관리-도구
편집 파일: metadatabase.db
SQLite format 3 @ * � * . � � � �5� �tt �E++�Atablemetavirt_segdirmetavirt_segdirCREATE TABLE 'metavirt_segdir'(level INTEGER,idx INTEGER,start_block INTEGER,leaves_end_block INTEGER,end_block INTEGER,root BLOB,PRIMARY KEY(level, idx))=Q+ indexsqlite_autoindex_metavirt_segdir_1metavirt_segdirg x//� tablemetavirt_segmentsmetavirt_segmentsCREATE TABLE 'metavirt_segments'(blockid INTEGER PRIMARY KEY, block BLOB)�%--�}tablemetavirt_contentmetavirt_contentCREATE TABLE 'metavirt_content'(docid INTEGER PRIMARY KEY, 'c0title', 'c1name', 'c2keywords', 'c3description', 'c4path')� �ctablemetavirtmetavirtCREATE VIRTUAL TABLE metavirt USING fts3(title text, name text, keywords text, description text, path text) � Q � Q �#o ]%�SWatch IO Activity on a Particular Devicetraceio2.stpioPrint out the executable name and process number as reads and writes to the specified device occur.io�n g!?�eTrace Time Spent in Read and Write for Files iotime.stpprofiling syscall io fileThe script watches each open, close, read, and write syscalls on the system. For each file the scripts observes opened it accumulates the amount of wall clock time spent in read and write operations and the number of bytes read and written. When a file is closed the script prints out a pair of lines for the file. Both lines begin with a timestamp in microseconds, the PID number, and the executable name in parentheses. The first line with the "access" keyword lists the file name, the attempted number of bytes for the read and write operations. The second line with the "iotime" keyword list the file name and the number of microseconds accumulated in the read and write syscalls.io � �Q�U� � �g i!!�9Graphing Disk and CPU Utilization with gnuplotgraphs.stpmonitoringThe script tracks the disk and CPU utilization. It prints a stream of data which, when piped directly into gnuplot, draws historical curves for each.general� 1 �9 block breaker gameblock.stpstapgamesA block game where you progressively break the ceiling blocks until clearing the levelstapgamesw '% � pingpong gamepingpong.stpstapgamesA simulated ball bounces around the terminal reflecting at the edgesstapgames� +�5 20482048.stp_best stapgamesThe modern classic 2048 sliding-tiles game, using local keyboard and ansi animation.stapgamesa % s life gamelifegame.stpstapgameswatch as your creation morphes into different formsstapgamesJ ! I eater gameeater.stpstapgameswalk through a maze, eat stuffstapgames t = ; : 9 8 7 6 5 4 3 2 1 0 / . - K _ b � o � Y h P t � � � � � v � e � Q ] � h t � � P � � � � v [ � � T � � ] � � v W � � [ G � C � ! � ! � � C � � < G [ m B G < � � � � I C C I � � � � � � � C � � � � � � p p C � � � � � ? p ) � g � X � � � g � ) s p � , s V ) { g i � { ' s , * ) @ V � � ( ) * + , s ~ ' # � s � � < Kk ��������������������������~ytoje`[VQLGB=83.)$ ���������������������}wqk v�6 ��4 ]�2 ��1 y�/ �. ��- ��, ��+ ��* �) ��( [�' T�$ �" �� � �� W� �� G� C� !� � � �� �� <� m� %� B �| �z Iw u #s o �m �j �g dc X_ �] �[ pY �U �O �L �K ?I )H gF E �B @ �= : s8 4 Z1 ,/ V+ {( i% �" � ' @ � � ( + ~ � � 9� |�������������������������~ytoje`[VQLGB=83.)$ ������������ �N �M LL xK kJ �I �P @ >= N< j; E: 9 �4 �3 S2 �0 "/ ^. �- \, H+ * }( D' & �% &$ c# �" R! � � � n � � u z M l r � J A q � � � � a F � � �������� � � � � e Q � �������������������������� n � ���\+��Q � �k �H0 350 a� ccess� nd� branches� cache� ounter� s� ycles� exit� insn� list� on� per� f � ormance� rofiling � vides� ratios� efs� script� how� ing� orted� tp � the� o� using� : ^ IQ81 8008Iby ev incr mm pf showi the wher, B 8>62 65728comp go me pe sm to7 X0553 52950 dispatche lik prints spent user& 6+/47 4656+er may prod tak# 0&*42 4166&eat me sc tt$ 2"%37 3703"devi ipis secu/ H!33 6120co for lo passi stp vi+ @24 4629dp loo processes the& 619 4232ev mou shown un& 8 14 5243 contr h of so u% :6 5095cr in pat stp vi � � �a �F 2 048 aJ , - O � � � ��" c+�Track Virtual Memory System Writing to Diskmmwriteback.stpmemoryThe mmwriteback.stp script uses the virtual memory tracepoints available in some kernels to report all of the file writebacks that occur form kupdate, pdflush and kjournald while the script is running. It's useful in determining where writes are coming from on a supposedly idle system that is experiencing unexpected IO.memory�k�! Q1/�=Collect Slab Allocation Statisticsvm.tracepoints.stpmemory statisticsThe script will probe all memory slab/slub allocations and collects information about the size of the object (bytes requested) and user-space process in execution. When run over a period of time, it helps to correlate kernel-space memory consumption owing to user-space processes.memory � �� � �h4 cAA�System-Wide Count of Network Packets by IPsnetfilter_summary_json.stp_best network traffic jsonThe script watches all IPv4 network traffic on the system. The data is output in JSON format and includes the number of packets sent along source IP address / destination IP address pair encountered, and the total number of bytes sent among the pair.network�%3 K91�+print tcp retransmission packettcp_retransmission.stptcp retransmissionThe tcp_retransmission.stp prints out a line for each tcp retransmission packetnetwork�[2 �?3�QTrack Start and Stop of Processes Due to Network Buffer Spacesk_stream_wait_memory.stpnetwork tcp processThe sk_stream-wait_memory.stp prints a time stamp, executable, and pid each time a process blocks due to the send buffer being full. A similar entry is printed each time a process continues because there is room in the buffer.network u �u �* �8 tty a spya urningh ypinga usedi rsg s a variousj walln tcho esn hen c , ichj ithi ? riteb A s b ingm ten j + xh �^) �@ script*a 1 sn ollinga si j econd [ [ �!�) a'+�! Show Time Spent on Each Line of a Functionlinetimes.stpprofiling _bestThe linetimes.stp script takes two arguments: where to find the function and the function name. linetimes.stp will instrument each line in the function. It will print out the number of times that the function is called, a table with the average and maximum time each line takes, and control flow information when the script exits.profiling � �� � : I''�TCP Connection Tracing Utilitytcp_trace.stpnetwork traceThis scripts traces a given TCP connection based on the filter parameters given by the user. The indexing is done by the 4 tuples local address, remote address, local port, remote port.network�Y9 w'1�yDisplay Network Statistics for Individual TCP Socketstcpipstat.stpnetwork statisticsThe tcpipstat script collects and displays network statistics related to individual TCP sockets or groups of sockets. The statistics that are collected are simular to that of the command netstat -s, only sorted and grouped by individual sockets.network ( ( �U �. socket s me urce ( pecified ulation tart >9 te ment istics pb uch$ ffixes m` pport yscall A s tem tap table pset cl hat 1 ? eu ir n� re,{ irdP s ose / read!2 � # id � me & d s 2 ing o* : p race d' ing , ked s * ial gger& ype writers � � � �� E#7�-System-Wide Futex Contentionfutexes.stpsyscall locking futexThe script watches the futex syscall on the system. On exit the futex's address, the number of contentions, and the average time for each contention on the futex are printed from lowest pid number to highest.process�U� s-/�qTrack Creation of Processes by process and execnamespawn_seeker.stpprocess schedulerThe spawn_seeker.stp script every minute (and on exit) prints out the local time and sorted lists of which processes and executables spawned tasks during the previous minute. This can be useful to determine what process is creating a flurry of short-lived processes. When a process exits its count of tasks created is added to its parent's count to better account for the indirect task creation by children processes. For more detailed examination of task creation consider using forktracker.stp.process � � � �!> �F toT�# % ) @ I = p �! tal � q race � back � points� s �) k� s� iggers�i wo� ype�0 unmapping� s� se� d� ful�" 4 9 I r� s� ing�R virtual� m� was� 2 " 4 tch� es � point � hat�% en$�/ " ! re�K ther�7 ich� le� / 1 ide� ll� ` th� , : , ork� rite� 5 "* able� s � x� 86�2 you �- " � . � �yu k!�aTrack a Particular Signal to a Specific Processsigmon.stpsignalsThe script watches for a particular signal sent to a specific process. When that signal is sent to the specified process, the script prints out the PID and executable of the process sending the signal, the PID and executable name of the process receiving the signal, and the signal number and name.process�Ot c# �Track the Migration of Specific Executablesmigrate.stpschedulerThe migrate.stp script takes an argument which is the executable name of the task it should monitor. Each time a task with that executable name migrates between processors an entry is printed with the process id (pid), the executable name, the processor off loading the task, and the process taking the task. Note that the task may or may not be executing at the time of the migration.process � \0 � pE G'yTracing of PHP code executionphp-trace.stptraceTrace of executing PHP code using the enabled markers.apps�)D S+)�AKeep track of NFS server statisticsnfsd-recent.stpnfs statisticsThis script tracks all nfsd server operations by client_ip address, and periodically lists those clients that have made recent requests. It's a way of finding out which nfs clients might be considered still connected.network�!C ;3+�?Network packet contentspacket_contents.stpnetwork trafficThe packet_contents.stp script displays the length of each network packet and its contents in both hexadecimal and ASCII. Systemtap strings are MAXSTRINGLEN in length by default which may not be enough for larger packets. In order to print larger packets, this limit can be increased by passing in the "-DMAXSTRINGLEN=65536" command line option.network 1 1 �L � creation urrent 3 ves data ebuginfo monstrate tect rmine iagnoseD fferent rectly 0 sk plays tributions: oingJ raws ue warf each ) t er dges ffective nable ) d ! vent s xceed- ing ecutable s ion it s pectedO tension $ file s tering rst ollowed* r0 # . mat 8 1 , s unc tion " ality s % game eneral7 iven , nuplot raphing s hardware ve eader istogram7 rical ow identical1 f ( llustrates � � �R� � ;�GSummarize the time each process spends doing copy on write for huge pageshugepage_cow_delays.stpmemoryThe system may attempt to reduce memory use by having two or more difference processes share the same writeable page in memory. However, when one of the processes write to shared page that is marked copy on write (COW) the page must be duplicated and then modified. On x86_64 machines the huge pages are 2MB in size, 512 times larger than a normal 4KB page. Thus, copy on write operation on a huge page can delay program execution by a hundred or more microseconds and maybe noticeable to latency sensitive programs. This script will tally the number of times that each process triggers a huge page copy on write, the total time