2015-03-23 18:47:52 +01:00
|
|
|
/*-------------------------------------------------------------------------
|
|
|
|
*
|
|
|
|
* copy_fetch.c
|
|
|
|
* Functions for using a data directory as the source.
|
|
|
|
*
|
2019-01-02 18:44:25 +01:00
|
|
|
* Portions Copyright (c) 2013-2019, PostgreSQL Global Development Group
|
2015-03-23 18:47:52 +01:00
|
|
|
*
|
|
|
|
*-------------------------------------------------------------------------
|
|
|
|
*/
|
|
|
|
#include "postgres_fe.h"
|
|
|
|
|
|
|
|
#include <sys/stat.h>
|
|
|
|
#include <dirent.h>
|
|
|
|
#include <fcntl.h>
|
|
|
|
#include <unistd.h>
|
|
|
|
|
|
|
|
#include "datapagemap.h"
|
|
|
|
#include "fetch.h"
|
|
|
|
#include "file_ops.h"
|
|
|
|
#include "filemap.h"
|
|
|
|
#include "pg_rewind.h"
|
|
|
|
|
|
|
|
static void recurse_dir(const char *datadir, const char *path,
|
2019-05-22 19:04:48 +02:00
|
|
|
process_file_callback_t callback);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
static void execute_pagemap(datapagemap_t *pagemap, const char *path);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Traverse through all files in a data directory, calling 'callback'
|
|
|
|
* for each file.
|
|
|
|
*/
|
|
|
|
void
|
|
|
|
traverse_datadir(const char *datadir, process_file_callback_t callback)
|
|
|
|
{
|
|
|
|
recurse_dir(datadir, NULL, callback);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* recursive part of traverse_datadir
|
2015-05-28 18:44:31 +02:00
|
|
|
*
|
2015-06-11 15:31:18 +02:00
|
|
|
* parentpath is the current subdirectory's path relative to datadir,
|
2015-05-28 18:44:31 +02:00
|
|
|
* or NULL at the top level.
|
2015-03-23 18:47:52 +01:00
|
|
|
*/
|
|
|
|
static void
|
|
|
|
recurse_dir(const char *datadir, const char *parentpath,
|
|
|
|
process_file_callback_t callback)
|
|
|
|
{
|
|
|
|
DIR *xldir;
|
|
|
|
struct dirent *xlde;
|
|
|
|
char fullparentpath[MAXPGPATH];
|
|
|
|
|
|
|
|
if (parentpath)
|
|
|
|
snprintf(fullparentpath, MAXPGPATH, "%s/%s", datadir, parentpath);
|
|
|
|
else
|
|
|
|
snprintf(fullparentpath, MAXPGPATH, "%s", datadir);
|
|
|
|
|
|
|
|
xldir = opendir(fullparentpath);
|
|
|
|
if (xldir == NULL)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not open directory \"%s\": %m",
|
|
|
|
fullparentpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
while (errno = 0, (xlde = readdir(xldir)) != NULL)
|
|
|
|
{
|
|
|
|
struct stat fst;
|
2017-04-11 20:13:31 +02:00
|
|
|
char fullpath[MAXPGPATH * 2];
|
|
|
|
char path[MAXPGPATH * 2];
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
if (strcmp(xlde->d_name, ".") == 0 ||
|
|
|
|
strcmp(xlde->d_name, "..") == 0)
|
|
|
|
continue;
|
|
|
|
|
2017-04-11 20:13:31 +02:00
|
|
|
snprintf(fullpath, sizeof(fullpath), "%s/%s", fullparentpath, xlde->d_name);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
if (lstat(fullpath, &fst) < 0)
|
|
|
|
{
|
2015-04-15 22:13:32 +02:00
|
|
|
if (errno == ENOENT)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* File doesn't exist anymore. This is ok, if the new master
|
|
|
|
* is running and the file was just removed. If it was a data
|
|
|
|
* file, there should be a WAL record of the removal. If it
|
|
|
|
* was something else, it couldn't have been anyway.
|
|
|
|
*
|
|
|
|
* TODO: But complain if we're processing the target dir!
