summaryrefslogtreecommitdiffstats
path: root/doc/src/sgml/html/runtime-config-developer.html
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2024-05-04 12:19:15 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2024-05-04 12:19:15 +0000
commit6eb9c5a5657d1fe77b55cc261450f3538d35a94d (patch)
tree657d8194422a5daccecfd42d654b8a245ef7b4c8 /doc/src/sgml/html/runtime-config-developer.html
parentInitial commit. (diff)
downloadpostgresql-13-upstream.tar.xz
postgresql-13-upstream.zip
Adding upstream version 13.4.upstream/13.4upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'doc/src/sgml/html/runtime-config-developer.html')
-rw-r--r--doc/src/sgml/html/runtime-config-developer.html291
1 files changed, 291 insertions, 0 deletions
diff --git a/doc/src/sgml/html/runtime-config-developer.html b/doc/src/sgml/html/runtime-config-developer.html
new file mode 100644
index 0000000..3e8d00e
--- /dev/null
+++ b/doc/src/sgml/html/runtime-config-developer.html
@@ -0,0 +1,291 @@
+<?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>19.17. Developer Options</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 V1.79.1" /><link rel="prev" href="runtime-config-custom.html" title="19.16. Customized Options" /><link rel="next" href="runtime-config-short.html" title="19.18. Short Options" /></head><body id="docContent" class="container-fluid col-10"><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">19.17. Developer Options</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="runtime-config-custom.html" title="19.16. Customized Options">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="runtime-config.html" title="Chapter 19. Server Configuration">Up</a></td><th width="60%" align="center">Chapter 19. Server Configuration</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 13.4 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="runtime-config-short.html" title="19.18. Short Options">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="RUNTIME-CONFIG-DEVELOPER"><div class="titlepage"><div><div><h2 class="title" style="clear: both">19.17. Developer Options</h2></div></div></div><p>
+ The following parameters are intended for work on the
+ <span class="productname">PostgreSQL</span> source code, and in some cases
+ to assist with recovery of severely damaged databases. There
+ should be no reason to use them on a production database.
+ As such, they have been excluded from the sample
+ <code class="filename">postgresql.conf</code> file. Note that many of these
+ parameters require special source compilation flags to work at all.
+ </p><div class="variablelist"><dl class="variablelist"><dt id="GUC-ALLOW-SYSTEM-TABLE-MODS"><span class="term"><code class="varname">allow_system_table_mods</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.1.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Allows modification of the structure of system tables as well as
+ certain other risky actions on system tables. This is otherwise not
+ allowed even for superusers. Ill-advised use of this setting can
+ cause irretrievable data loss or seriously corrupt the database
+ system. Only superusers can change this setting.
+ </p></dd><dt id="GUC-BACKTRACE-FUNCTIONS"><span class="term"><code class="varname">backtrace_functions</code> (<code class="type">string</code>)
+ <a id="id-1.6.6.20.3.2.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ This parameter contains a comma-separated list of C function names.
+ If an error is raised and the name of the internal C function where
+ the error happens matches a value in the list, then a backtrace is
+ written to the server log together with the error message. This can
+ be used to debug specific areas of the source code.
+ </p><p>
+ Backtrace support is not available on all platforms, and the quality
+ of the backtraces depends on compilation options.
+ </p><p>
+ This parameter can only be set by superusers.
+ </p></dd><dt id="GUC-IGNORE-SYSTEM-INDEXES"><span class="term"><code class="varname">ignore_system_indexes</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.3.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Ignore system indexes when reading system tables (but still
+ update the indexes when modifying the tables). This is useful
+ when recovering from damaged system indexes.
+ This parameter cannot be changed after session start.
+ </p></dd><dt id="GUC-POST-AUTH-DELAY"><span class="term"><code class="varname">post_auth_delay</code> (<code class="type">integer</code>)
+ <a id="id-1.6.6.20.3.4.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ The amount of time to delay when a new
+ server process is started, after it conducts the
+ authentication procedure. This is intended to give developers an
+ opportunity to attach to the server process with a debugger.
+ If this value is specified without units, it is taken as seconds.
+ A value of zero (the default) disables the delay.
+ This parameter cannot be changed after session start.
+ </p></dd><dt id="GUC-PRE-AUTH-DELAY"><span class="term"><code class="varname">pre_auth_delay</code> (<code class="type">integer</code>)
+ <a id="id-1.6.6.20.3.5.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ The amount of time to delay just after a
+ new server process is forked, before it conducts the
+ authentication procedure. This is intended to give developers an
+ opportunity to attach to the server process with a debugger to
+ trace down misbehavior in authentication.