in microseconds spent copying the pages, and the average time in microseconds of the huge page copy.memory �" � @ ]-9�SShow Process Ancestry for IP Connectionsconnect_stat.stpnetwork socket processThe connect_stat.stp script prints a task's entire ancestry (parent process name/uid/gid) whenever it attempts an outgoing socket connection to a given IP address.network�? a31�UTrack Creation of Incoming TCP Connectionstcp_connections.stpnetwork tcp socketThe tcp_connections.stp script prints information for each new incoming TCP connection accepted by the computer. The information includes the UID, the command accepting the connection, the PID of the command, the port the connection is on, and the IP address of the originator of the request.network�V> e/;�sIncrease Initial TCP Congestion Window to 10tcp_init_cwnd.stpnetwork tcp socket guruRun the tcp_init_cwnd.stp script in the background to override a kernel's default tcp cwnd value to 10, which has been found to improve latency for web server type workloads. The script prints a count of cwnd value changes when it is stopped.network , , �Q �& ofG j " n � e � ly . pen ! tional |! s r e ut , ver para meter s er � iodically laces y oints 1 st retty vent ious' int ed s + obe 1 s I+ cess file ing y3example thon quantity ick read + # cv- lated ive porting spect ults turn uns time $ s aid cript: & econd 6 e % nd. t howing G s imple ngle� xth}- ze of $ $ �Y& �6 5 k a?a - boutp ccessnR umulatec dns sn tivitya ioc llp ongp mountn nc d9b 1 yf red af gumentj so ttempted f Y ributed sd udita backtracec! sedl si isa eginn> stp tweenj$ lksj ockf ootf thn< reakdownj usinessp yh 9 tesn(6 ch allc edi$ pture g hangedd sd lockn sen dn0 ommonc nditione solee tainsi* unts i umulativel decryptiong scendingl riptorh viceb s f isableh ke topm ummyh ring c eachb rlye syg � �a� I!%�IPrint Process File Descriptorspfiles.stpprocess fileRun pfiles.stp to produce a human-readable summary of all open file descriptors of a given process. Specify the process-id as -x PID for fastest performance.process�� c#G�OGenerate backtraces for kernel audit eventsauditbt.stpmonitoring security backtraceAttaches to the kernel audit-log paths (also used by libaudit), and log every record being sent, along with a user-space backtrace of the process that caused it.process h � � � � � � �^9 �@ completed�# uter�G nsider�Q tains � ention� s� vert� `�^8 �@ 2mb�9 4kb�B 512�< 64�3 aY�) & C R U �x5 �t user �" s� ing� variable� ety� ous� ersions�" wait�+ 4 � time� ed�7 ing � 3 keup� s �2 T tches� hat� en � ich � le�E ide� th �0 out� oken� x�4 # xx� � � �4�. G �y Profile Kernel/User Functionspf3.stpprofilingThe pf3.stp script sets up time-based sampling. Every five seconds it prints out a sorted list with the top twenty kernel and/or user functions with samples. Use any of --ldd, --all-modules, -d MODULE, -d /PATH/TO/EXEC to add more symbolic info.profiling � �C� �1�OLog the kernel splitting huge pages into normal sized pageshugepage_split.stpmemoryBecause some portions of the kernel code only work with normal-sized pages the kernel may convert a huge page into a set of normal-sized pages using a split operation. This operation may be relatively expensive and noticeable to latency sensitive programs. You can identify if split operations are occurring with the systemtap hugepage_split.stp script.memory�+� �-/� Print trace of process ancestors for matching exec commandspstrace_exec.stpprocess backtraceThe pstrace_exec.stp script watches each exec operation. If the exec contains a substring that matches the script's command-line argument, it prints out that process and all of its ancestors.process ' ' �V/ �0 takes t ingt4 skt ermination� hat!s ! ' e�q is�" read {) s} id{+ me q 0 d~ sy tampq o*s & okv tal}) race q point� kt iggered{@ un� sefuls rsy via| wait} ing} sv" tch~ er � sq hen u ich t lev ide w th q out| x | G �7 G �ms ]!+�GGenerates a process diagram in DOT form.pstree.stpprocess diagramThe pstree.stp script generates a process diagram in DOT form. For instance, it may be useful on a 'make' command to see all the processes that are started.processzr I+} Signal Counts by Process Namesig_by_proc.stpsignalsPrint signal counts by process name in descending order.process�aq Y'/�)Trace Time Spent in Nanosleep Syscallssleeptime.stpsyscall nanosleepThe script watches each nanosleep syscall on the system. At the end of each nanosleep syscall the script prints out a line with a timestamp in microseconds, the pid, the executable name in parentheses, the "nanosleep:" key, and the duration of the sleep in microseconds.process�fp ;37�=monitor syscall historythread-business.stp_best process syscallPrints a periodic tabular report about the counts of syscall activity of all threads on the system, along with a textual recent-syscall-history for eachprocess �# ocking� g� trace � uck� machine� kes� y� icroseconds� grates�4 onitoring� ot� st� utex � es� name � ! s� ew� nnn� ofile� ptrace� t�; p�) tl� umber � / observed�) ccurred�P f>� ! 0 f�? n"� ' e� pen� ration� s� tionally�* s �5 $ r�3 " der� ed� ther� ut� put�- verflow�& ride� s� parentheses�# ths� er� formance� haps� iodically� files � id%�; limit� ts� rint � � � � �a� � )O�UTrack Time Processes Spend in Various States Using Tracepointsschedtimes.stpprocess scheduler time tracepointThe schedtimes.stp script instruments the scheduler to track the amount of time that each process spends in running, sleeping, queuing, and waiting for io. On exit the script prints out the accumulated time for each state of processes observed. Optionally, this script can be used with the '-c' or '-x' options to focus on a specific PID and its children.process�n� �3W�cMonitor process creation/termination and module [un]loadingprocmod_watcher.stpprocess monitoring syscall tracepointThe procmod_watcher.stp script monitors calls to fork(), exec(), exit(), init_module(), and delete_module(). Event-specific details are also printed out (e.g. for exec(), the file being exec'ed). This script does not require debuginfo.process / / �N$ � ipisZ s ^ kernel Z least ^. ( ikelier[ ne [ N uxZ st \" onger[ major^3 ny\- xmapentries\; y\5 easures ^ icroseconds[ nor^4 ost ^ sY name] umber[ 9 5 : o] f&Z 3 n Y $ J ly(Q pen[ s[ 0 ration[ s\1 9 / tion\I al` r [ der] ut] put ^ standing\. per Y iodic` ally] rint] sZ( ! ocess] esZ orZ filesY ing ^ queue\ read^ % ing^ s ^: 1 ceivedY ordY ference(Q gexp` mote[ quests\ scfZ ript&Y ! econds \% tors` � �N � � G' �#Callgraph Tracing of Tcl codetcl-trace.stpcallgraphPrint a timed per-thread microsecond-timed callgraph of executing Tcl code.general�> /'9�Callgraph Tracingcallgraph.stpsimple trace callgraphPrint a timed per-thread microsecond-timed nested callgraph. The first parameter names the function probe points to trace.general�2 /) �Profile Tcl callstcl-funtop.stpprofilingThe tcl-funtop.stp script displays a top-like view of most frequently used Tcl functions for the each five second interval.general�9 c%-�SWatch a Variable Changing Value in a Threadvarwatch.stp_best monitoringThis script places a set of probes (specified by $1), each of which monitors the state of some context $variable expression (specified by $2). Whenever the value changes, with respect to the active thread, the event is traced.general $ X $ �1 S%1�OWatchdog Timer for Arbitrary Eventswatchdog.stpwatchdog backtraceThe watchdog.stp script provides a watchdog timer mechanism for arbitrary events. The script takes three arguments: the events to start watchdog timer, the event to stop the watchdog timer, and the time in millseconds for the watchdog. If the watchdog timer is exceed, the script will trigger a stack backtrace of the user-process that timed out using pstack. This script can be used to diagnose what the userspace application is doing when a slower than expected operation occurs.general�% =33�=Function Time Statisticsfunc_time_stats.stpfunction statisticsThe func_time_stats.stp script tracks the wall clock time for each invocation of the function probe listed as the first command line argument. When the script exits it prints out the minimum, average, and maximum times in microseconds followed by a count of times that the function was called and a histogram showing the distributions of times.general � �� �H M/�Tracing glib2 memory allocationsgmalloc_watch.stptraceThe gmalloc_watch.stp script from Colin Walters' blog (https://blog.verbum.org/2011/03/19/analyzing-memory-use-with-systemtap/) traces the allocation of glib2 memory using the markers in glib2.apps�lG =1/�WTrace libguestfs startuplibguestfs_log.stpapplication traceThe libguestfs_log.stp script prints a log of when various libgueststartup steps are encountered. The first column is microseconds since the script started. The second column is the time elapsed in microseconds between the previous and current events and the third column is the event name.apps sure quantity between different probe points, aggregate the measurements, and display the results. The first argument to the script is the quantity to measure such as time in microseconds via "gettimeofday_us()". There could be multiple intervals being measured concurrently and the second argument is the means to match the start probe points listed in the fourth argument with the end probes listed in the fifth argument. The third argument is how to display the aggregated data. The data could be displayed as a sum (@sum), an average (@avg), or a histogram (@hist_log). The fourth and fifth arguments are lists of probes marking the start and end of an interval. An optional sixth argument allows grouping of the intervals. One could use "tid()" for the matching argument to measure the intervals on a per-thread basis, but then use "execname()" for the optional argument to group the measurements by executable name. If there is no sixth argument, all the intervals are aggregated into a single entry.general � 9 � �w g%!�WKeep a tally of executables run on the systemalso_ran.stpprometheusThe also_ran.stp script tallies each time a executable is started or a shared library is loaded for execution. This information can be useful to determine what software is actually being used on the system. The script makes the information available via procfs in Prometheus readable format.general�D M17�iCallgraph Tracing with Argumentspara-callgraph.stp_best trace callgraphPrint a timed per-thread microsecond-timed callgraph, complete with function parameters and return values. The first parameter names the function probe points to trace. The optional second parameter names the probe points for trigger functions, which acts to enable tracing for only those functions that occur while the current thread is nested within the trigger.general � &$� � �%/ O!