|
|
|
|
*/
|
|
|
|
}
|
|
|
|
else
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not stat file \"%s\": %m",
|
|
|
|
fullpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
if (parentpath)
|
2017-04-11 20:13:31 +02:00
|
|
|
snprintf(path, sizeof(path), "%s/%s", parentpath, xlde->d_name);
|
2015-03-23 18:47:52 +01:00
|
|
|
else
|
2017-04-11 20:13:31 +02:00
|
|
|
snprintf(path, sizeof(path), "%s", xlde->d_name);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
if (S_ISREG(fst.st_mode))
|
|
|
|
callback(path, FILE_TYPE_REGULAR, fst.st_size, NULL);
|
|
|
|
else if (S_ISDIR(fst.st_mode))
|
|
|
|
{
|
|
|
|
callback(path, FILE_TYPE_DIRECTORY, 0, NULL);
|
|
|
|
/* recurse to handle subdirectories */
|
|
|
|
recurse_dir(datadir, path, callback);
|
|
|
|
}
|
|
|
|
#ifndef WIN32
|
|
|
|
else if (S_ISLNK(fst.st_mode))
|
|
|
|
#else
|
|
|
|
else if (pgwin32_is_junction(fullpath))
|
|
|
|
#endif
|
|
|
|
{
|
|
|
|
#if defined(HAVE_READLINK) || defined(WIN32)
|
|
|
|
char link_target[MAXPGPATH];
|
2015-05-28 18:17:22 +02:00
|
|
|
int len;
|
2015-03-23 18:47:52 +01:00
|
|
|
|
2015-05-28 18:17:22 +02:00
|
|
|
len = readlink(fullpath, link_target, sizeof(link_target));
|
|
|
|
if (len < 0)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not read symbolic link \"%s\": %m",
|
|
|
|
fullpath);
|
2015-05-28 18:17:22 +02:00
|
|
|
if (len >= sizeof(link_target))
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("symbolic link \"%s\" target is too long",
|
2015-03-23 18:47:52 +01:00
|
|
|
fullpath);
|
2015-05-28 18:17:22 +02:00
|
|
|
link_target[len] = '\0';
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
callback(path, FILE_TYPE_SYMLINK, 0, link_target);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If it's a symlink within pg_tblspc, we need to recurse into it,
|
2015-05-28 18:44:31 +02:00
|
|
|
* to process all the tablespaces. We also follow a symlink if
|
2016-10-20 17:24:37 +02:00
|
|
|
* it's for pg_wal. Symlinks elsewhere are ignored.
|
2015-03-23 18:47:52 +01:00
|
|
|
*/
|
2015-05-28 18:44:31 +02:00
|
|
|
if ((parentpath && strcmp(parentpath, "pg_tblspc") == 0) ||
|
2016-10-20 17:24:37 +02:00
|
|
|
strcmp(path, "pg_wal") == 0)
|
2015-03-23 18:47:52 +01:00
|
|
|
recurse_dir(datadir, path, callback);
|
|
|
|
#else
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("\"%s\" is a symbolic link, but symbolic links are not supported on this platform",
|
2015-03-23 18:47:52 +01:00
|
|
|
fullpath);
|
Phase 2 of pgindent updates.
Change pg_bsd_indent to follow upstream rules for placement of comments
to the right of code, and remove pgindent hack that caused comments
following #endif to not obey the general rule.
Commit e3860ffa4dd0dad0dd9eea4be9cc1412373a8c89 wasn't actually using
the published version of pg_bsd_indent, but a hacked-up version that
tried to minimize the amount of movement of comments to the right of
code. The situation of interest is where such a comment has to be
moved to the right of its default placement at column 33 because there's
code there. BSD indent has always moved right in units of tab stops
in such cases --- but in the previous incarnation, indent was working
in 8-space tab stops, while now it knows we use 4-space tabs. So the
net result is that in about half the cases, such comments are placed
one tab stop left of before. This is better all around: it leaves
more room on the line for comment text, and it means that in such
cases the comment uniformly starts at the next 4-space tab stop after
the code, rather than sometimes one and sometimes two tabs after.