+ If this value is specified without units, it is taken as seconds.
+ A value of zero (the default) disables the delay.
+ This parameter can only be set in the <code class="filename">postgresql.conf</code>
+ file or on the server command line.
+ </p></dd><dt id="GUC-TRACE-NOTIFY"><span class="term"><code class="varname">trace_notify</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.6.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Generates a great amount of debugging output for the
+ <code class="command">LISTEN</code> and <code class="command">NOTIFY</code>
+ commands. <a class="xref" href="runtime-config-client.html#GUC-CLIENT-MIN-MESSAGES">client_min_messages</a> or
+ <a class="xref" href="runtime-config-logging.html#GUC-LOG-MIN-MESSAGES">log_min_messages</a> must be
+ <code class="literal">DEBUG1</code> or lower to send this output to the
+ client or server logs, respectively.
+ </p></dd><dt id="GUC-TRACE-RECOVERY-MESSAGES"><span class="term"><code class="varname">trace_recovery_messages</code> (<code class="type">enum</code>)
+ <a id="id-1.6.6.20.3.7.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Enables logging of recovery-related debugging output that otherwise
+ would not be logged. This parameter allows the user to override the
+ normal setting of <a class="xref" href="runtime-config-logging.html#GUC-LOG-MIN-MESSAGES">log_min_messages</a>, but only for
+ specific messages. This is intended for use in debugging Hot Standby.
+ Valid values are <code class="literal">DEBUG5</code>, <code class="literal">DEBUG4</code>,
+ <code class="literal">DEBUG3</code>, <code class="literal">DEBUG2</code>, <code class="literal">DEBUG1</code>, and
+ <code class="literal">LOG</code>. The default, <code class="literal">LOG</code>, does not affect
+ logging decisions at all. The other values cause recovery-related
+ debug messages of that priority or higher to be logged as though they
+ had <code class="literal">LOG</code> priority; for common settings of
+ <code class="varname">log_min_messages</code> this results in unconditionally sending
+ them to the server log.
+ This parameter can only be set in the <code class="filename">postgresql.conf</code>
+ file or on the server command line.
+ </p></dd><dt id="GUC-TRACE-SORT"><span class="term"><code class="varname">trace_sort</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.8.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If on, emit information about resource usage during sort operations.
+ This parameter is only available if the <code class="symbol">TRACE_SORT</code> macro
+ was defined when <span class="productname">PostgreSQL</span> was compiled.
+ (However, <code class="symbol">TRACE_SORT</code> is currently defined by default.)
+ </p></dd><dt><span class="term"><code class="varname">trace_locks</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.9.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If on, emit information about lock usage. Information dumped
+ includes the type of lock operation, the type of lock and the unique
+ identifier of the object being locked or unlocked. Also included
+ are bit masks for the lock types already granted on this object as
+ well as for the lock types awaited on this object. For each lock
+ type a count of the number of granted locks and waiting locks is
+ also dumped as well as the totals. An example of the log file output
+ is shown here:
+</p><pre class="screen">
+LOG: LockAcquire: new: lock(0xb7acd844) id(24688,24696,0,0,0,1)
+ grantMask(0) req(0,0,0,0,0,0,0)=0 grant(0,0,0,0,0,0,0)=0
+ wait(0) type(AccessShareLock)
+LOG: GrantLock: lock(0xb7acd844) id(24688,24696,0,0,0,1)
+ grantMask(2) req(1,0,0,0,0,0,0)=1 grant(1,0,0,0,0,0,0)=1
+ wait(0) type(AccessShareLock)
+LOG: UnGrantLock: updated: lock(0xb7acd844) id(24688,24696,0,0,0,1)
+ grantMask(0) req(0,0,0,0,0,0,0)=0 grant(0,0,0,0,0,0,0)=0
+ wait(0) type(AccessShareLock)
+LOG: CleanUpLock: deleting: lock(0xb7acd844) id(24688,24696,0,0,0,1)
+ grantMask(0) req(0,0,0,0,0,0,0)=0 grant(0,0,0,0,0,0,0)=0
+ wait(0) type(INVALID)
+</pre><p>
+ Details of the structure being dumped may be found in
+ <code class="filename">src/include/storage/lock.h</code>.
+ </p><p>
+ This parameter is only available if the <code class="symbol">LOCK_DEBUG</code>
+ macro was defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt><span class="term"><code class="varname">trace_lwlocks</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.10.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If on, emit information about lightweight lock usage. Lightweight
+ locks are intended primarily to provide mutual exclusion of access
+ to shared-memory data structures.