+�ETrace Activity on Network Devicesnetdev.stpnetwork trafficThe netdev.stp script traces configuration and transmit/receive activity on network devices.network� . =##�+Watch autofs4 Operationsautofs4.stpnetwork nfsTrace key autofs4 operations such as mounting or unmounting remote filesystems.network�- M++�qDump of Received UDP/TCP Packetstcpdumplike.stpnetwork trafficThe tcpdumplike.stp prints out a line for each TCP & UDP packet received. Each line includes the source and destination IP addresses, the source and destination ports, and flags.network�W, q'?�mWatch Where Socket Buffers Are Freed in the Kerneldropwatch.stpnetwork tracepoint socketEvery five seconds the dropwatch.stp script lists the number of socket buffers freed at locations in the kernel.network � �Y � �A� 5!�'uprobes-based ltraceltrace.stpprocessThe ltrace.stp script lists calls that the designated process makes through PLTs (procedure linkage tables), generally into shared libraries.process�� Q'+�'Trace Time Spent in wait4 Syscallswait4time.stpsyscall processThe script watches each wait4 syscall on the system. At the end of each wait4 syscall the script prints out a line with a timestamp in microseconds, the pid, the executable name in parentheses, the "wait4:" key, the duration of the wait and the PID that the wait4 was waiting for. If the waited for PID is not specified , it is "-1".process�� U3�}System-Wide Count of Syscalls by PIDsyscalls_by_pid.stpsyscallThe script watches all syscall on the system. On exit the script prints a list showing the number of systemcalls executed by each PID ordered from greatest to least number of syscalls.process f f �= �2 small� ome� rted � ( pace� wn � ed� ends� t�u lit � ting� tack � p� rting� tistics �< pl�Q : 4 1 ubstring� mmarize� ym� bol� s� call � tem� tap�7 . t�7 ake� lly�a sk�E s� erminated �$ & han�? t(� , . "F e��) ) / - n�/ is3� ! 1 D 6 ^ us�D ime0� @ r s�=* tamp� lb� � � �eI S5=� )Time Statistics on KVM Exit Reasonskvm_service_time.stp_best virtualization kvmThe kvm_service_time.stp script tracks the statistics about the amount of time that the processor left the guest virtual machine for each exit reason (for example fixing up a page table or handling an IO operation). When the script exits it prints out the number of times each exit reason was encountered, the total duration of time it left the guest VM, the minimum time, the average time, and the maximum time in microseconds for that exit reason. On Linux 2.6.38 and newer kernel the script can automatically determine whether it is running on Intel or AMD processors. For older kernels with a kernel.trace("kvm_exit") tracepoint that does not have the $isa parameter you can explicitly state the kvm type with a "-G kvm=intel" or "-G kvm=amd" on the command line.virtualization � ] � � M3!�Measure intervals between eventsmeasureinterval.stpmonitoringThe measureinteval.stp script allows quick creation of instrumentation to mea *�@ M+�Color Table for ansi_set_color()ansi_colors.stpformatThe script prints a table showing the available color combinations for the ansi_set_color() function in the ansi.stp tapset.general�] ]A+�Callgraph Tracing with Verbose Argumentspara-callgraph-verbose.stptrace callgraphPrint a timed per-thread microsecond-timed callgraph, complete with pretty-printed function parameters and return values. The first parameter names the function probe points to trace. The optional second parameter names the probe points for trigger functions, which acts to enable tracing for only those functions that occur while the current thread is nested within the trigger.general ) ) �T �, 1 $( a2! 0 ccept% 2close% tivity/ ddress0 es- ka+ ll " so() ndG! re) gument( s$ t, utofs4 . verages$$ basic! e$ ing ) st! yond$" iolatency$ tesize# lock # uffers, t " y $& call" n $ tch(! haracteristics$ lient0 ose% ommand' on " nfiguration/ nections% sumed( unt& s& pu # debuginfo " stination- tails" vices/ isk $ tribution# s$4 o " ropwatch, ump- ration% s+ ynamically " each" 1 � 1 �h I +�UPrint Process File Descriptorspsig.stpprocess signalsRun psig.stp to produce a human-readable summary of the signal handling configuration of a given process. Specify the process-id as -x PID for fastest performance.process� ~ [++{Watch semop(2)/semtimedop(2) operationssemop-watch.stpprocess lockingPrints a timed trace of semop(2)/semtimedop(2) syscallsprocess�T} }59�SGenerate Backtraces of Threads Waiting for IO OperationssleepingBeauties.stpio scheduler backtraceThe script monitors the time that threads spend in waiting for IO operations (in "D" state) in the wait_for_completion function. If a thread spends over 10ms, its name and backtrace is printed, and later so is the total delay.process � % � �@� e'�sTrack Virtual Memory System Page Reclamationmmreclaim.stpmemoryThe mmreclaim.stp script uses the virtual memory tracepoints available in some kernels to track page reclaim activity that occurred while the script was running. It's useful in debugging performance problems that occur due to page reclamation.memory�W� a+#�Summarize Process Misses across NUMA Nodesnuma_faults.stpmemory numaThe numa_faults.stp script tracks the read and write pages faults for each process. When the script exits it prints out the total read and write pages faults for each process. The script also provide a break down of page faults per node for each process. This script is useful for determining whether the program has good locality (page faults limited to a single node) on a NUMA computer.memory ) ) �T' �, eatmydatah nospce triesm veryk xecutableb , I sl itsc fasterh ewf ileb s k ysteme rst f N ore ,& romc synch unci tionsi getm nutlsg reatesti uru a histogramc ryp ooksa il d b entifiedh mplementationse nc % + dicationse ode b watchb 2d ternale oh o|a statj timenh si / tn kerneli ystrokesa wordnS least c ' inenO sn8 st c j snT ogk gere machinej jor b ster g brwatchf X X �%: �N good�; roup� hardware �$ ( s�: ving� ighest�& owever� uge� 6" page�8 / ndred�S w � identify �/ $ f� 0 % mplementation� rove� n"� 0 @ &; direct�D s� ternal� o � vocations� s,�' 3 # %" # t� G % s �3 ! kernel"� s� malloc � larger�> tency �* [ eaks �< L imited�? s � ne� st�) s� ved�- ocal� ity�< king� g� > west�" m�> achines�4 jor� 0 lloc � ny� pped�? ing� rked�$ tches� ing� y� be�X g g � �0 pattern er ingpong ped oints wer rint s obe cess - 8 F or * s fs 4 - gressively metheus 6 / vided" s stack= y2example thon ran eadable 7 0 cord s8 duce flecting gex peatedlyB ort turn un time' script2 # " $ econd s" e 7 t hared ow ing % 8 imple ulated ze of5 liding owerM oftware urce9 pace/ eed nds tack3 pgames( rt ed tes istics s op watch es �q� })�;Track Virtual Memory System Actions on File Backed Pagesmmfilepage.stpmemoryThe mmfilepage.stp script uses the virtual memory tracepoints available in some kernels to track the number of faults, copy on writes mapping, and unmapping operations for file backed pages. When the script is terminated the counts are printed for each process that allocated pages while the script was running. The mmfilepage.stp script is useful in debugging leaks in the mapped file regions of a process.memory $ $ �Y+ �6 10ms} 2~ 0x, 5v 5 v aGq ! bout v ll s so� n t d3q' 1 8 pplied| res gumentt s t q A veragew backtrace {: ! s} es ? ing� st v tweent y r c| allx s x ny hanges{ ng{ ombinationy mands pletion} nfigurationv' tainsv& entionw sw untz s r pu{ reation� d} ebuginfo�' faultx lay}* ete� scendingr riptors tails� iagram s fferent{" oes�$ ts urationq) e� achq d�! mulates| ndq tire| ryt \ �\ �(w S%7�7System-Wide Shared Futex Contentionfutexes2.stpsyscall locking futexThe script watches just shared futex syscalls on the system. On exit the futex's key, the number of contentions, and the average time for each contention on the futex are printed from lowest pid number to highest.process�vv s#;�1Periodically Print Process Information With Historyproctop.stpprocess scheduler _bestEvery 5 seconds, print out a list of 25 processes that took the most system time with information about the processes. Includes information on processes that may have exited while the script was running. The script contains configuration options listed in the script source.process $ �$ �a �F mounting. ulti$2 name# d2" ��_ �B every (+ xample$* ec( s(3 noop( ution( its! file" w ! ! �\L �< increased�a dication�. fo�3 / : rmation-� 6 � itially� strument� terval� o�5 ctl� s9� ! & > ( sues�C t'�# " ! s� / . , kernel �' 1 s� journald� retprobes�\ update� larger�' st � tency �@ r�T S P dd�& " - ead�I H F king� st �0 ibc� mitations�= < : ne� i � times � ux� st � . s � og� nger� wer�X W T machines� l k h kes�! nipulate� ximum � . y �] ) be�= B A ? emory)�$ ssage�( icroseconds�; �] �> lookups5 mac7 y1 emory2 name1 s= et8 filter1 stat9 top6 work�1 w? fs5 d = top5 ot1 umber4 of92 ( n4 ' ly 1( ! peration= s= tion<# s< r 1 / dered;/ iginator?. ut3 going 1 put4 verride> packet3 s1 ! ir 4! s;< rameters: ent@ ticular< eriodic6 ally< id2 ort1 # ssible= rint3 ed2 ing1 s-2 % obes1 cess2 " ed6 & & �W< �2 per�, formance � l� faults � id �# % ortions� revious� int � 0 ed� ( * s$� & oblems�! c� essM�& + ( - h es� gram�9 O s �, ] vide�% strace � ran�A ead � 3 claim� mation�' duce� fuses� gions �A P latively�% ports� quest� stricted� oot � ! un � ning� 3 5 s� s�= H ame� cheduler� ript_� ; ! _ 2 eeker � nsitive �+ \ rvicing�! t� hare� d� ort�, w� ignificant� mple� ngle�B ze�; d � ccumulates ! tive@ s ' ually ? lso mount # n # d7 % " imation s i y pplicationH rbitrary chive * e! gument s ound s t & tribues tes vailable 2 + erage$ oid backtrace4 s ll sed e ( A A cause6 ing ! st lock s old unces reak er y + c alled4 graph n @ @ eiling lassic earing ock olor s2 mbinations mand5 parable1 ilation lete nstraints rresponding unt- pu � � � � 0 762 19041� a � ) ction�5 nd�) pplications� re� based� ecause�3 log� com� developers� oing� entry� find� or� help� tml� tp� is� livejournal� mentioning� onitors�$ need� occurred�6 n� ce�< perations� ut�/ polling� rocesses�C filing � rather�0 ecords�* turned�2 screen�; ipt� econd�> how� ome�4 tp � ystemcall�% s�, than�1 t� e � imed�. out � o� ol� p�A wenty�B udrepper� pdates�9 sed�& with�? � � �A! � virtual I izationI mIA waitMQ edM> ingLM sM ltersH s I6 | tchH yD ellL� hen F- ( T ther I_ ichC ideA llLc# thA' l k rittenJ0 youIy $ �$ �a2 �F like� mits � ne � kage� st � s � A ved $�`1 �D dispatched� maxmapentries� ue�' ration�( each� f Y N Y �q�$ ?1=�MLog recent free(3) calls.last_100_frees.stpmemory process backtraceThis script reports on the last few free(3) libc calls done by processes (possibly restricted by stap -x/-c), along with a userspace backtrace at those moments.memory�.�# �?�Summarize the time each process spends clearing huge pageshugepage_clear_delays.stpmemoryWhen a huge page is initially used by a process it must be cleared of its original contents to avoid leaking information between processes. On x86_64 machines the huge pages are 2MB in size, 512 times larger than a normal 4KB page. Thus, clearing a huge page can delay program execution by a hundred or more microseconds and maybe noticeable to latency sensitive programs. This script will tally the number of times that each process triggers a huge page clear, the total time in microseconds spent clearing the pages, and the average time in microseconds of the huge page clear.memory 5 @ ccess�2 $ ed � s � ount�A ross� tions� ve�F ity� dded�9 r� ess� * ffect� ll� ocated �- / ion� w � # so�$ mount� ncestors�" dJ� ' 2 .