and published early in 1992, in volume three of comp.sources.sun. The initial README file contained the following: [5]
strace(1) is a system call tracer for Sun(tm) systems much like the Sun supplied program trace(1). strace(1) is a useful utility to sort of debug programs for which no source is available which unfortunately includes almost all of the Sun supplied system software.
Later, Branko Lankester ported this version to Linux, releasing his version in November 1992 with the second release following in 1993. [6] [7] Richard Sladkey combined these separate versions of strace in 1993, and ported the program to SVR4 and Solaris in 1994, [8] resulting in strace 3.0 that was announced in comp.sources.misc in mid-1994. [9]
Beginning in 1996, strace was maintained by Wichert Akkerman. During his tenure, strace development migrated to CVS; ports to FreeBSD and many architectures on Linux (including ARM, IA-64, MIPS, PA-RISC, PowerPC, s390, SPARC) were introduced. In 2002, the burden of strace maintainership was transferred to Roland McGrath. Since then, strace gained support for several new Linux architectures (AMD64, s390x, SuperH), bi-architecture support for some of them, and received numerous additions and improvements in syscalls decoders on Linux; strace development migrated to git during that period. Since 2009, strace is actively maintained by Dmitry Levin. strace gained support for AArch64, ARC, AVR32, Blackfin, Meta, Nios II, OpenSISC 1000, RISC-V, Tile/TileGx, Xtensa architectures since that time.
The last version of strace that had some (evidently dead) [10] code for non-Linux operating systems was 4.6, released in March 2011. [11] In strace version 4.7, released in May 2012, [12] all non-Linux code had been removed; [13] since strace 4.13, [14] the project follows Linux kernel's release schedule, and as of version 5.0, [15] it follows Linux's versioning scheme as well.
In 2012 strace also gained support for path tracing and file descriptor path decoding. [16] In August 2014, strace 4.9 was released, [17] [18] where support for stack traces printing was added. In December 2016, [19] [20] syscall fault injection feature was implemented.
Version | Release date | Notable changes |
---|---|---|
6.2 | 26 February 2023 [21] | Implemented collision resolution for overlapping terminal/sound ioctl commands |
6.1 | 12 December 2022 [22] | Enhanced decoding of statx syscall and terminal ioctl commands |
6.0 | 29 October 2022 [23] | Enhanced decoding of setns syscall and PTP_* ioctl commands |
5.19 | 12 August 2022 [24] | The "(deleted)" marker of unlinked file descriptor paths is no longer interpreted as a part of a path; enhanced decoding of arch_prctl , io_uring_register , and prctl syscalls; enhanced decoding of siginfo_t structure |
5.18 | 18 June 2022 [25] | |
5.17 | 26 March 2022 [26] | Added support for a new set_mempolicy_home_node syscall on Linux |
5.16 | 10 January 2022 [27] | Added support for printing SELinux context mismatches (--secontext=mismatch option); added support for a new futex_waitv syscall on Linux |
5.15 | 1 December 2021 [28] | |
5.14 | 2 September 2021 [29] | Added support for the new memfd_secret and quotactl_fd syscalls decoding |
5.13 | 18 July 2021 [30] | Implemented System Call Vectored ABI support on the IBM POWER architecture |
5.12 | 26 April 2021 [31] | Added support for the new mount_setattr syscall decoding |
5.11 | 17 February 2021 [32] | Added option for injecting data on syscall entry/exit (poke_enter= and poke_exit= parameters of the --inject option); added support for the new epoll_pwait2 syscall decoding; imlemented decoding of FS_IOC_FS[GS]ETXATTR , FS_IOC{,32}_[GS]ETFLAGS , GPIO_* , SIOCADDMULTI , SIOCDELMULTI , SIOCGIFENCAP , SIOCOUTQNSD , SIOCSIFENCAP , SIOCSIFHWBROADCAST , UBI_IOCRPEB , UBI_IOCSPEB , V4L2_BUF_TYPE_META_CAPTURE , V4L2_BUF_TYPE_META_OUTPUT , and VIDIOC_QUERY_EXT_CTRL ioctl commands |
5.10 | 14 December 2020 [33] | Stack tracing is supported for non-native personalities when compiled with libdw ; added support for process_madvise syscall decoding |
5.9 | 24 September 2020 [34] | Added support for close_range syscall decoding; added support for TEE_* ioctl decoding; implemented decoding of {msg,msq,sem}info and {msg,msq,sem}id_ds structures used in the respective System V IPC syscalls commands |
5.8 | 6 August 2020 [35] | Changed the list of syscalls covered by the %process group to contain syscalls associated with process lifecycle: kill , tkill , tgkill , pidfd_send_signal , and rt_sigqueueinfo were added, arch_prctl and unshare were removed; added support for faccessat2 syscall decoding; added support for LOOP_CONFIGURE ioctl decoding |
