1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
|
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>28.5. Dynamic Tracing</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="progress-reporting.html" title="28.4. Progress Reporting" /><link rel="next" href="diskusage.html" title="Chapter 29. Monitoring Disk Usage" /></head><body id="docContent" class="container-fluid col-10"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">28.5. Dynamic Tracing</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="progress-reporting.html" title="28.4. Progress Reporting">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="monitoring.html" title="Chapter 28. Monitoring Database Activity">Up</a></td><th width="60%" align="center">Chapter 28. Monitoring Database Activity</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 16.2 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="diskusage.html" title="Chapter 29. Monitoring Disk Usage">Next</a></td></tr></table><hr /></div><div class="sect1" id="DYNAMIC-TRACE"><div class="titlepage"><div><div><h2 class="title" style="clear: both">28.5. Dynamic Tracing <a href="#DYNAMIC-TRACE" class="id_link">#</a></h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="dynamic-trace.html#COMPILING-FOR-TRACE">28.5.1. Compiling for Dynamic Tracing</a></span></dt><dt><span class="sect2"><a href="dynamic-trace.html#TRACE-POINTS">28.5.2. Built-in Probes</a></span></dt><dt><span class="sect2"><a href="dynamic-trace.html#USING-TRACE-POINTS">28.5.3. Using Probes</a></span></dt><dt><span class="sect2"><a href="dynamic-trace.html#DEFINING-TRACE-POINTS">28.5.4. Defining New Probes</a></span></dt></dl></div><a id="id-1.6.15.10.2" class="indexterm"></a><p>
<span class="productname">PostgreSQL</span> provides facilities to support
dynamic tracing of the database server. This allows an external
utility to be called at specific points in the code and thereby trace
execution.
</p><p>
A number of probes or trace points are already inserted into the source
code. These probes are intended to be used by database developers and
administrators. By default the probes are not compiled into
<span class="productname">PostgreSQL</span>; the user needs to explicitly tell
the configure script to make the probes available.
</p><p>
Currently, the
<a class="ulink" href="https://en.wikipedia.org/wiki/DTrace" target="_top">DTrace</a>
utility is supported, which, at the time of this writing, is available
on Solaris, macOS, FreeBSD, NetBSD, and Oracle Linux. The
<a class="ulink" href="https://sourceware.org/systemtap/" target="_top">SystemTap</a> project
for Linux provides a DTrace equivalent and can also be used. Supporting other dynamic
tracing utilities is theoretically possible by changing the definitions for
the macros in <code class="filename">src/include/utils/probes.h</code>.
</p><div class="sect2" id="COMPILING-FOR-TRACE"><div class="titlepage"><div><div><h3 class="title">28.5.1. Compiling for Dynamic Tracing <a href="#COMPILING-FOR-TRACE" class="id_link">#</a></h3></div></div></div><p>
By default, probes are not available, so you will need to
explicitly tell the configure script to make the probes available
in <span class="productname">PostgreSQL</span>. To include DTrace support
specify <code class="option">--enable-dtrace</code> to configure. See <a class="xref" href="install-make.html#CONFIGURE-OPTIONS-DEVEL" title="17.3.3.6. Developer Options">Section 17.3.3.6</a> for further information.
</p></div><div class="sect2" id="TRACE-POINTS"><div class="titlepage"><div><div><h3 class="title">28.5.2. Built-in Probes <a href="#TRACE-POINTS" class="id_link">#</a></h3></div></div></div><p>
A number of standard probes are provided in the source code,
as shown in <a class="xref" href="dynamic-trace.html#DTRACE-PROBE-POINT-TABLE" title="Table 28.48. Built-in DTrace Probes">Table 28.48</a>;
<a class="xref" href="dynamic-trace.html#TYPEDEFS-TABLE" title="Table 28.49. Defined Types Used in Probe Parameters">Table 28.49</a>
shows the types used in the probes. More probes can certainly be
added to enhance <span class="productname">PostgreSQL</span>'s observability.