Also, ensure that comments following #endif are indented the same
as comments following other preprocessor commands such as #else.
That inconsistency turns out to have been self-inflicted damage
from a poorly-thought-through post-indent "fixup" in pgindent.
This patch is much less interesting than the first round of indent
changes, but also bulkier, so I thought it best to separate the effects.
Discussion: https://postgr.es/m/E1dAmxK-0006EE-1r@gemulon.postgresql.org
Discussion: https://postgr.es/m/30527.1495162840@sss.pgh.pa.us
2017-06-21 21:18:54 +02:00
|
|
|
#endif /* HAVE_READLINK */
|
2015-03-23 18:47:52 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (errno)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not read directory \"%s\": %m",
|
|
|
|
fullparentpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
if (closedir(xldir))
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not close directory \"%s\": %m",
|
|
|
|
fullparentpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Copy a file from source to target, between 'begin' and 'end' offsets.
|
|
|
|
*
|
|
|
|
* If 'trunc' is true, any existing file with the same name is truncated.
|
|
|
|
*/
|
|
|
|
static void
|
2018-01-03 21:00:11 +01:00
|
|
|
rewind_copy_file_range(const char *path, off_t begin, off_t end, bool trunc)
|
2015-03-23 18:47:52 +01:00
|
|
|
{
|
2018-09-01 21:27:12 +02:00
|
|
|
PGAlignedBlock buf;
|
2015-03-23 18:47:52 +01:00
|
|
|
char srcpath[MAXPGPATH];
|
|
|
|
int srcfd;
|
|
|
|
|
|
|
|
snprintf(srcpath, sizeof(srcpath), "%s/%s", datadir_source, path);
|
|
|
|
|
|
|
|
srcfd = open(srcpath, O_RDONLY | PG_BINARY, 0);
|
|
|
|
if (srcfd < 0)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not open source file \"%s\": %m",
|
|
|
|
srcpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
if (lseek(srcfd, begin, SEEK_SET) == -1)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not seek in source file: %m");
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
open_target_file(path, trunc);
|
|
|
|
|
|
|
|
while (end - begin > 0)
|
|
|
|
{
|
|
|
|
int readlen;
|
|
|
|
int len;
|
|
|
|
|
|
|
|
if (end - begin > sizeof(buf))
|
|
|
|
len = sizeof(buf);
|
|
|
|
else
|
|
|
|
len = end - begin;
|
|
|
|
|
2018-09-01 21:27:12 +02:00
|
|
|
readlen = read(srcfd, buf.data, len);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
|
|
|
if (readlen < 0)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not read file \"%s\": %m",
|
|
|
|
srcpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
else if (readlen == 0)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("unexpected EOF while reading file \"%s\"", srcpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
|
2018-09-01 21:27:12 +02:00
|
|
|
write_target_range(buf.data, begin, readlen);
|
2015-03-23 18:47:52 +01:00
|
|
|
begin += readlen;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (close(srcfd) != 0)
|
Unified logging system for command-line programs
This unifies the various ad hoc logging (message printing, error
printing) systems used throughout the command-line programs.
Features:
- Program name is automatically prefixed.
- Message string does not end with newline. This removes a common
source of inconsistencies and omissions.
- Additionally, a final newline is automatically stripped, simplifying
use of PQerrorMessage() etc., another common source of mistakes.
- I converted error message strings to use %m where possible.
- As a result of the above several points, more translatable message
strings can be shared between different components and between
frontends and backend, without gratuitous punctuation or whitespace
differences.
- There is support for setting a "log level". This is not meant to be
user-facing, but can be used internally to implement debug or
verbose modes.
- Lazy argument evaluation, so no significant overhead if logging at
some level is disabled.
- Some color in the messages, similar to gcc and clang. Set
PG_COLOR=auto to try it out. Some colors are predefined, but can be
customized by setting PG_COLORS.
- Common files (common/, fe_utils/, etc.) can handle logging much more
simply by just using one API without worrying too much about the
context of the calling program, requiring callbacks, or having to
pass "progname" around everywhere.