+$ onymous � re$� 3 ' O ) 8 * gument� s� ttempt � vailable� erage � { backed� trace� e$�! $ G - , cause � tter�@ reak�' point �# ' y�G I c� all� n � . F , M # ertain�A hildren�H ode� llapse �( ing� ect� mmand� s� � H� � �E+ �'�aSummarize read() and write() syscall latency (non-debuginfo)rwtime-nd.stpioThis traces read() and write() syscalls, producing a histogram summary of their durations (aka latencies).lwtools� * �A-'�;Measure the distribution of file system synchronous read and write latency (non-debuginfo)fslatency-nd.stpio filesystemThis dynamically traces two common file system functions: do_sync_read() and do_sync_write(), and reports a histogram distribution of latency. Many, but not all, file systems and workloads use these functions. Tracing their time provides one view of suffered file system latency.lwtools�5) {-�GTrace open() syscalls showing filenames (non-debuginfo)opensnoop-nd.stpioThis traces the open() syscall system-wide, to show which files are being opened, and by who.lwtools � � � �� �)'�CLog Failed Process Memory Allocation Due to Overcommit Limitsovercommit.stpmemory limitsThe overcommit.stp script prints a line each time the kernel refuses a memory allocation request from a process because of /proc/sys/vm/overcommit* limits.memory�B� y)�aTrack Virtual Memory System Actions on Anonymous Pagesmmanonpage.stpmemoryThe mmanonpage.stp script uses the virtual memory tracepoints available in some kernels to track the number of faults, user space frees, page ins, copy on writes and unmaps for anonymous pages. When the script is terminated the counts are printed for each process that allocated pages while the script was running. This script displays the anonymous page statistics for each process that ran while the script is active. It's useful in debugging leaks in the anonymous regions of a process.memory H H �:_ q#%�[List Executables Reading and Writing the Most Dataiostats.stpio profiling The iostat.stp script measures the amount of data successfully read and written by all the executables on the system. The output is sorted from most greatest sum of bytes read and written by an executable to the least. The output contains the count of operations (opens, reads, and writes), the totals and averages for the number of bytes read and written.io�x^ �)%�;List Executables Reading and Writing the Most Data by Deviceiodevstats.stpio profiling The iodevstats.stp script measures the amount of data successfully read and written by all the executables for each io device on the system. The output is sorted from greatest sum of bytes read and written to a device by an executable to the least. The output contains device major/minor number, the count of operations (reads and writes), the totals and averages for the number of bytes read and written.io T T �,1 � +E�qTrace threads sending network traffic to given host and/or portwho_sent_it.stpnetwork trace traffic simpleThis script traces outgoing network packets using the netfilter probes (not requiring debuginfo), printing the source thread name/id and destination host:port. It may be filtered with the_dport and the_daddr globals, e.g., to watch only for DNS traffic (port 53), and/or only to the localhost (127.0.0.1).network�z0 o+ �SFind Which Client Is Removing NFS Files on Servernfsd_unlink.stpnfs diskThe nfsd_unlink.stp script lists the ip address and file name each time time a file is being removed or unlinked by the nfsd. This script is run on the nfs server.network P �1 P �]�' [! �1 Show Processes Doing Polling Operationstopsys.stpprofiling The topsys.stp script lists out the top twenty systemcalls for the previous 5 seconds. The output is sorted from most frequent to least frequent.profiling�S�& [#+� Show Functions Taking Longer Than Usualfntimes.stp_best profilingThe fntimes.stp script monitors the execution time history of a given function family (assumed non-recursive). Each time (beyond a warmup interval) is then compared to the historical maximum. If it exceeds a certain threshold (250%), a message is printed.profiling�t�% M71�E Count Times Functions Are Calledfunctioncallcount.stpprofiling functionThe functioncallcount.stp script takes one argument, a list of functions to probe. The script will run and count the number of times that each of the functions on the list is called. On exit the script will print a sorted list from most frequently to least frequently called function.profiling $ $ �Y, �6 errorsx snoopx vent� ryv xec� utable q st edz ingt@ itw edv failingx urex stest ew| ile teredy ingy rstx or s k� ms und{ rom w" unction} s{> tex w es2w g |! enerate} ss iven sx! reatestz handling vev eightx$ ierarchies| ghestw' storyv uman idt& * sy f} n(q * cludesv dividual| formationv it� stances tervalx o } st " ts s} justw kernel{= y q& later}% eastz ineq sx' st v edv) oadingt. cking w osely| westw# makes �a�2 [# �7 Show Processes Doing Polling Operationstimeout.stpprofilingThe timeout.stp script is based on a blog entry (http://udrepper.livejournal.com/19041.html) mentioning a need for a tool to help developers find applications that are polling. The timeout.stp script monitors systemcall used for polling and records the systemcalls that timed out rather than returned because some action occurred. The script updates the screen once a second with the top twenty processes.profiling �]. �> produce sig trees readable ceivingu- portx& ingx quire�& un ning v# sw chedulert reenx# iptCq ! ! econdx s v es mop ~ timedop~ ndingu" tu t| haredw elly ould t wing z ; igr monu nalr sr leepq, ingbeauties} timeq o}& urcev- pecifict edu sx y nd} s} tq tarteds te {7 isticsy pZq race| ummary yscall.q sq ! timesy tem(q callsz tapy tabularx tex py� %"* ing�v unt�4 s�& ( ' w �( reated�7 ing�( on�F debugging� ; K lay�N s� tailed�L ermine �$ J ing�6 ifference� splays�8 oing� wn�( ue �$ plicated�- ring� e�D ach>� * , g lapsed�? very� xamination�M ed�H clude�@ ec � name� utables� ion�P it � s�2 $ pensive�& failed� ult � s � N ile� % tered�. rst�3 lurry�* or2�B 9 ktracker�S rees� om �! unction� tex � es� generate� s� libc � % % �X �4 stpe 6 ream uff upport ystem % tap table kes llies y pset erminal tedH hanN t*( # / 0 9 3 eq $ ! m & rmal is) > ose - read 0 e ottle d , ing ugh iles me& % " d : r s' o*, # * # race ing * ks ial gger $ 1 ype s " underline til D se d " B ful r / 7 space G s ing < tilization values riable ous ! ! �\# �< deviceY seeks` ifferent` sk` maxmapentries\G uring] eachY ducation(Q mergency(Q very[ # xceed\6 ecutable ^+ % s] itZ% pensiveZ filesystems[ ve] orDQ 0 8 requency Z- om ^ unctionZ generates` ing] iven` reatest ^ uru(Q highly[ stogram` s` rical(Q t[ i] dentifyZ nZ+ & creased\E terZ processorZ ruptY sY val] o7[ blktime\ devstats^ stat_ s_ top] piZ# = +N = �c o'%�Tally Reschedule Reason During AIO io_submit Callio_submit.stpio backtraceWhen a reschedule occurs during an AIO io_submit call, accumulate the traceback in a histogram. When the script exits prints out a sorted list from most common to least common backtrace.io�Zb W)�CMonitoring Reads and Writes to a Fileinodewatch.stpio The inodewatch.stp outputs the executable name and process id each time a read or write occurs to the specified inode on the specified major/minor device.io�]a 1!9�OMonitor TTY Typingttyspy.stpio tty monitoring guruThe ttyspy.stp script uses tty_audit hooks to monitor recent typing activity on the system, printing a scrolling record of recent keystrokes, on a per-tty basis.io�r` c+�aHistograms of Seek Behavior for Each Devicedeviceseeks.stpdiskThe deviceseeks.stp script generates a periodic histogram showing the frequency of different sized seeks (in sectors) on each device, or optional given device-name regexpio �y�b0 747 again�M ll�4 nd�$% re�0 gument� s� ( s� based� e� / ing�2 can� / urrent�C display� s�3 first� unction� gets�J happening�) if�- n�L dicate� s� t�2 name� o�. on�! r� ut�A passed� id� rint�% ocess� filing � sched � ule�@ d�K r� ' ript� econd� tp � witch � es�( takes� sk� s�? he� is�8 o�- race� d�F wo� understand�= sed�; what� en�H ich�> ill� th�* nd� try� vents� ry� xec� utable � ed� it � fastest� ile� s� lurry� ocus �7 & r*� 4 k� tracker � rom� utex� generally� te� iven � libc� reatest� uru � handler� eavy�) ierarchies� stogram� s�7 uman� id� entified� y� f �5 # n� ! itialization� struments � terrupt� s�> o� voke� o� rq� s� : H t� = s�= kernel� s� y�& large�! tency � dd� east� ngth�: ibaudit� c�% raries� � � �<F � '/�/Provide elapsed times for Passes of SystemTap script compilationstap_time.stpapplication traceSystemTap has multiple passes to convert the text of a SystemTap script into instrumentation that actually collect data on the system. The stap_time.stp script uses the markers in SystemTap to note the amount of time that each of the passes requires. When SystemTap completes pass 4 (compiling the instrumentation into a kernel module) the script print out the script name followed by the amount of time in milliseconds required for Pass 0 (command line option parsing), Pass 1 (script parsing), Pass 2 (elaboration), Pass 3 (code generation), and Pass 4 (module compilation). The resulting data can be analyzed to determine if there are issues with the amount of time that systemtap takes to generate instrumentation.apps P �o P � % � 3)�gShow socket lifespan, from accept() to close() (non-debuginfo)accept2close-nd.stpnetwork socketThis traces socket duration from the accept() syscall to close(), and provides details on the lifespan of these passive connections, showing the distribution as a histogram.lwtools�t$ y/�EMeasure block I/O latency distribution (non-debuginfo)biolatency-nd.stpioThis measures block I/O latency (storage I/O, ie, disk I/O), and shows the distribution as a histogram. This can be useful to identify the characteristics of I/O latency, beyond the averages shown by iostat(1). For example, to study I/O latency outliers, or multi-modal distributions.lwtools�# s+�Measure block I/O size distribution (non-debuginfo)bitesize-nd.stpioThis uses the kernel tracepoint block_rq_insert to read the size of I/O. The output includes the name of the process or thread that was on-CPU when the I/O request was inserted on the issue queue.lwtools % % �X; �4 memory}� icroseconds � V nimum�< or�0 ute� sses� manonpage � filepage �4 reclaim � odified�0 re �K I ust�+ n�4 eeds � ode�- s� rmal� A ticeable �( Y uma�E ber� c observed� ccur�# red� ing�4 P + s� fP� ) 1 & @ K n2� D & ' e� ly � 1 peration� H s�2 ) r�4 ? I ut� ' put�+ vercommit � view� page5� * % " s*� ! 7C rent�< ths� # # �ZJ �8 by)� / D C % A tes� c� alled�" ' v s� n �2 ; : � 8 pabilies� ) tes�) ies� lities� used�D ertain�$ heck � ildren� lear �R ed� ing�.