</p><div class="table" id="DTRACE-PROBE-POINT-TABLE"><p class="title"><strong>Table 28.48. Built-in DTrace Probes</strong></p><div class="table-contents"><table class="table" summary="Built-in DTrace Probes" border="1"><colgroup><col class="col1" /><col class="col2" /><col class="col3" /></colgroup><thead><tr><th>Name</th><th>Parameters</th><th>Description</th></tr></thead><tbody><tr><td><code class="literal">transaction-start</code></td><td><code class="literal">(LocalTransactionId)</code></td><td>Probe that fires at the start of a new transaction.
arg0 is the transaction ID.</td></tr><tr><td><code class="literal">transaction-commit</code></td><td><code class="literal">(LocalTransactionId)</code></td><td>Probe that fires when a transaction completes successfully.
arg0 is the transaction ID.</td></tr><tr><td><code class="literal">transaction-abort</code></td><td><code class="literal">(LocalTransactionId)</code></td><td>Probe that fires when a transaction completes unsuccessfully.
arg0 is the transaction ID.</td></tr><tr><td><code class="literal">query-start</code></td><td><code class="literal">(const char *)</code></td><td>Probe that fires when the processing of a query is started.
arg0 is the query string.</td></tr><tr><td><code class="literal">query-done</code></td><td><code class="literal">(const char *)</code></td><td>Probe that fires when the processing of a query is complete.
arg0 is the query string.</td></tr><tr><td><code class="literal">query-parse-start</code></td><td><code class="literal">(const char *)</code></td><td>Probe that fires when the parsing of a query is started.
arg0 is the query string.</td></tr><tr><td><code class="literal">query-parse-done</code></td><td><code class="literal">(const char *)</code></td><td>Probe that fires when the parsing of a query is complete.
arg0 is the query string.</td></tr><tr><td><code class="literal">query-rewrite-start</code></td><td><code class="literal">(const char *)</code></td><td>Probe that fires when the rewriting of a query is started.
arg0 is the query string.</td></tr><tr><td><code class="literal">query-rewrite-done</code></td><td><code class="literal">(const char *)</code></td><td>Probe that fires when the rewriting of a query is complete.
arg0 is the query string.</td></tr><tr><td><code class="literal">query-plan-start</code></td><td><code class="literal">()</code></td><td>Probe that fires when the planning of a query is started.</td></tr><tr><td><code class="literal">query-plan-done</code></td><td><code class="literal">()</code></td><td>Probe that fires when the planning of a query is complete.</td></tr><tr><td><code class="literal">query-execute-start</code></td><td><code class="literal">()</code></td><td>Probe that fires when the execution of a query is started.</td></tr><tr><td><code class="literal">query-execute-done</code></td><td><code class="literal">()</code></td><td>Probe that fires when the execution of a query is complete.</td></tr><tr><td><code class="literal">statement-status</code></td><td><code class="literal">(const char *)</code></td><td>Probe that fires anytime the server process updates its
<code class="structname">pg_stat_activity</code>.<code class="structfield">status</code>.
arg0 is the new status string.</td></tr><tr><td><code class="literal">checkpoint-start</code></td><td><code class="literal">(int)</code></td><td>Probe that fires when a checkpoint is started.
arg0 holds the bitwise flags used to distinguish different checkpoint
types, such as shutdown, immediate or force.</td></tr><tr><td><code class="literal">checkpoint-done</code></td><td><code class="literal">(int, int, int, int, int)</code></td><td>Probe that fires when a checkpoint is complete.