5.7 | 1 June 2020 [36] | |
5.6 | 7 April 2020 [37] | Added long option aliases along with possibilities for more precise option specifications via -e quiet /--quiet , -e decode-fds /--decode-fds , --absolute-timestamps , --relative-timestamps , and --syscall-times options; implemented various syscall statistics improvements |
5.5 | 6 February 2020 [38] | |
5.4 | 28 November 2019 [39] | |
5.3 | 25 September 2019 [40] | |
5.2 | 12 July 2019 [41] | |
5.1 | 22 May 2019 [42] | |
5.0 | 19 March 2019 [43] | |
4.26 | 26 December 2018 [44] | |
4.25 | 30 October 2018 [45] | |
4.24 | 14 August 2018 [46] | |
4.23 | 14 June 2018 [47] | |
4.22 | 5 April 2018 [48] | |
4.21 | 13 February 2018 [49] | |
4.20 | 13 November 2017 [50] | |
4.19 | 5 September 2017 [51] | |
4.18 | 5 July 2017 [52] | |
4.17 | 24 May 2017 [53] | |
4.16 | 14 February 2017 [54] | |
4.15 | 14 December 2016 [55] | |
4.14 | 4 October 2016 [56] | |
4.13 | 26 July 2016 [57] | |
4.12 | 31 May 2016 [58] | |
4.11 | 21 December 2015 [59] | |
4.10 | 6 March 2015 [60] | |
4.9 | 15 August 2014 [61] | |
4.8 | 3 June 2013 [62] | |
4.7 | 2 May 2012 [63] | All non-Linux code had been removed |
4.6 | 15 March 2011 [64] | |
4.5.20 | 13 April 2010 | Added support for new inotify_init1 , perf_event_open , preadv , pwritev , recvmmsg , rt_tgsigqueueinfo syscalls on Linux; -C option added, that combines regular and summary output; added Tile architecture support on Linux |
4.5.19 | 21 October 2009 | Maintainership passed to Dmitry Levin; strace now terminates with the same exit code/signal as the traced program (if it was started by strace); added support for new accept4 , dup3 , epoll_create1 , eventfd2 , inotify_init1 , pipe2 , signalfd4 syscalls on Linux; added Blackfin, AVR32, and CRIS architedcture support on Linux |
4.5.18 | 28 August 2008 | Added support for subpage_prot POWER-specific syscall on Linux |
4.5.17 | 21 July 2008 | -F flag is deprecated, as -f traces vfork on Linux since long time |
4.5.16 | 3 August 2007 | |
4.5.15 | 16 January 2007 | Added support for new *at , inotify* , pselect6 , ppoll and unshare syscalls on Linux |
4.5.14 | 16 January 2007 | System call number can be supplied in -e specification |
4.5.13 | 3 August 2005 | Add desc syscall group support to -e trace= |
4.5.12 | 8 June 2005 | |
4.5.11 | 22 March 2005 | |
4.5.10 | 13 March 2005 | |
4.5.9 | 4 February 2004 | |
4.5.8 | 19 October 2004 | Decode mbind , [sg]et_mempolicy , waitid , fadvise64{,_64} , and epoll_* syscalls, RTC_* ioctls on Linux |
4.5.7 | 31 August 2004 | |
4.5.6 | 12 July 2004 | Added support for 64-bit SPARC architecture on Linux. |
4.5.5 | 27 June 2004 | |
4.5.4 | 3 June 2004 | -p attaches to all NPTL threads on Linux only when -f is supplied |
4.5.3 | 16 April 2004 | Added support for mq_* syscalls on Linux; -p now attaches to all NPTL threads on Linux |
4.5.2 | 1 March 2004 | |
4.5.1 | 13 November 2003 | |
4.5 | 24 September 2003 | Maintainership passed to Roland McGrath; added x86-64 support on Linux with support of tracing of compat processes; added support for SH and SH64 architectures on Linux; -E option added |
4.4 | 19 August 2001 | |
4.3.1 | 7 April 2001 | |
4.3 | 1 April 2001 | Added support for HP PA/RISC and IA-64 architectures on Linux; added support for 32-bit UID/GID syscalls on Linux; added support for FreeBSD on x86 |
4.2 | 21 January 2000 | Added support for IBM Z architecture on Linux |
4.1 | 26 November 1999 | Added support for MIPS architecture on Linux; strace-graph script added |
4.0.1 | 25 July 1999 | |
4.0 | 9 July 1999 | Fixed 64-bit struct stat decoding on Linux; Irix 64 updates; Solaris updates |
3.99.1 | 9 June 1999 | |
3.99 | 27 April 1999 | New maintainer, Wichert Akkerman; added support for IBM POWER, SPARC, and ARM architectures on Linux; added support for many syscalls on Linux |
3.1 | 1 June 1996 | Added support for the Irix OS, m68k and DEC Alpha architectures on Linux; added support for -o! option syntax; added support for syscall classes (file , process ); added support for IPC syscalls on Sun OS |
3.0 | 9 July 1994 [65] | Initial cross-platform version by Richard Sladkey. Includes support for -x , -q , -e (trace , abbrev , verbose , raw , signal , read , and write qualifiers), -c , -i options |
The most common use is to start a program using strace, which prints a list of system calls made by the program. This is useful if the program continually crashes, or does not behave as expected; for example using strace may reveal that the program is attempting to access a file which does not exist or cannot be read.