- ode� 20 llect� s� ming�* mand�h � N pared� nsumption�* tainer �O ents� rol�4 rrelate�& unt � M s�P O L pu�! d�) % 0 ata�0 / - elay�3 s� pend� signed �? ? tails �4 C ermine� ing�& vice�G s� iagnose�B sappearing�Q P M k� kretactive�c oes�_ ^ [ ing� ne� river�_ uring�: each3�L # S liminated�3 2 0 nds� tries�A @ > rror �O N s� & & �W �2 dport1 ue2 mp 7 ring6# e1$ ach2 ncountered 4" ! tire@ ry2 very6 xecutable2 it; file= ter: ed1 ing< ve6 lags7 ocus< r 1) mat 4 und> rom;0 ull2 g1% athers5 id@ ven1 lobals1# reatest;1 ouped9$ s9 uru> h<" as> ighest< ost1 id 1 f= mprove> n2) cludes 4 oming? rease> dexing: ividual9 formation 8 it> ial> terfaces6 val6% p 4 * s4 v4 4 s$2 . . & t1 - json4 keep5 rnel> latency> east;3 ine 3 st6 ing6 s<$ ocal: host13 * ��B * �� 7#�ITrack SIGKILL Signalssigkill.stpsignalsThe script traces any SIGKILL signals. When that SIGKILL signal is sent to a process, the script prints out the signal name, the destination executable and process ID, the executable name and user ID that sents the signal.processu� E)y Signal Counts by Process IDsig_by_pid.stpsignalsPrint signal counts by process ID in descending order.process�� C+/�#Trace Creation of Processesforktracker.stpprocess schedulerThe forktracker.stp script prints out a time-stamped entry showing each fork and exec operation on the machine. This can be useful to determine what process is creating a flurry of short-lived processes.process�)� Y/O�track wakeup-to-dispatch latency statssched-latency.stpprocess scheduler time tracepointThis script periodically reports a histogram of the latency between a task (thread) being woken up and it actually being dispatched to a CPU: the amount of time it's spent in the runnable queue.process + + �R �( for3C# I 4 9 &# rom B# unctionB4 sB gI� enerateFu ionFZ lib2H mallocH uest I- ruA handlingI$ sF 4 ] veD u eldL} pL xadecimalC istoricalN olderLq ing L6 e ttpsH ifA h ~ n)B ) ! L d comingA reasedC0 dicateM s A/ 3 formationM_ putJ strumentationF'D sB tegerA* lId" oF) oI& pD s-A a aIw suesFk tD ,% " sC keepD rnel B 7 Y G sIj vmIn largerC$ stB; eftI- ngthC ibguestfs G startupG mitC- neB $ 8 M � ux B S stK s D &$ u b�&u �.Y [7 �= Record Interrupts on a Per-Device Basisinterrupts-by-dev.stpinterruptThe interrupts-by-dev.stp script profiles interrupts received by each device per 100 ms.interrupt�X K/'�1cve-2015-0235 security band-aidcve-2015-0235.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�W K/'�1cve-2017-6074 security band-aidcve-2017-6074.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�V K/'�1cve-2015-3456 security band-aidcve-2015-3456.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids U U �( �T hack eader isth ogram g F ow S i f� n ' ) cludes formation + strumentation teractive val z s 0U o� vocation ked C o s! 7 V & t * kernel y board codes hack strokes language eroy vel ife ke stedA sp ogi make s rkings tch< ing� ximum 1 eans: sure l d2 interval val ments� mory icrosecond s ( 4 ght nimum . ode ified " ule nitoring s re * st ultiple/ name � s ested 8 o� ises umerous o ccur2 c �c �_ �B 015 117 0> 2714 4: 531- a-2 ��, �\ unlink0 ed0 mounting. se "% ful$ s# various' iew "- ( was# tch, hen# re, ich ( o) ide& ll( th& orkloads "$ s( ld! rite" s" � {Np � �c8 i/;� Tracks time between packet queue and transmit.net_xmit_json.stpnetwork statistics jsonThis script tracks time between packet queue and transmit. The information is provided to userspace via procfs in JSON format.network�[7 3%+�IDump of STP packetsstp_dump.stpnetwork trafficThe stp_dump.stp prints out the packet contents. Each block contains the STP protocol ID, version ID, flags, root and bridge MAC addresses, and various times.network�*6 y!+�%Periodic Listing of Processes Using Network Interfacesnettop.stpnetwork trafficEvery five seconds the nettop.stp script prints out a list of processed (PID and command) with the number of packets sent/received and the amount of data sent/received by the process during that interval.network�5 S#)}Keep track of NFS server statisticsnfsdtop.stpnfs statisticsThe nfsdtop.stp script gathers and displays NFS lookups,network = = � �_ �B thez1 re2& is1 read1 s1 ime 2 s7 o/1& 2 tal 4% $ race1 s1 ing: k 2 s8 ffic#1+ nsmit8 uples: ype>! s< uid ? ser: space8 ing1 tility: value> rious7 ersion7 ia8 wait2 tch1' es 4 eb> hen>, ever@ ich> o1 ide4 ndow> th1 orkloads>" xmit8 � � � �G�6 = �) Profile Kernel Functionspf2.stpprofilingThe pf2.stp script sets up time-based sampling. Every five seconds it prints out a sorted list with the top ten kernel functions with samples.profiling�C�5 s% �a Show the Period of the Various Timers on the Systemperiodic.stpprofilingThe periodic.stp script uses the kernel.trace("timer_expire_entry") tracepoint to collect data on period and frequency of the various timers on the system. The script displays a sorted list of the timers observed on the system from most frequent to least frequent. The script needs to be compiled with the '--all-modules' option to produce list the function names. Optionally, this script can be used with a numerical argument to indicate the interval in seconds between printing output.profiling names) s) ystem " s. nd0 ve, lags- or $) k( reed, om % slatency* slower" unctions " general! hello! world! istogram$ i# d& entify $ e$ f(, mmediately! plemented! n ! cludes # ing ' - sert# ed#% o-" stat$' p - s" sue#( t ! kernel # y. ill' snoop' latencies+ y"2 ifespan% ne ( sts , ved( ocations, wtools2" many " eans( sure # s$ ssage! odal$3 $ $ �YK �6 every � xceeding�: 9 7 s�" ssive � ec�. * 5 utable �$ T' s� ion� 5 it�$ E s�. : - + periencing�3 failed�9 mily� ew� ile� line � s� nd� ve � low�5 ntimes � or$� ] M * R' Y bidden�# m� at�* ) ' ree� s� quent� ly�. om$�+ , 4 3 1 unction �3 q callcount � s� g�F athered�! ive�+ n� ood�- reatly� uru�D handle�c d� r �; ing� 5 ve�j elp�A s�$ istorical� y� uge � " page� ndred�8 idle�/ f� Q Y n2� #5 & N $ @ clude�< d�� � � ��/ o/W�9 Provide a per-process syscall tally on the systemsyscallsbypid.stpprometheus process syscall tracepointThe syscallsbypid.stp script tallies each syscall for each running process. This information can be useful to determine the activity of various processes on the system. The script makes the information available via procfs in Prometheus readable format. When a process exits its data will be eliminated from the prometheus output. To avoid exceeding the storage limitations of SystemTap older entries maybe overwritten by newer entries. This can lead to some active process syscall counts disappearing and/or later reappearing with a lower value. Also note that the script does not properly name syscalls for 32-bit applications running on 64-bit machines.profiling F F �7 �r processes 2 fs8 tocol7 vided8 queue8 received6 lated9 mote: quest?1 s= iring1 transmission 3 oom2( t7 un> s9 ' cript@1 s: econds6 en;7 nd2 ing1 t1 rver 5 how@ ing; imilar2 ple1 ular9 k2 ocket< s 9 top <$ rted9" urce1 " pace2 tamp2 rt2 t@ istics5 op2 ped>/ pg1 ream2 ummarize< y 4 ystem4 task@ cp;2 ipstat9 hat 6$ � "N � � ( �-�QTrace process exec() with command line argument details (non-debuginfo)execsnoop-nd.stpioThis can identify if CPU is consumed by short-lived processes, by tracing new process execution. It works by tracing exec() from the fork()->exec() sequence, which means it will not catch new processes that only fork(). It will also show every exec(), including those if a process re-execs.lwtools�Q' �-�]Trace kill() signals showing process and signal details (non-debuginfo)killsnoop-nd.stpioThis traces signals system-wide, including those sent by the kill(1) command, and shows various details.lwtools�[& s3�Count syscalls with process details (non-debuginfo)syscallbypid-nd.stpioThis traces syscalls system-wide, and produces a summary report showing their counts by process ID, process name, and syscall types.lwtools �oQ �b where �' ther � ich �% le� ill+� E 1 0 i . th/� V * U b R ritebacks� s�( ing� x� 86� yes�J H H �5( �n messagek icrosecondsnC1 nor b onitor a inga se stc name b - J fs i umberi &" ol bservedl sn ccuro s b f6a & n2a ej penn edn sslg rationsn#A r b derl therh utc 4 put i s b pairn6 renthesesnL ticularo er a iodm icp idnE resumedh intk inga sc 3 ocess#b esh filingi videsj rawf eadb 8 ingm s b sonc cent a orda porte se % schedulec sj @ � @ �> s �iReport opened files whose names match a given regexregex.stpregexUses the regex functionality to detect opened files whose names match a pattern given on the command line. If no command line parameter is given, demonstrate by filtering for files that end with an extension showing them to be an archive.general� O)�SystemTap python 2 support tapsetpy2example.stpsimpleA python support tapset that displays backtraces and variable values general�p �-/�Monitor Intel processors for throttling due to power or thermal limitscpu_throttle.stpprometheus limitsIntel processors have hardware mechanisms that will reduce the effective processors speed to avoid exceeding set thermal or power constraints. The cpu_throttle.stp script monitors when CPU throttling occurs and accumulates the amount of time in milliseconds that each processor is throttled. The script makes the information available via procfs in Prometheus readable format.general k d sj# rets g torf sf imple f omeh rtedc pacee ecifiedb ntn sl g taph tisticsj usm pba ubmit c mmarizem ppressesh witchk filek yscalln s h p tema tape tabularp kesj llyc en l xtualp hatm eqa mh see ise readp sp imeb stampnA lsg o"a % p l tali3 racen backc iol 2o kl ffic g } } � �p0 370 a� nd� call� s� ombination� errno � or� s� xit� frequently� list� most� occur� f� n� pid� rocesses� filing � vides� return� script� how� ing� orted� tp � ystem� the� which� �� � �=� m/�[Trace processes running out of file descriptors rlimit_nofile.stplimitsThis script watches processes being scheduled and which try to allocate a file descriptor without luck.process�'� M!