(The probes listed next fire in sequence during checkpoint processing.)
arg0 is the number of buffers written. arg1 is the total number of
buffers. arg2, arg3 and arg4 contain the number of WAL files added,
removed and recycled respectively.</td></tr><tr><td><code class="literal">clog-checkpoint-start</code></td><td><code class="literal">(bool)</code></td><td>Probe that fires when the CLOG portion of a checkpoint is started.
arg0 is true for normal checkpoint, false for shutdown
checkpoint.</td></tr><tr><td><code class="literal">clog-checkpoint-done</code></td><td><code class="literal">(bool)</code></td><td>Probe that fires when the CLOG portion of a checkpoint is
complete. arg0 has the same meaning as for <code class="literal">clog-checkpoint-start</code>.</td></tr><tr><td><code class="literal">subtrans-checkpoint-start</code></td><td><code class="literal">(bool)</code></td><td>Probe that fires when the SUBTRANS portion of a checkpoint is
started.
arg0 is true for normal checkpoint, false for shutdown
checkpoint.</td></tr><tr><td><code class="literal">subtrans-checkpoint-done</code></td><td><code class="literal">(bool)</code></td><td>Probe that fires when the SUBTRANS portion of a checkpoint is
complete. arg0 has the same meaning as for
<code class="literal">subtrans-checkpoint-start</code>.</td></tr><tr><td><code class="literal">multixact-checkpoint-start</code></td><td><code class="literal">(bool)</code></td><td>Probe that fires when the MultiXact portion of a checkpoint is
started.
arg0 is true for normal checkpoint, false for shutdown
checkpoint.</td></tr><tr><td><code class="literal">multixact-checkpoint-done</code></td><td><code class="literal">(bool)</code></td><td>Probe that fires when the MultiXact portion of a checkpoint is
complete. arg0 has the same meaning as for
<code class="literal">multixact-checkpoint-start</code>.</td></tr><tr><td><code class="literal">buffer-checkpoint-start</code></td><td><code class="literal">(int)</code></td><td>Probe that fires when the buffer-writing portion of a checkpoint
is started.
arg0 holds the bitwise flags used to distinguish different checkpoint
types, such as shutdown, immediate or force.</td></tr><tr><td><code class="literal">buffer-sync-start</code></td><td><code class="literal">(int, int)</code></td><td>Probe that fires when we begin to write dirty buffers during
checkpoint (after identifying which buffers must be written).
arg0 is the total number of buffers.
arg1 is the number that are currently dirty and need to be written.</td></tr><tr><td><code class="literal">buffer-sync-written</code></td><td><code class="literal">(int)</code></td><td>Probe that fires after each buffer is written during checkpoint.
arg0 is the ID number of the buffer.</td></tr><tr><td><code class="literal">buffer-sync-done</code></td><td><code class="literal">(int, int, int)</code></td><td>Probe that fires when all dirty buffers have been written.
arg0 is the total number of buffers.
arg1 is the number of buffers actually written by the checkpoint process.
arg2 is the number that were expected to be written (arg1 of
<code class="literal">buffer-sync-start</code>); any difference reflects other processes flushing
buffers during the checkpoint.</td></tr><tr><td><code class="literal">buffer-checkpoint-sync-start</code></td><td><code class="literal">()</code></td><td>Probe that fires after dirty buffers have been written to the
kernel, and before starting to issue fsync requests.</td></tr><tr><td><code class="literal">buffer-checkpoint-done</code></td><td><code class="literal">()</code></td><td>Probe that fires when syncing of buffers to disk is
complete.</td></tr><tr><td><code class="literal">twophase-checkpoint-start</code></td><td><code class="literal">()</code></td><td>Probe that fires when the two-phase portion of a checkpoint is
started.</td></tr><tr><td><code class="literal">twophase-checkpoint-done</code></td><td><code class="literal">()</code></td><td>Probe that fires when the two-phase portion of a checkpoint is
complete.</td></tr><tr><td><code class="literal">buffer-extend-start</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int, unsigned int)</code></td><td>Probe that fires when a relation extension starts.