An alternative application is to use the -p
flag to attach to a running process. This is useful if a process has stopped responding, and might reveal, for example, that the process is blocking whilst attempting to make a network connection.
Among other features, strace allows the following:
-e trace=
option): by name, like clone,fork,vfork; using one of the predefined groups, like %ipc or %file; or (since strace 4.17) using regular expression syntax, like -e trace=/clock_.*
.-P /etc/ld.so.cache
, for example).-e read=
and -e write=
options).-T
, -c
, -C
, and -w
options; -U
option enables printing of additional information, like minimum and maximum syscall execution time).-t
and -r
options).-e inject=syscall specification:tampering specification
option): modifying return (:retval=
; since strace 4.16) and error code (:error=
; since strace 4.15) of the specified syscalls, inject signals (:signal=
; since strace 4.16), delays (:delay_enter=
and :delay_exit=
; since strace 4.22), and modify data pointed by syscall arguments (:poke_enter=
and :poke_exit=
; since strace 5.11) upon their execution.-y
option; -yy
option provides some additional information, like endpoint addresses for sockets, paths and device major/minor numbers for files).-k
option).-e status=
option; since strace 5.2 [note 3] ).--pidns-translation
option; since strace 5.9).--secontext
option; since strace 5.12).strace supports decoding of arguments of some classes of ioctl commands, such as BTRFS_*, V4L2_*, DM_*, NSFS_*, MEM*, EVIO*, KVM_*, and several others; it also supports decoding of various netlink protocols.
As strace only details system calls, it cannot be used to detect as many problems as a code debugger such as GNU Debugger (gdb). It is, however, easier to use than a code debugger, and is a very useful tool for system administrators. It is also used by researchers to generate system call traces for later system call replay. [66] [67] [68]
The following is an example of typical output of the strace
command:
user@server:~$ stracels ...open(".", O_RDONLY|O_NONBLOCK|O_LARGEFILE|O_DIRECTORY|O_CLOEXEC) = 3fstat64(3, {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0fcntl64(3, F_GETFD) = 0x1 (flags FD_CLOEXEC)getdents64(3, /* 18 entries */, 4096) = 496getdents64(3, /* 0 entries */, 4096) = 0close(3) = 0fstat64(1, {st_mode=S_IFIFO|0600, st_size=0, ...}) = 0mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7f2c000write(1, "autofs\nbackups\ncache\nflexlm\ngames"..., 86autofsA
The above fragment is only a small part of the output of strace when run on the 'ls' command. It shows that the current working directory is opened, inspected and its contents retrieved. The resulting list of file names is written to standard output.
Different operating systems feature other similar or related instrumentation tools, offering similar or more advanced features; some of the tools (although using the same or a similar name) may use completely different work mechanisms, resulting in different feature sets or results. Such tools include the following:
-z
option, was originally added in strace 4.5, but was never documented as it did not work properly.The GNU Debugger (GDB) is a portable debugger that runs on many Unix-like systems and works for many programming languages, including Ada, Assembly, C, C++, D, Fortran, Haskell, Go, Objective-C, OpenCL C, Modula-2, Pascal, Rust, and partially others.
A Berkeley (BSD) socket is an application programming interface (API) for Internet domain sockets and Unix domain sockets, used for inter-process communication (IPC). It is commonly implemented as a library of linkable modules. It originated with the 4.2BSD Unix operating system, which was released in 1983.
In computing, a system call is the programmatic way in which a computer program requests a service from the operating system on which it is executed. This may include hardware-related services, creation and execution of new processes, and communication with integral kernel services such as process scheduling. System calls provide an essential interface between a process and the operating system.
In computing, ls
is a command to list computer files and directories in Unix and Unix-like operating systems. It is specified by POSIX and the Single UNIX Specification.
In Unix and Unix-like computer operating systems, a file descriptor is a process-unique identifier (handle) for a file or other input/output resource, such as a pipe or network socket.