+ </p><p>
+ This parameter is only available if the <code class="symbol">LOCK_DEBUG</code>
+ macro was defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt><span class="term"><code class="varname">trace_userlocks</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.11.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If on, emit information about user lock usage. Output is the same
+ as for <code class="symbol">trace_locks</code>, only for advisory locks.
+ </p><p>
+ This parameter is only available if the <code class="symbol">LOCK_DEBUG</code>
+ macro was defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt><span class="term"><code class="varname">trace_lock_oidmin</code> (<code class="type">integer</code>)
+ <a id="id-1.6.6.20.3.12.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If set, do not trace locks for tables below this OID (used to avoid
+ output on system tables).
+ </p><p>
+ This parameter is only available if the <code class="symbol">LOCK_DEBUG</code>
+ macro was defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt><span class="term"><code class="varname">trace_lock_table</code> (<code class="type">integer</code>)
+ <a id="id-1.6.6.20.3.13.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Unconditionally trace locks on this table (OID).
+ </p><p>
+ This parameter is only available if the <code class="symbol">LOCK_DEBUG</code>
+ macro was defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt><span class="term"><code class="varname">debug_deadlocks</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.14.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If set, dumps information about all current locks when a
+ deadlock timeout occurs.
+ </p><p>
+ This parameter is only available if the <code class="symbol">LOCK_DEBUG</code>
+ macro was defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt><span class="term"><code class="varname">log_btree_build_stats</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.15.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If set, logs system resource usage statistics (memory and CPU) on
+ various B-tree operations.
+ </p><p>
+ This parameter is only available if the <code class="symbol">BTREE_BUILD_STATS</code>
+ macro was defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt id="GUC-WAL-CONSISTENCY-CHECKING"><span class="term"><code class="varname">wal_consistency_checking</code> (<code class="type">string</code>)
+ <a id="id-1.6.6.20.3.16.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ This parameter is intended to be used to check for bugs in the WAL
+ redo routines. When enabled, full-page images of any buffers modified
+ in conjunction with the WAL record are added to the record.
+ If the record is subsequently replayed, the system will first apply
+ each record and then test whether the buffers modified by the record
+ match the stored images. In certain cases (such as hint bits), minor
+ variations are acceptable, and will be ignored. Any unexpected
+ differences will result in a fatal error, terminating recovery.
+ </p><p>
+ The default value of this setting is the empty string, which disables
+ the feature. It can be set to <code class="literal">all</code> to check all
+ records, or to a comma-separated list of resource managers to check
+ only records originating from those resource managers. Currently,
+ the supported resource managers are <code class="literal">heap</code>,
+ <code class="literal">heap2</code>, <code class="literal">btree</code>, <code class="literal">hash</code>,
+ <code class="literal">gin</code>, <code class="literal">gist</code>, <code class="literal">sequence</code>,
+ <code class="literal">spgist</code>, <code class="literal">brin</code>, and <code class="literal">generic</code>. Only
+ superusers can change this setting.
+ </p></dd><dt id="GUC-WAL-DEBUG"><span class="term"><code class="varname">wal_debug</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.17.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If on, emit WAL-related debugging output. This parameter is
+ only available if the <code class="symbol">WAL_DEBUG</code> macro was
+ defined when <span class="productname">PostgreSQL</span> was
+ compiled.
+ </p></dd><dt id="GUC-IGNORE-CHECKSUM-FAILURE"><span class="term"><code class="varname">ignore_checksum_failure</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.18.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Only has effect if <a class="xref" href="app-initdb.html#APP-INITDB-DATA-CHECKSUMS">data checksums</a> are enabled.
+ </p><p>
+ Detection of a checksum failure during a read normally causes
+ <span class="productname">PostgreSQL</span> to report an error, aborting the current
+ transaction. Setting <code class="varname">ignore_checksum_failure</code> to on causes
+ the system to ignore the failure (but still report a warning), and
+ continue processing. This behavior may <span class="emphasis"><em>cause crashes, propagate
+ or hide corruption, or other serious problems</em></span>. However, it may allow
+ you to get past the error and retrieve undamaged tuples that might still be
+ present in the table if the block header is still sane. If the header is
+ corrupt an error will be reported even if this option is enabled. The
+ default setting is <code class="literal">off</code>, and it can only be changed by a superuser.