arg0 contains the fork to be extended. arg1, arg2, and arg3 contain the
tablespace, database, and relation OIDs identifying the relation. arg4
is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared
buffer. arg5 is the number of blocks the caller would like to extend
by.</td></tr><tr><td><code class="literal">buffer-extend-done</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int, unsigned int, BlockNumber)</code></td><td>Probe that fires when a relation extension is complete.
arg0 contains the fork to be extended. arg1, arg2, and arg3 contain the
tablespace, database, and relation OIDs identifying the relation. arg4
is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared
buffer. arg5 is the number of blocks the relation was extended by, this
can be less than the number in the
<code class="literal">buffer-extend-start</code> due to resource
constraints. arg6 contains the BlockNumber of the first new
block.</td></tr><tr><td><code class="literal">buffer-read-start</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int)</code></td><td>Probe that fires when a buffer read is started.
arg0 and arg1 contain the fork and block numbers of the page.
arg2, arg3, and arg4 contain the tablespace, database, and relation OIDs
identifying the relation.
arg5 is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared buffer.
</td></tr><tr><td><code class="literal">buffer-read-done</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int, bool)</code></td><td>Probe that fires when a buffer read is complete.
arg0 and arg1 contain the fork and block numbers of the page.
arg2, arg3, and arg4 contain the tablespace, database, and relation OIDs
identifying the relation.
arg5 is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared buffer.
arg6 is true if the buffer was found in the pool, false if not.</td></tr><tr><td><code class="literal">buffer-flush-start</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid)</code></td><td>Probe that fires before issuing any write request for a shared
buffer.
arg0 and arg1 contain the fork and block numbers of the page.
arg2, arg3, and arg4 contain the tablespace, database, and relation OIDs
identifying the relation.</td></tr><tr><td><code class="literal">buffer-flush-done</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid)</code></td><td>Probe that fires when a write request is complete. (Note
that this just reflects the time to pass the data to the kernel;
it's typically not actually been written to disk yet.)
The arguments are the same as for <code class="literal">buffer-flush-start</code>.</td></tr><tr><td><code class="literal">wal-buffer-write-dirty-start</code></td><td><code class="literal">()</code></td><td>Probe that fires when a server process begins to write a
dirty WAL buffer because no more WAL buffer space is available.
(If this happens often, it implies that
<a class="xref" href="runtime-config-wal.html#GUC-WAL-BUFFERS">wal_buffers</a> is too small.)</td></tr><tr><td><code class="literal">wal-buffer-write-dirty-done</code></td><td><code class="literal">()</code></td><td>Probe that fires when a dirty WAL buffer write is complete.</td></tr><tr><td><code class="literal">wal-insert</code></td><td><code class="literal">(unsigned char, unsigned char)</code></td><td>Probe that fires when a WAL record is inserted.
arg0 is the resource manager (rmid) for the record.
arg1 contains the info flags.</td></tr><tr><td><code class="literal">wal-switch</code></td><td><code class="literal">()</code></td><td>Probe that fires when a WAL segment switch is requested.</td></tr><tr><td><code class="literal">smgr-md-read-start</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int)</code></td><td>Probe that fires when beginning to read a block from a relation.
arg0 and arg1 contain the fork and block numbers of the page.
arg2, arg3, and arg4 contain the tablespace, database, and relation OIDs
identifying the relation.
arg5 is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared buffer.</td></tr><tr><td><code class="literal">smgr-md-read-done</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int, int, int)</code></td><td>Probe that fires when a block read is complete.
arg0 and arg1 contain the fork and block numbers of the page.
arg2, arg3, and arg4 contain the tablespace, database, and relation OIDs
identifying the relation.
arg5 is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared buffer.
arg6 is the number of bytes actually read, while arg7 is the number
requested (if these are different it indicates trouble).</td></tr><tr><td><code class="literal">smgr-md-write-start</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int)</code></td><td>Probe that fires when beginning to write a block to a relation.
arg0 and arg1 contain the fork and block numbers of the page.