�[Print Resource Limits of Processplimit.stpprocessThe script prints a variety of resource limits for a given pid, like /proc/$$/limits on recent kernels.process�� e%7� Disable ptrace from Hierarchies of Processesnoptrace.stpprocess security guruBlocks ptrace(2) attempts from processes identified by stap -c/-x, as also specifiable from /proc/systemtap/stap_XXX/ control files. Processes may be added or removed from the blocked list.process�[� ]-#� Override default new-pthread stack sizesthreadstacks.stpthread guruOverrides default NPTL pthread_create stack size for all new threads created by target processes. Reports one line per process when the related glibc variable __default_stacksize is updated. Moot for glibc versions that support $LIBC_PTHREAD_DEFAULT_STACKSIZE_NP.process ! !0 �\I �< 100 � d 250�& mb�" 3� 2�e d a 4kb�+ 5� 12�% 64� j i f ab� - ! " ,- ( +- $ ,"! ), bout � U ccumulates� ion� tive�L K [ I ity� ually�4 dd�0 , 7 resses� ll � ' { # . ocation� s� ong� so�Z Y V n�L dN� <" R ! Q ! &O N 6 y�6 $ + pplication�M s�g f c re�) ! gument� s� ssociated� umed� t� vailable� $ # ! erage �^ , oid� 9 8 6 backtrace� s� sed � is�" e"� $ # ` ! ing� st� tween� yond� it�f e b � � � �j�4 � � Show Which Processes and System Calls Return Errors Most Frequentlyerrno.stpprofilingOn exit the errno.stp script provides a sorted list showing which combination of PID, system call, and error occur most frequently.profiling�&�3 u-1� Display the Task Switches Happening in the Schedulersched_switch.stpprofiling functionThe sched_switch.stp script takes two arguments, first argument can be "pid" or "name" to indicate what is being passed as second argument. The script will trace the process based on pid/name and print the scheduler switches happening with the process. If no arguments are passed, it displays all the scheduler switches. This can be used to understand which tasks schedule out the current process being traced, and when it gets scheduled in again.profiling ccepted? ing? tivity< ddress!4 + es7 ll4 ow< ong 4 mong 4* ) unt6 n@ cestry@ d@1 " ( re9 ttempts@ background> sed: e1 cause2% en> ing2 st 4 tween8 lock7 s2 ridge7 uffer2 y 4 % tes 4( ' changes>+ lient= ollected9 s9 mmand6 plete= uter? ngestion> nect@ ion: s ? tains7 ents7 inues2$ unt 4 ' reation? wnd > daddr1" ta 4 ebuginfo1 fault> stination1 isplay9 s 5 ns1* one: + + �R" �( 0056T 235X 600R 728S 10\$ 0Y 00\H 2008R 12T 3Q 4U 5 V 6S 7W 94Q 3456V 6074W 7169U aY ' ctivity] id0Q s(Q ll ^ owed\< mount\ n ^* $ d Z re[ verage\ s ^@ 7 backtraceZ sZ ndHQ sisY e\+ havior` lock\ yY tes ^"% callZ n\* se\? s\( useZ sZ ingZ ommand\M pletion\ utes\ nfigurable[ tains ^1 + ended[ unt ^7 - ve0Q data ^ efault\8 scending Z, vY Q Q �+�- �;/ �] Monitor which executables use ioctl syscalls and what kernel code is handling the ioctlioctl_handler.stpprofilingThe ioctl systemcall is used to manipulate devices setting or special files. The way that ioctl syscalls are handled depend greatly on the device the special file is associated with. Using strace to monitor the open and ioctl syscalls may not give a good indication of what kernel code is actually handling the ioctl operations. The ioctl_handler.stp script is designed to provide more details. On exit the ioctl_handler.stp script provides a count of the ioctl syscalls for each executable run on the system. If there was some special device driver code used to handle to the ioctl, the output will have a tally of the times the function name and module was called for that executable. The "--all-modules" option should be included on the command line so the script can provide function name information.profiling �y�b0 491 a� nd�$ s� based� est� roken� cpus� down� every� five� unctions� it� kernel�% list� occupying� f�" ut� percentage�! rints� ofile� ing � sampling� cript� econds� ts� orted� tp � the� read � s� ime� s � op� wenty� up� ser�# with� Z �Z �AK a);�E)Tally the Number of User-Space QEMU Eventsqemu_count.stpvirtualization qemu kvmThe qemu_count.stp script tallies the number of times each of the user-space qemu probepoints is encountered. When the script exits, it prints a list of the number of times each user-space qemu probepoint is encountered.virtualization�_J y#A�i)Tally the Number of User-Space QEMU IO on Each IO Portqemu_io.stpvirtualization qemu kvm ioThe qemu_io.stp script tallies the number of times each of the IO port on the guest virtual machines is touched by a input or output operation. When the script exits, it prints a count of the number of times each IO port read and written.virtualization Q Q �, �\ codeB Y linH lectF umnG mmandB, 7 L � pilationF _ ingF2 letesF/ ulsoryLF nnectedD% sideredD# tentionL s C vertF unt J% urrentG' veN data B Q efaultC ltaB termineFg ^ isplaysC maxstringlenC5 oesIs ropA# pedA ingA sA ueL rationI: each'B ( ! i ducationN laborationFV psedF ementB$ mergencyN nableL, d E � counteredG 7 oughC" tryB& ventG/ sG( ryM/ xampleI ceededL` ssive L ecutingE onE itB8 " sA9 + ! plicitlyI{ fileB ndingD rstB % 8 veM0 xingI ollowedB) @ B � � �T�, I?� Profile Kernel/User Backtracespf4.stp_best profiling backtraceThe pf4.stp script sets up time-based sampling. Every five seconds it prints out a sorted list with the top twenty kernel and/or user stack backtraces (on a per-cpu basis). Use any of --ldd, --all-modules, -d MODULE, -d /PATH/TO/EXEC to add more symbolic info.profiling & & �W4 �2 spent � tack� size� mped� p� te�& s� s� ealing� p`� uffering� mmary� pport�$ ymbol� scall� s �! tem � calls� tap� tables� rget� sk� hat%� / # O eq� ief � s�+ read � s� tacks� ough� ime#� < s�1 tamp� o7� / race � point� s� s� k� s� y� underlying� p� dated� robes� se� d �/ ful� " " �[ �: 0 A. K 3H 1FQ 4634P 9H 2 FU T 011H 5O 6N 8P 3FX 8IV 4F1. 5195N 6IU 5536C6 7547O a6A( - N ! - boutI ctuallyF ddressD idN sN llA ( ocationH sH mdIf$ ountF$!- w nI% alyzedFe ingH d6B. [ & H / )0 B pplication F sE re!A j � gumentsL# sL� ciiC utomaticallyI] verage IF S backtraceL� sL-T ndN e A% ! " d � enL5 stI tweenG# ig L G kl L logH othC y B* A C cB alledB n A$ . c \ usingL lientD sD A A �< �| passingC2 erM iodicallyD sM hp E idB0 ortJ sitiveA) revious B% % intC) ; /7 ed L� 5 s B - # % obepointK& sK cess L3 J esLL 2 orI sIg tocolA sA videF qemuJ reachedLA dJ. sonI sI centD ferenceN questsD ireM, dFH sF, spectivelyB9 ultingFa unningIb s B criptJA ! %3 D econdG C Z sM1 urity$N rializationL verD iceI houldA2 imilarM^ nceG ocket B pace J ecified A < ntMc tampLg pF rtedG upG teI| isticsD s M epsG illD$ ! ! �\ �< shown$% s" ignal' s' mple! ze# low" er" ocket% urce- torage$ p\! udy$, uch. ffered "/ * mmarize+ y & ync " hronous" scall% bypid& s& tem "+ & s "" tap! tcp- dumplike- han" t" $ eG# ir") $ n! se"& ! is8" ose ' . read# shold" ime"* % o# race" point # s(" ing"( # ffic - nsmit/ wo " ypes& udp- # # �Z0 �8 1�? 2� a?�9 ( ccumulated�" tivity�* ually� dded� ll� ocate� ong� so � mount � nalysis� d4�' , / y� re� rays�% s � t� taches� empts� udit� bt� backtrace � s � sed� e�. ing� st� tween � 5 locked� s� y"� c�2 ! alls� n �- used� hildren�> ompeting� ntention � rol� unt� s� pu � reate� d� ing� on� ycle � s� d� efault � scending� riptor� s � ignated� tination� termine � isable� patch� Y +� Y �7� k-/�?Watch a Kernel Symbol Using Breakpoint Hardwarehw_watch_sym.stpmemory watchpointThe script will watch accesses to the starting address of a single kernel symbol and prints a traceback each time the symbol is accessed. This script needs to be run as root to allow access to the breakpoint hardware.memory�� {7�sLog the kernel collapsing normal pages into a huge pagehugepage_collapse.stpmemoryThe kernel may attempt to group many normal sized pages into a single huge page to improve TLB performance. This operation may take a significant amount of time and affect program performance. You can identify if huge page collapse operations are occurring with the systemtap hugepage_collapse.stp script.memory�Q� _-)�Overview glibc malloc internal operationsglibc-malloc.stpmemory processThis script reports on internal statistics of the glibc malloc implementation, as used by a process restricted by stap -x/-cmemory � � �"�40 788 a� * ll�7 nd� rgument�I be�3 tween�P can�C ollect� mpiled�4 data� isplays� entry� xpire� frequency� t�+ om�) unction�> in�N dicate�K terval�M kernel� least�- ist�! modules�8 st�* names�? eeds�1 umerical�H observed�% f� n � ption�9 ally�@ utput�R period� ic � rinting�Q oduce�; filing � script � econds�O how� orted� tp � ystem� the� is�A imer� s� o � race� point� used�E s� various� with�5 > �#� m//�Watch a Kernel Address Using Breakpoint Hardwarehw_watch_addr.stpmemory watchpointThe script will watch accesses to a single kernel address and prints a traceback each time the address is accessed. This script needs to be run as root to allow access to the breakpoint hardware.memory�>� [#�}Show Paths to Kernel Malloc Invocationskmalloc-topmemoryThe kmalloc-top perl program runs a small systemtap script to collect stack traces for each call to the kmalloc function and counts the time that each stack trace is observed. When kmalloc-top exits it prints out sorted list. The output can be filtered to print only the first N stack traces (-t), stack traces with a minimum counts (-m), or exclude certain stack traces (-e).memory & & �W �2 stpmA ringsC ystem A tapC F table I" 9$ sM3 kes Fs ! ingM llies J yJ cpA extF hat A0 L A% + e�A . mL0 reFi irdG+ sB' , oseD ime1B !- / f 1 sF 2 ! G oA" ( G olA talI9 uchedJ race)B n pointIq sH ing E kD s D fficC wo L" 2 ypeI udpA ntilA5 pI seH dL* rA7 ; sF ing E validA ueA, riousG erbumH � � �� c#�Generate Log of Major and Minor Page Faultspfaults.stpmemoryThe pfaults.stp script generates a simple log for each major and minor page fault that occurs on the system. Each line contains a timestamp (in microseconds) when the page fault servicing was completed, the pid of the process, the address of the page fault, the type of access (read or write), the type of fault (major or minor), and the elapsed time for page fault. This log can be examined to determine where the page faults are occurring.memory T @�� T �" �+'�YTrace slow file system synchronous reads and writes (non-debuginfo)fsslower-nd.stpio filesystemThis dynamically traces two common file system functions: do_sync_read() and do_sync_write(), and shows details of each call that is slower than a threshold. Many, but not all, file systems and workloads use these functions. Tracing their time provides one view of suffered file system latency.lwtools�F! K)%� SystemTap "Hello World" Programhelloworld.stp_best simpleA basic "Hello World" program implemented in SystemTap script. It prints out "hello world" message and then immediately exits.general� O)�SystemTap python 3 support tapsetpy3example.stpsimpleA python support tapset that displays backtraces and variable valuesgeneral�= /##�Hack the keyboardkeyhack.stpsimple guruThis script makes it appear that one's keyboard is broken, by changing keycodes at the kernel device driver level. Annoy your friends!general � �� � �z c5� System-Wide Count of Syscalls by Executablesyscalls_by_proc.stpsyscallThe script watches all syscall on the system. On exit the script prints a list showing the number of systemcalls executed by each executable ordered from greatest to least number of syscalls.process�Ty g%�System-Wide Syscall Statistics with FilteringsyscalltimessyscallCombination shell/systemtap script to measure system call counts and times. Can be filtered by process IDs, process names and users.process�[x C%+�9Tabulate System Call Errorserrsnoop.stpprocess syscallPrints a periodic tabular report about failing system calls, by process and by syscall failure. The first optional argument specifies the reporting interval (in seconds, default 5); the second optional argument gives a screen height (number of lines in the report, default 20).process } � } ��8 =-+�' Profile Kernel Functionsthread-times.stp_best profilingThe thread-times.stp script sets up time-based sampling. Every five seconds it prints out a sorted list with the top twenty threads occupying the CPUs, broken down as a percentage of user and kernel time.profiling�f�7 o)+� Show Reasons and Durations for Processes Sleepinglatencytap.stp_best profilingThe latencytap.stp script collects data on the intervals processes are deactivated (sleeping). The script categorizes the reasons for the sleeps by analyzing the backtraces and displays a sorted list of the top 20 causes from largest total sum time sleeping to smallest. The output is updated every 30 seconds. The script needs to be compiled with the '--all-modules' option to produce reasons for sleeps caused by modules. Optionally, this script can be used with the '-c' or '-x' options to focus on a specific PID.profiling & & �W �2 lockL G W ing L s L g G ngM machineI sJ deD nuallyA8 rkersE ximum IJ V stringlenC yC emoryH icrosecondsB" M ghtD! llisecondsFG nimum IC M oduleF8( ultipleF rocessorL nameB- ? 