+ </p></dd><dt id="GUC-ZERO-DAMAGED-PAGES"><span class="term"><code class="varname">zero_damaged_pages</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.19.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Detection of a damaged page header normally causes
+ <span class="productname">PostgreSQL</span> to report an error, aborting the current
+ transaction. Setting <code class="varname">zero_damaged_pages</code> to on causes
+ the system to instead report a warning, zero out the damaged
+ page in memory, and continue processing. This behavior <span class="emphasis"><em>will destroy data</em></span>,
+ namely all the rows on the damaged page. However, it does allow you to get
+ past the error and retrieve rows from any undamaged pages that might
+ be present in the table. It is useful for recovering data if
+ corruption has occurred due to a hardware or software error. You should
+ generally not set this on until you have given up hope of recovering
+ data from the damaged pages of a table. Zeroed-out pages are not
+ forced to disk so it is recommended to recreate the table or
+ the index before turning this parameter off again. The
+ default setting is <code class="literal">off</code>, and it can only be changed
+ by a superuser.
+ </p></dd><dt id="GUC-IGNORE-INVALID-PAGES"><span class="term"><code class="varname">ignore_invalid_pages</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.20.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If set to <code class="literal">off</code> (the default), detection of
+ WAL records having references to invalid pages during
+ recovery causes <span class="productname">PostgreSQL</span> to
+ raise a PANIC-level error, aborting the recovery. Setting
+ <code class="varname">ignore_invalid_pages</code> to <code class="literal">on</code>
+ causes the system to ignore invalid page references in WAL records
+ (but still report a warning), and continue the recovery.
+ This behavior may <span class="emphasis"><em>cause crashes, data loss,
+ propagate or hide corruption, or other serious problems</em></span>.
+ However, it may allow you to get past the PANIC-level error,
+ to finish the recovery, and to cause the server to start up.
+ The parameter can only be set at server start. It only has effect
+ during recovery or in standby mode.
+ </p></dd><dt id="GUC-JIT-DEBUGGING-SUPPORT"><span class="term"><code class="varname">jit_debugging_support</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.21.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If LLVM has the required functionality, register generated functions
+ with <span class="productname">GDB</span>. This makes debugging easier.
+ The default setting is <code class="literal">off</code>.
+ This parameter can only be set at server start.
+ </p></dd><dt id="GUC-JIT-DUMP-BITCODE"><span class="term"><code class="varname">jit_dump_bitcode</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.22.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Writes the generated <span class="productname">LLVM</span> IR out to the
+ file system, inside <a class="xref" href="runtime-config-file-locations.html#GUC-DATA-DIRECTORY">data_directory</a>. This is only
+ useful for working on the internals of the JIT implementation.
+ The default setting is <code class="literal">off</code>.
+ This parameter can only be changed by a superuser.
+ </p></dd><dt id="GUC-JIT-EXPRESSIONS"><span class="term"><code class="varname">jit_expressions</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.23.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Determines whether expressions are JIT compiled, when JIT compilation
+ is activated (see <a class="xref" href="jit-decision.html" title="31.2. When to JIT?">Section 31.2</a>). The default is
+ <code class="literal">on</code>.
+ </p></dd><dt id="GUC-JIT-PROFILING-SUPPORT"><span class="term"><code class="varname">jit_profiling_support</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.24.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ If LLVM has the required functionality, emit the data needed to allow
+ <span class="productname">perf</span> to profile functions generated by JIT.
+ This writes out files to <code class="filename">$HOME/.debug/jit/</code>; the
+ user is responsible for performing cleanup when desired.
+ The default setting is <code class="literal">off</code>.
+ This parameter can only be set at server start.
+ </p></dd><dt id="GUC-JIT-TUPLE-DEFORMING"><span class="term"><code class="varname">jit_tuple_deforming</code> (<code class="type">boolean</code>)
+ <a id="id-1.6.6.20.3.25.1.3" class="indexterm"></a>
+ </span></dt><dd><p>
+ Determines whether tuple deforming is JIT compiled, when JIT
+ compilation is activated (see <a class="xref" href="jit-decision.html" title="31.2. When to JIT?">Section 31.2</a>).
+ The default is <code class="literal">on</code>.
+ </p></dd></dl></div></div><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navfooter"><hr></hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="runtime-config-custom.html" title="19.16. Customized Options">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="runtime-config.html" title="Chapter 19. Server Configuration">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="runtime-config-short.html" title="19.18. Short Options">Next</a></td></tr><tr><td width="40%" align="left" valign="top">19.16. Customized Options </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 13.4 Documentation">Home</a></td><td width="40%" align="right" valign="top"> 19.18. Short Options</td></tr></table></div></body></html> \ No newline at end of file