arg2, arg3, and arg4 contain the tablespace, database, and relation OIDs
identifying the relation.
arg5 is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared buffer.</td></tr><tr><td><code class="literal">smgr-md-write-done</code></td><td><code class="literal">(ForkNumber, BlockNumber, Oid, Oid, Oid, int, int, int)</code></td><td>Probe that fires when a block write is complete.
arg0 and arg1 contain the fork and block numbers of the page.
arg2, arg3, and arg4 contain the tablespace, database, and relation OIDs
identifying the relation.
arg5 is the ID of the backend which created the temporary relation for a
local buffer, or <code class="symbol">InvalidBackendId</code> (-1) for a shared buffer.
arg6 is the number of bytes actually written, while arg7 is the number
requested (if these are different it indicates trouble).</td></tr><tr><td><code class="literal">sort-start</code></td><td><code class="literal">(int, bool, int, int, bool, int)</code></td><td>Probe that fires when a sort operation is started.
arg0 indicates heap, index or datum sort.
arg1 is true for unique-value enforcement.
arg2 is the number of key columns.
arg3 is the number of kilobytes of work memory allowed.
arg4 is true if random access to the sort result is required.
arg5 indicates serial when <code class="literal">0</code>, parallel worker when
<code class="literal">1</code>, or parallel leader when <code class="literal">2</code>.</td></tr><tr><td><code class="literal">sort-done</code></td><td><code class="literal">(bool, long)</code></td><td>Probe that fires when a sort is complete.
arg0 is true for external sort, false for internal sort.
arg1 is the number of disk blocks used for an external sort,
or kilobytes of memory used for an internal sort.</td></tr><tr><td><code class="literal">lwlock-acquire</code></td><td><code class="literal">(char *, LWLockMode)</code></td><td>Probe that fires when an LWLock has been acquired.
arg0 is the LWLock's tranche.
arg1 is the requested lock mode, either exclusive or shared.</td></tr><tr><td><code class="literal">lwlock-release</code></td><td><code class="literal">(char *)</code></td><td>Probe that fires when an LWLock has been released (but note
that any released waiters have not yet been awakened).
arg0 is the LWLock's tranche.</td></tr><tr><td><code class="literal">lwlock-wait-start</code></td><td><code class="literal">(char *, LWLockMode)</code></td><td>Probe that fires when an LWLock was not immediately available and
a server process has begun to wait for the lock to become available.
arg0 is the LWLock's tranche.
arg1 is the requested lock mode, either exclusive or shared.</td></tr><tr><td><code class="literal">lwlock-wait-done</code></td><td><code class="literal">(char *, LWLockMode)</code></td><td>Probe that fires when a server process has been released from its
wait for an LWLock (it does not actually have the lock yet).
arg0 is the LWLock's tranche.
arg1 is the requested lock mode, either exclusive or shared.</td></tr><tr><td><code class="literal">lwlock-condacquire</code></td><td><code class="literal">(char *, LWLockMode)</code></td><td>Probe that fires when an LWLock was successfully acquired when the
caller specified no waiting.
arg0 is the LWLock's tranche.
arg1 is the requested lock mode, either exclusive or shared.</td></tr><tr><td><code class="literal">lwlock-condacquire-fail</code></td><td><code class="literal">(char *, LWLockMode)</code></td><td>Probe that fires when an LWLock was not successfully acquired when
the caller specified no waiting.
arg0 is the LWLock's tranche.
arg1 is the requested lock mode, either exclusive or shared.</td></tr><tr><td><code class="literal">lock-wait-start</code></td><td><code class="literal">(unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, LOCKMODE)</code></td><td>Probe that fires when a request for a heavyweight lock (lmgr lock)
has begun to wait because the lock is not available.
arg0 through arg3 are the tag fields identifying the object being
locked. arg4 indicates the type of object being locked.
arg5 indicates the lock type being requested.</td></tr><tr><td><code class="literal">lock-wait-done</code></td><td><code class="literal">(unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, LOCKMODE)</code></td><td>Probe that fires when a request for a heavyweight lock (lmgr lock)
has finished waiting (i.e., has acquired the lock).