0 etB filterA work#A werIX fs D dD ot C t eF" secondsL? umber&A 0 ! = E ofaA $ - % & ! lderIi n F R( ceL1 eL& lyN peration I' sD tionC9 N D alL( rA #D" _ derC' gH utB < . 6 putJ packet A sA % geI! rameterIx singFO ss F0 esF( �]3 �> prints � oc � edure� ss�� es&�( $ 5 or�@ s�6 duce� thread�# race� queue�$ ing� readable� cent� ord� lated� moved� ports � source� limit� un� nable�# ing� D s � ched� times � uled� r� ript:� ecurity � nt � s�% hared� lib� ort�" wing� 8 ig� kill � nal� s� ze� s� leeping� ome� pace� ecifiable� c �: ) ed �< y� nd� s� 5 # n� re � Z is"�C < ; < 9 ose� reshold�% us�- ime-�" U / $ s�&& " o o`�% - $ ( ' @! - $ ) % 4 p� sys � tal�T racepoint� s � k� iggers�N wenty� o� unexpected�4 p � se� # * d� ! ] ful�$ r� space� s� ing� ual� value�Y X U rious � ia�% $ " rtual� m� warmup� ns�T s�[ y� hat� 0 en$� + 7 * J ( � �b�1 � �y Show performance ratios using perf.counter to access performance countersperf.stpprofilingOn exit the perf.stp script provides a sorted list showing cycles per insn, branches per insn, and cache refs per insnprofiling�~�0 G5 �s Profile Kernel/User Functionsfileline-profile.stpprofilingThe fileline-profile.stp script ends by printing out a sorted list of the top twenty kernel and/or user processes providing file:line information, if available, from the samples addresses gathered over the time period the script is run. Use any of --ldd, --all-modules, -d MODULE, -d /PATH/TO/EXEC to add more symbolic info. To include the symbol name in the output, specify guru mode (-g) and add symbolname="yes" to the stap command.profiling etdev/ work.% w( fs. ! d 0 on " t" umber, o# f*" n# e ", ' ly(% pen) ed) snoop) rations. r# 1 ut ! liers$0 put# packet- s- ssive% orts- rints ! ocess# # es( duces& ing+ gram! vides"+ & queue#) re(2 ad" s" ceive/ d- mote. ved0 ing0 port& s* quest## q# un0 wtime+ script! econds, nt' quence( rver0 " hort( w % * ing% ' �' �?] i �Periodically Print IO Activity by Process Nameiotop.stpioEvery five seconds print out the top ten executables generating I/O traffic during that interval sorted in descending order.io�\ i'�'Average Time Block IO Requests Spend in Queue ioblktime.stpioThe ioblktime.stp script tracks the amount of time that each block IO requests spend waiting for completion. The script computes the average waiting time for block IO per device and prints list every 10 seconds. In some cases there can be too many outstanding block IO operations and the script may exceed the default number of MAXMAPENTRIES allowed. In this case the allowed number can be increased with "-DMAXMAPENTRIES=10000" option on the stap command line.io � � � � [1'�qCount I/O Syscalls using Alias Suffixesalias_suffixes.stpio statisticsalias_suffixes.stp is a demonstration of how alias suffixes in the systemtap language might be used. The script tracks the wall clock time for each invocation of the system calls open, close, read, and write. When the script exists it prints out the minimum, average, and maximum times in microseconds for each system call, followed by a count of times that each syscall was invoked and a histogram showing the distributions of times.general�q Q)W�'Interactive Count Specified Eventseventcount.stpinteractive statistics thread processThe script periodically prints a count of specified events and their related tid's over the course of execution. Numerous configuration options exist to control filtering / reporting, some of which can be modified at runtime. See the script source for more information.general � |� �p[ 7# �Trace slow vfs opens.slowvfs.stpio simpleThis script prints a line for every kernel vfs_open operation that takes longer than a configurable number of microseconds. Highly contended or remote filesystems are likelier to hit this.io�Z i3�] Tally Backtraces for Inter-Processor Interruptscf.stpinterrupt backtraceThe Linux kernel function smp_call_function causes expensive inter-processor interrupts (IPIs). The scf.stp script tallies the processes and backtraces causing the interprocessor interrupts to identify the cause of the expensive IPI. On exit the script prints the tallies in descending frequency.interrupt � � �A�( �1W�w Provide a per-process accumulation of syscall latency on the systemsyscalllatency.stpprometheus process syscall tracepointThe syscalllatency.stp script accumulates syscall latency for each running process. This information can be useful to whether excessive time is being spent in particular syscalls on the system. The script makes the information available via procfs in Prometheus readable format. When a process exits its data will be eliminated from the prometheus output. To avoid exceeding the storage limitations of SystemTap older entries maybe overwritten by newer entries. This can lead to some active process syscall error counts disappearing and/or later reappearing with a lower value. Also note that the script does not properly name syscalls for 32-bit applications running on 64-bit machines.profiling Y � �O K/'�1cve-2015-7547 security band-aidcve-2015-7547.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�N K/'�1cve-2016-5195 security band-aidcve-2016-5195.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�$M g'�9Per Process Statistics on Big Kernel Lock Usebkl_stats.stplockingThe bkl_stats.stp script can indicate which processes have excessive waits for the Big Kernel Lock (BKL) and which processes are taking the BKL for long periods of time. The bkl_stats.stp script prints lists of all the processes that require the BKL. Every five seconds two tables are printed out. The first table lists the processes that waited for the BKL followed by the number of times that the process waited, the minimum time of the wait, the average and the maximum time waited. The second table lists has similar information for the time spent in holding the lock for each of the processes.locks $ $ �YM �6 mmwriteback � ode�E ule�* t & 1 s�( | $ / ments� nitor � # ing�; s � re�: 1 B - 8 st �- ust� name �b a r ^ ? eed�^ wer�E D B on� rmal�* t�` _ * \ e�[ Z W iceable�> umber�H Y object� ccur� fT� ; > ! = '- % /!! ! ' lder�@ ? = nB�, O O 3 /- N e� pen�% rations � 8 tion �e } ally� r)�9 S R O iginal� ut � put � 7 6 h 4 B ver � " written�C B @ wing�+ page �)" s� = rticular� th�, ( 3 dflush� er� iod � % � �� % � � via 3 , ( walk l clock s3 tch dog hat E en K re ich & le 1 ose ill 0 th " in 7 you r o o � � undesirable p s+ e� d rspace ing value s riable ous watch erbose ia) ew wall s B tch hen & ever ich( le3 y thefail ith F in9 rite , % your zing ! ! �\ �< in# " & . ( # formation 1 < put* tel ractive o verse ocation s(. + 5 , 0 t keep rnel / yboard ) level ibrary fe game mits ne sted oaded cal tions ! makes / ( tch ximum& ze echanism s mory icrosecond s $ ) lliseconds # ' seconds$ nimum# odern ule nitor ing s re; phes uch ltiple name s noseconds& ested 6 o t/ e' occur 0 s Q f( + $ 5 n # ly , pened rationP tional r ut ! ; put* para meter s J `�$� � J �U K/'�1cve-2014-7169 security band-aidcve-2014-7169.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�T K/'�1cve-2012-0056 security band-aidcve-2012-0056.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�S K/'�1cve-2016-0728 security band-aidcve-2016-0728.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�R K/'�1cve-2008-0600 security band-aidcve-2008-0600.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids�Q K/'�1cve-2013-2094 security band-aidcve-2013-2094.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids� P M1'�1cve-2018-14634 security band-aidcve-2018-14634.stpsecurity guruhistorical emergency security band-aid, for reference/education onlysecurity-band-aids " " �[N �: pf3 � 4 � olling� ssibly� revious� int �( ed�* ing� s� obe � s�W cess9� D #" "" " es"�/ fs�& % # file� ingX� gram�4 s�B metheus �( ' % perly�a ` ] vide� AJ s�K ing� readable�) ( & ppearing�U T Q cent� ursive� port� s� quested� stricted � F un� U ) ning �! _ ^ K [ s�# amples � ing � cripta� D @ 6 @ % =? ? # econds� nsitive�A ts � ting� hould�~ w � � [ � �UB g-)�Trace Functions Called in Network Socket Codesocket-trace.stpnetwork socketThe script instruments each of the functions in the Linux kernel's net/socket.c file. The script prints out trace data. The first element of a line is time delta in microseconds from the previous entry. This is followed by the command name and the PID. The "->" and "<-" indicates function entry and function exit, respectively. The last element of the line is the function name.network�"A ]15�System-Wide Network Packet Dropping Toolnetfilter_drop.stpnetwork packets guruThe script drops the specified number of packets of the specified protocol. Valid protocols are TCP, UDP, or ALL. If ALL is specified, all incoming packets are dropped. The number of packets to drop can be specified with a positive integer. A value of 0 indicates that packets should be dropped until the user manually exits.network �| 1!7�Trace system callsstrace.stp_best process syscallThe script loosely emulates strace, when applied to individual processes or hierarchies (via -c/-x), or the entire system (without -c/-x). A few output configuration parameters may be set with -G.process�h{ g% �;Monitor Changes in Processor Executing a Taskchng_cpu.stpschedulerThe chng_cpu.stp script takes an argument which is the executable name of the task it should monitor. Each time a task with that executable name is found running on a different processor, the script prints out the thread id (tid), the executable name, the processor now running the task, the thread state, and a backtrace showing the kernel functions that triggered the running of the task on the processor.process � � �3�* {9W�m Provide a per-process syscall error tally on the systemsyscallerrorsbypid.stpprometheus process syscall tracepointThe syscallerrorsbypid.stp script tallies syscall errors for each running process. This information can be useful to whether there are excessive errors for various processes on the system. The script makes the information available via procfs in Prometheus readable format. When a process exits its data will be eliminated from the prometheus output. To avoid exceeding the storage limitations of SystemTap older entries maybe overwritten by newer entries. This can lead to some active process syscall error counts disappearing and/or later reappearing with a lower value. Also note that the script does not properly name syscalls for 32-bit applications running on 64-bit machines.profiling � � ��+ �3C�9 Monitor capabilities and syscalls used by a process and it childrencontainer_check.stpprofiling container syscallThe container_check.stp script monitors the use of linux capablities and optionally forbidden syscalls by a process and its children. On exit the script prints out lists showing the capabilies used by each executable, which syscall used specific capabilites for each executable, a list of forbidden syscalls used, and details on any syscalls that failed during monitoring. This script is designed to help diagnose issues caused by restricted capabilies and syscalls when running an application in a container. If the script warns about skipped probes, the number of active kretprobes may need to be increased with "-DKRETACTIVE=100" option on the command lineprofiling � � �_- �B mays ; easurey icrosecondsq gratet st iontF � �`P �D the�� + $ $ ��~O � showing� ize � $ kipped�V lab� ub� o�� me� K J \ H rted�* pace� ecial� D fic�( y�C nds� t�X tack� p � M tistics� orage�< ; 9 pz� < race�! ummarize� pposedly�. ymbol�> ic�2 . 