DTrace is a comprehensive dynamic tracing framework originally created by Sun Microsystems for troubleshooting kernel and application problems on production systems in real time. Originally developed for Solaris, it has since been released under the free Common Development and Distribution License (CDDL) in OpenSolaris and its descendant illumos, and has been ported to several other Unix-like systems.
seccomp is a computer security facility in the Linux kernel. seccomp allows a process to make a one-way transition into a "secure" state where it cannot make any system calls except exit
, sigreturn
, read
and write
to already-open file descriptors. Should it attempt any other system calls, the kernel will either just log the event or terminate the process with SIGKILL or SIGSYS. In this sense, it does not virtualize the system's resources but isolates the process from them entirely.
inotify is a Linux kernel subsystem created by John McCutchan, which monitors changes to the filesystem, and reports those changes to applications. It can be used to automatically update directory views, reload configuration files, log changes, backup, synchronize, and upload. The inotifywait and inotifywatch commands allow using the inotify subsystem from the command line. One major use is in desktop search utilities like Beagle, where its functionality permits reindexing of changed files without scanning the filesystem for changes every few minutes, which would be very inefficient.
The GNU Binary Utilities, or binutils, is a collection of programming tools maintained by the GNU Project for working with executable code including assembly, linking and many other development operations.
truss is a system tool available on some Unix-like operating systems. When invoked with an additional executable command-line argument, truss makes it possible to print out the system calls made by and the signals received by this executable command-line argument. As of version IEEE Std 1003.1-2008, truss is not part of the Single UNIX Specification (POSIX).
In computing, SystemTap is a scripting language and tool for dynamically instrumenting running production Linux-based operating systems. System administrators can use SystemTap to extract, filter and summarize data in order to enable diagnosis of complex performance or functional problems.
ltrace is a debugging utility in Linux, used to display the calls a userspace application makes to shared libraries. It does this by hooking into the dynamic loading system, allowing it to insert shims which display the parameters which the applications uses when making the call, and the return value which the library call reports. ltrace can also trace Linux system calls. Because it uses the dynamic library hooking mechanism, ltrace cannot trace calls to libraries which are statically linked directly to the target binary. Since 0.7.3, ltrace can also trace calls to libraries which are loaded using dlopen.
ktrace is a utility included with certain versions of BSD Unix and Mac OS X that traces kernel interaction with a program and dumps it to disk for the purposes of debugging and analysis. Traced kernel operations include system calls, name translations, signal processing, and I/O.
ProbeVue is IBM's implementation of a lightweight dynamic tracing environment introduced in AIX version 6.1. ProbeVue provides the ability to probe running processes in order to provide statistical analysis as well as retrieve data from the probed process. The dynamic nature of ProbeVue allows it to be used as a global system performance tool while retaining the ability to drill into very specific events on a single process or thread.
LTTng is a system software package for correlated tracing of the Linux kernel, applications and libraries. The project was originated by Mathieu Desnoyers with an initial release in 2005. Its predecessor is the Linux Trace Toolkit.
ptrace is a system call found in Unix and several Unix-like operating systems. By using ptrace one process can control another, enabling the controller to inspect and manipulate the internal state of its target. ptrace is used by debuggers and other code-analysis tools, mostly as aids to software development.
ftrace is a tracing framework for the Linux kernel. Although its original name, Function Tracer, came from ftrace's ability to record information related to various function calls performed while the kernel is running, ftrace's tracing capabilities cover a much broader range of kernel's internal operations.
kdump is a feature of the Linux kernel that creates crash dumps in the event of a kernel crash. When triggered, kdump exports a memory image that can be analyzed for the purposes of debugging and determining the cause of a crash. The dumped image of main memory, exported as an Executable and Linkable Format (ELF) object, can be accessed either directly through /proc/vmcore during the handling of a kernel crash, or it can be automatically saved to a locally accessible file system, to a raw device, or to a remote system accessible over network.
Entropy-supplying system calls are system calls in Unix-like operating system kernels through which processes can obtain entropic or random data. The first of these was getentropy
, introduced to the OpenBSD operating system in release 5.6, as a refactoring of the sysctl(3) KERN_ARND approach used since 1997. Linux offers a very similar system call, getrandom
, which was based on getentropy
. It was first available in Linux 3.17, released in October 2014. In July 2015, Solaris introduced slightly modified versions of getentropy
and getrandom
. In August 2015, FreeBSD introduced the read_random
system call for obtaining random data from the kernel.
A virtual kernel architecture (vkernel) is an operating system virtualisation paradigm where kernel code can be compiled to run in the user space, for example, to ease debugging of various kernel-level components, in addition to general-purpose virtualisation and compartmentalisation of system resources. It is used by DragonFly BSD in its vkernel implementation since DragonFly 1.7, having been first revealed in September 2006, and first released in the stable branch with DragonFly 1.8 in January 2007.