The arguments are the same as for <code class="literal">lock-wait-start</code>.</td></tr><tr><td><code class="literal">deadlock-found</code></td><td><code class="literal">()</code></td><td>Probe that fires when a deadlock is found by the deadlock
detector.</td></tr></tbody></table></div></div><br class="table-break" /><div class="table" id="TYPEDEFS-TABLE"><p class="title"><strong>Table 28.49. Defined Types Used in Probe Parameters</strong></p><div class="table-contents"><table class="table" summary="Defined Types Used in Probe Parameters" border="1"><colgroup><col /><col /></colgroup><thead><tr><th>Type</th><th>Definition</th></tr></thead><tbody><tr><td><code class="type">LocalTransactionId</code></td><td><code class="type">unsigned int</code></td></tr><tr><td><code class="type">LWLockMode</code></td><td><code class="type">int</code></td></tr><tr><td><code class="type">LOCKMODE</code></td><td><code class="type">int</code></td></tr><tr><td><code class="type">BlockNumber</code></td><td><code class="type">unsigned int</code></td></tr><tr><td><code class="type">Oid</code></td><td><code class="type">unsigned int</code></td></tr><tr><td><code class="type">ForkNumber</code></td><td><code class="type">int</code></td></tr><tr><td><code class="type">bool</code></td><td><code class="type">unsigned char</code></td></tr></tbody></table></div></div><br class="table-break" /></div><div class="sect2" id="USING-TRACE-POINTS"><div class="titlepage"><div><div><h3 class="title">28.5.3. Using Probes <a href="#USING-TRACE-POINTS" class="id_link">#</a></h3></div></div></div><p>
The example below shows a DTrace script for analyzing transaction
counts in the system, as an alternative to snapshotting
<code class="structname">pg_stat_database</code> before and after a performance test:
</p><pre class="programlisting">
#!/usr/sbin/dtrace -qs
postgresql$1:::transaction-start
{
@start["Start"] = count();
self->ts = timestamp;
}
postgresql$1:::transaction-abort
{
@abort["Abort"] = count();
}
postgresql$1:::transaction-commit
/self->ts/
{
@commit["Commit"] = count();
@time["Total time (ns)"] = sum(timestamp - self->ts);
self->ts=0;
}
</pre><p>
When executed, the example D script gives output such as:
</p><pre class="screen">
# ./txn_count.d `pgrep -n postgres` or ./txn_count.d <PID>
^C
Start 71
Commit 70
Total time (ns) 2312105013
</pre><p>
</p><div class="note"><h3 class="title">Note</h3><p>
SystemTap uses a different notation for trace scripts than DTrace does,
even though the underlying trace points are compatible. One point worth
noting is that at this writing, SystemTap scripts must reference probe
names using double underscores in place of hyphens. This is expected to
be fixed in future SystemTap releases.
</p></div><p>
You should remember that DTrace scripts need to be carefully written and
debugged, otherwise the trace information collected might
be meaningless. In most cases where problems are found it is the
instrumentation that is at fault, not the underlying system. When
discussing information found using dynamic tracing, be sure to enclose
the script used to allow that too to be checked and discussed.