- Some programs called setvbuf() to make sure that stderr is
unbuffered, even on Windows. But not all programs did that. This
is now done centrally.
Soft goals:
- Reduces vertical space use and visual complexity of error reporting
in the source code.
- Encourages more deliberate classification of messages. For example,
in some cases it wasn't clear without analyzing the surrounding code
whether a message was meant as an error or just an info.
- Concepts and terms are vaguely aligned with popular logging
frameworks such as log4j and Python logging.
This is all just about printing stuff out. Nothing affects program
flow (e.g., fatal exits). The uses are just too varied to do that.
Some existing code had wrappers that do some kind of print-and-exit,
and I adapted those.
I tried to keep the output mostly the same, but there is a lot of
historical baggage to unwind and special cases to consider, and I
might not always have succeeded. One significant change is that
pg_rewind used to write all error messages to stdout. That is now
changed to stderr.
Reviewed-by: Donald Dong <xdong@csumb.edu>
Reviewed-by: Arthur Zakirov <a.zakirov@postgrespro.ru>
Discussion: https://www.postgresql.org/message-id/flat/6a609b43-4f57-7348-6480-bd022f924310@2ndquadrant.com
2019-04-01 14:24:37 +02:00
|
|
|
pg_fatal("could not close file \"%s\": %m", srcpath);
|
2015-03-23 18:47:52 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Copy all relation data files from datadir_source to datadir_target, which
|
|
|
|
* are marked in the given data page map.
|
|
|
|
*/
|
|
|
|
void
|
|
|
|
copy_executeFileMap(filemap_t *map)
|
|
|
|
{
|
|
|
|
file_entry_t *entry;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < map->narray; i++)
|
|
|
|
{
|
|
|
|
entry = map->array[i];
|
|
|
|
execute_pagemap(&entry->pagemap, entry->path);
|
|
|
|
|
|
|
|
switch (entry->action)
|
|
|
|
{
|
|
|
|
case FILE_ACTION_NONE:
|
|
|
|
/* ok, do nothing.. */
|
|
|
|
break;
|
|
|
|
|
|
|
|
case FILE_ACTION_COPY:
|
2018-01-03 21:00:11 +01:00
|
|
|
rewind_copy_file_range(entry->path, 0, entry->newsize, true);
|
2015-03-23 18:47:52 +01:00
|
|
|
break;
|
|
|
|
|
|
|
|
case FILE_ACTION_TRUNCATE:
|
|
|
|
truncate_target_file(entry->path, entry->newsize);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case FILE_ACTION_COPY_TAIL:
|
2018-01-03 21:00:11 +01:00
|
|
|
rewind_copy_file_range(entry->path, entry->oldsize,
|
|
|
|
entry->newsize, false);
|
2015-03-23 18:47:52 +01:00
|
|
|
break;
|
|
|
|
|
|
|
|
case FILE_ACTION_CREATE:
|
|
|
|
create_target(entry);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case FILE_ACTION_REMOVE:
|
|
|
|
remove_target(entry);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
close_target_file();
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
execute_pagemap(datapagemap_t *pagemap, const char *path)
|
|
|
|
{
|
|
|
|
datapagemap_iterator_t *iter;
|
|
|
|
BlockNumber blkno;
|
|
|
|
off_t offset;
|
|
|
|
|
|
|
|
iter = datapagemap_iterate(pagemap);
|
|
|
|
while (datapagemap_next(iter, &blkno))
|
|
|
|
{
|
|
|
|
offset = blkno * BLCKSZ;
|
2018-01-03 21:00:11 +01:00
|
|
|
rewind_copy_file_range(path, offset, offset + BLCKSZ, false);
|
2015-03-23 18:47:52 +01:00
|
|
|
/* Ok, this block has now been copied from new data dir to old */
|
|
|
|
}
|
2015-06-11 15:31:18 +02:00
|
|
|
pg_free(iter);
|
2015-03-23 18:47:52 +01:00
|
|
|
}
|