9 name�I scall'�I H & E errorsbypid � latency � s"�J b # + _ bypid � tem �0 X call� s� tap�? > < table�) kes � . ing� llies � y�F l han �( t0� K \ # [ 8 j X ' ? � ' �>= U) �uTrace NFSD requests with file names.nfsd-trace.stpnfs traceThis script traces all nfsd server operations by client_ip address, operation, and complete file name (if possible).network�T< u)�Periodically Summarize Socket Activity on the Systemsocktopnetwork socketThe socktop script periodically prints out a list of the processes with the highest socket activity. Command line options for the script allow filtering to focus on particular types of sockets. The "-h" option lists socktop script's filtering options.network�>; c77�ASystem-Wide Count of Network Packets by IPsnetfilter_summary.stp_best network trafficThe script watches all IPv4 network traffic on the system. On exit the script prints a list showing the number of packets sent along source IP address / destination IP address pair encountered, and the total number of bytes sent among the pair. The list is ordered from greatest to least number of packets seen among the source/destination pairs.network � �� � � A)c�wWhy did the function fail?whythefail.stpspeculation monitoring function trace _bestThe whythefail.stp script prints a statement-execution trace for a given function, but only for those runs of the function that ended up with a (configurable) post-return condition.general�, 3#+�mBad Filename Filterbadname.stpfilesystem guruThe badname.stp script shows how one could prevent the creation of files with undesirable names using guru mode.general�t Q/)�UTrace Socket Activity of a Processsocket-events.stpsocket syscallPrints the life cycle of all sockets associated with a process. This includes bytes and timing. The timing information that is tracked includes event completion relative to the start of said event and the end of the previous event. Currently tracks read, write, recv, send, connect and close.general odule� nitort ing� s } stv name$q! ) sy nosleep q ot t> % et7 w{1 umberu3 % of=q 8 & 6 ft- n"q ( perations } tionalx sv( r t< derr edz utq ' put| ver} parameters| enthesesq# ticularu erformance iodicx allyv idq ' $ rintr edt" ! # s q & oc r ess�q % ess or t, # st mod � topv ! ! �\ �< 1 2 3 aS _ -! 7 ctive" ity s) ggregate dWY lias l � ows{ nb d* !: $ erson noy si y ppear reo@ gument / s n ranges s%; sociated t # vailable erage c / gd backtraces d name sed is� e.0 ! ing1 st tween roken ut � y� : tes c all 8 graph! s n hanges ing lick ock se 1 " ode lor s mbinations pilation lete ion ncurrently3 dition figurable tion nect/ text � � �0L W�mTracing Contention on Big Kernel Lockbkl.stplockingThe bkl.stp script can help determine whether the Big Kernel Lock (BKL) is causing serialization on a multiprocessor system due to excessive contention of the BKL. The bkl.stp script takes two arguments. The first one is optional, and used to enable backtraces, and print them once a process has been holding the BKL for a user specified number of nseconds is reached. The second option is compulsory and is the number of processes waiting for the Big Kernel Lock (BKL). When the number of processes waiting for the BKL is reached or exceeded, the script will print a time stamp, the number of processes waiting for the BKL, the holder of the BKL, and the amount of time the BKL was held. If backtraces are enabled, a backtrace will be printed as well.locks k � k �' 5�{make keyboard noiseskey.stpsimpleFor fans of Leroy Anderson and Typewriters only, this script arranges to play a click or a zing for various keystrokes.general�Y A!/�7Print the Size of a C Typesizeof.stpstatistics memoryThis script prints the size of a type, based on dwarf debuginfo for any kernel or userspace module, or trial-compilation of a given header file name.general� Y9G�UInteractive Print the Size of a C Typesizeof_interactive.stpstatistics memory interactiveThis script prints the size of a type, based on dwarf debuginfo for any kernel or userspace module, or trial-compilation of a given header file name. Types and corresponding locations are provided to the script at runtime via keyboard input. The format of the input is identical to that of sizeof.stp (see script source for more information). Types and locations can be repeatedly given until the process is terminated. general � � �g�>0 345 a� based� every� five� unctions� it� kernel� list� out� pf2 � rints� ofile� ing � samples� ing� cript� econds� ts� orted� tp � ten� he� ime� op� up� with� y F y �Jg �I'�=Capture SSL/TLS master secrets from gnutls and openssl userscapture_ssl_master_secrets.stpio monitoringCapture SSL/TLS master secrets from gnutls and openssl users for easy traffic decryptionio�Vf �%?�cMonitor Read/Write of the Boot Sector Area of Block Devicesmbrwatch.stpio monitoring disk simple The mbrwatch.stp script reports any attempted reads/writes of the first few sectors of a raw block device.io�fe Y!+�GReport an disk out-of-space condition.enospc.stpdisk filesystemThis script monitors a filesystem implementations for early internal indications of ENOSPC, and reports these to the system logger and the systemtap console.io�ud Y+�uMonitoring Attribute Changes to a Fileinodewatch2.stpio The inodewatch2.stp script outputs the executable name, process id, and attributes each time the attributes are changed on the specified inode on the specified major/minor device.io " " �[ �: control uld -0, nt < rse reation urrent5 ly) ycle dataX ebuginfo monstration vice id fferent splayA ed] s tributions I river warf each nable+ d $ H1 ed try� vent % count s xecname� utable� ing on ist s ) pression fail ns ifthM" le name s ystem ter ing rst ve ollowed 9 r4 % � ) mat urthD) requently iends unction s ' top generalP ttimeofday* iven roup� ing� uru x x ��z0 887 20�# 30�2 a� ; ll�< nalyzing� d� re� backtraces� e�8 st� y�0 c�O an�J tegorizes� used�D s�$ ollects� mpiled�9 data� eactivated� isplays� urations� every�1 focus�T r�0 rom�% intervals� s�/ largest�& tencytap � ist� modules�= needs�6 of� n�O ption�> ally�G s�R r�P utput�. pid�X rocesses� duce�@ filing � reasons�0 script � ' econds�3 how� leeping� s�/ mallest�, orted� pecific�W tp � um�( the� is�H ime�) o �+ p�" tal�' updated�0 sed�L with�: x�Q � � � % � security`Q ek` s` howing` imple[ zed` low[ vfs[ mpZ ome\' rted] pend\ tap\L pbQ uccessfully ^ m ^ ystem ^ takes[ lliesZ yZ en] han[ t[ eFY re\) is [ > ime\ oZ & & o\, p] tals ^> 5 race[ ks\ ffic] vfs[ waiting\ ith\F rites ^< 3 ing^ ten^% 6 ��6 � m O#�QSummarize Disk Read/Write Trafficdisktop.stpdiskGet the status of reading/writing disk every 5 seconds, output top ten entries during that period.io�6l e#�sTrack Cumulative IO Activity by Process Nametraceio.stpioEvery second print out the top ten executables sorted in descending order based on cumulative I/O traffic observed.iopk -)�Switch log filesswitchfile.stpioEvery second print a log message and switch log files every 5 seconds.io � � � �{ e-�aShow Attribues in Table for ansi_set_color()ansi_colors2.stpformatThe script prints a table showing the available attributes (bold, underline, and inverse) with color combinations for the ans_set_color() function in the ansi.stp tapset.general� �+�OSee the amount of wall clock time a process spends in various statesstopwatches.stptimeThe stopwatch.stp script illustrates how to use multiple stopwatches record how much wallclock time a process spends in kernel- and user-space. On exit the script prints out the time in seconds, milliseconds, microseconds, and nanoseconds. Note that this output of this script is not directly comparable to the time command because time records the time that the process is actually active in kernel- and user-space.general y 4� y �j I+/�#IO Statistics for SCSI Devicesiostat-scsi.stpio profiling scsiThe iostat-scsi.stp script provides a breakdown of the number of blks read and written on the machine's various SCSI devices. The script takes one argument which is the number of seconds between reports.io�"i y1%�Tally the Number of NFS Functions Used by Each Processnfs_func_users.stpio profilingThe nfs_func_users.stp script counts the uses of NFS functions in the kernel on a per process bases. The output is sorted from the process with the greatest number of NFS functions called to the least. The output contains the executable name, the process number, and the total number of NFS functions called by the process.io�Ih '')�;disable fsynceatmydata.stpio guru simpleSuppresses fsync() syscalls from processes identified by stap -c/-x by turning them into presumed-faster fsync() on some dummy or other file descriptorio � �� �+o�yTrack IRQ's and Other Processes Stealing Cycles from a Taskcycle_thief.stp_best process scheduler time tracepoint interruptThe cycle_thief.stp script instruments the scheduler and IRQ handler to determine which processes and interrupts are competing with the specified task for the cpu cycles. This script uses the '-c' or '-x' options to focus on a specific task. The script output the number of times the task migrates between processors, histograms showing the length of time on and off processor, lists of processes running while the task is off the processor, and the interrupts that occurred while the task was running.process�i� O5�Ipthread mutex contention analysismutex-contention.stplockingTracks pthread-mutex initialization/use and underlying futex operations, to identify (with backtraces/symbol-names) the mutexes suffering most contention. Invoke with "-d SHLIB --ldd", perhaps with -DMAXMAPENTRIES=NNNN for some large NNNN, if the arrays overflow due to heavy activity.process