</p></div><div class="sect2" id="DEFINING-TRACE-POINTS"><div class="titlepage"><div><div><h3 class="title">28.5.4. Defining New Probes <a href="#DEFINING-TRACE-POINTS" class="id_link">#</a></h3></div></div></div><p>
New probes can be defined within the code wherever the developer
desires, though this will require a recompilation. Below are the steps
for inserting new probes:
</p><div class="procedure"><ol class="procedure" type="1"><li class="step"><p>
Decide on probe names and data to be made available through the probes
</p></li><li class="step"><p>
Add the probe definitions to <code class="filename">src/backend/utils/probes.d</code>
</p></li><li class="step"><p>
Include <code class="filename">pg_trace.h</code> if it is not already present in the
module(s) containing the probe points, and insert
<code class="literal">TRACE_POSTGRESQL</code> probe macros at the desired locations
in the source code
</p></li><li class="step"><p>
Recompile and verify that the new probes are available
</p></li></ol></div><p><strong>Example: </strong>
Here is an example of how you would add a probe to trace all new
transactions by transaction ID.
</p><div class="procedure"><ol class="procedure" type="1"><li class="step"><p>
Decide that the probe will be named <code class="literal">transaction-start</code> and
requires a parameter of type <code class="type">LocalTransactionId</code>
</p></li><li class="step"><p>
Add the probe definition to <code class="filename">src/backend/utils/probes.d</code>:
</p><pre class="programlisting">
probe transaction__start(LocalTransactionId);
</pre><p>
Note the use of the double underline in the probe name. In a DTrace
script using the probe, the double underline needs to be replaced with a
hyphen, so <code class="literal">transaction-start</code> is the name to document for
users.
</p></li><li class="step"><p>
At compile time, <code class="literal">transaction__start</code> is converted to a macro
called <code class="literal">TRACE_POSTGRESQL_TRANSACTION_START</code> (notice the
underscores are single here), which is available by including
<code class="filename">pg_trace.h</code>. Add the macro call to the appropriate location
in the source code. In this case, it looks like the following:
</p><pre class="programlisting">
TRACE_POSTGRESQL_TRANSACTION_START(vxid.localTransactionId);
</pre><p>
</p></li><li class="step"><p>
After recompiling and running the new binary, check that your newly added
probe is available by executing the following DTrace command. You
should see similar output:
</p><pre class="screen">
# dtrace -ln transaction-start
ID PROVIDER MODULE FUNCTION NAME
18705 postgresql49878 postgres StartTransactionCommand transaction-start
18755 postgresql49877 postgres StartTransactionCommand transaction-start
18805 postgresql49876 postgres StartTransactionCommand transaction-start
18855 postgresql49875 postgres StartTransactionCommand transaction-start
18986 postgresql49873 postgres StartTransactionCommand transaction-start
</pre><p>
</p></li></ol></div><p>
There are a few things to be careful about when adding trace macros
to the C code:
</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
You should take care that the data types specified for a probe's
parameters match the data types of the variables used in the macro.
Otherwise, you will get compilation errors.
</p></li><li class="listitem"><p>
On most platforms, if <span class="productname">PostgreSQL</span> is
built with <code class="option">--enable-dtrace</code>, the arguments to a trace
macro will be evaluated whenever control passes through the
macro, <span class="emphasis"><em>even if no tracing is being done</em></span>. This is
usually not worth worrying about if you are just reporting the
values of a few local variables. But beware of putting expensive
function calls into the arguments. If you need to do that,
consider protecting the macro with a check to see if the trace
is actually enabled:
</p><pre class="programlisting">
if (TRACE_POSTGRESQL_TRANSACTION_START_ENABLED())
TRACE_POSTGRESQL_TRANSACTION_START(some_function(...));
</pre><p>
Each trace macro has a corresponding <code class="literal">ENABLED</code> macro.
</p></li></ul></div><p>
</p></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="progress-reporting.html" title="28.4. Progress Reporting">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="monitoring.html" title="Chapter 28. Monitoring Database Activity">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="diskusage.html" title="Chapter 29. Monitoring Disk Usage">Next</a></td></tr><tr><td width="40%" align="left" valign="top">28.4. Progress Reporting </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 16.2 Documentation">Home</a></td><td width="40%" align="right" valign="top"> Chapter 29. Monitoring Disk Usage</td></tr></table></div></body></html>
|
