database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-11-22 12:22:45 +03:00
Code Activity

Revert pg_wal_replay_wait() stored procedure

Browse Source 
This commit reverts 3c5db1d6b0, and subsequent improvements and fixes
including 8036d73ae3, 867d396ccd, 3ac3ec580c, 0868d7ae70, 85b98b8d5a,
2520226c95, 014f9f34d2, e658038772, e1555645d7, 5035172e4a, 6cfebfe88b,
73da6b8d1b, and e546989a26.

The reason for reverting is a set of remaining issues.  Most notably, the
stored procedure appears to need more effort than the utility statement
to turn the backend into a "snapshot-less" state.  This makes an approach
to use stored procedures questionable.

Catversion is bumped.

Discussion: https://postgr.es/m/Zyhj2anOPRKtb0xW%40paquier.xyz
This commit is contained in:
Alexander Korotkov
2024-11-04 22:43:08 +02:00
parent 3293b718a0
commit 3a7ae6b3d9
22 changed files with 9 additions and 1020 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
src/backend/access/transam/Makefile
View File

@@ -36,8 +36,7 @@ OBJS = \
xlogreader.o \
xlogrecovery.o \
xlogstats.o \
xlogutils.o \
xlogwait.o
xlogutils.o
include $(top_srcdir)/src/backend/common.mk

1
src/backend/access/transam/meson.build
View File

@@ -24,7 +24,6 @@ backend_sources += files(
'xlogrecovery.c',
'xlogstats.c',
'xlogutils.c',
'xlogwait.c',
)
# used by frontend programs to build a frontend xlogreader

6
src/backend/access/transam/xact.c
View File

@@ -31,7 +31,6 @@
#include "access/xloginsert.h"
#include "access/xlogrecovery.h"
#include "access/xlogutils.h"
#include "access/xlogwait.h"
#include "catalog/index.h"
#include "catalog/namespace.h"
#include "catalog/pg_enum.h"
@@ -2827,11 +2826,6 @@ AbortTransaction(void)
*/
LWLockReleaseAll();
/*
* Cleanup waiting for LSN if any.
*/
WaitLSNCleanup();
/* Clear wait information and command progress indicator */
pgstat_report_wait_end();
pgstat_progress_end_command();

7
src/backend/access/transam/xlog.c
View File

@@ -62,7 +62,6 @@
#include "access/xlogreader.h"
#include "access/xlogrecovery.h"
#include "access/xlogutils.h"
#include "access/xlogwait.h"
#include "backup/basebackup.h"
#include "catalog/catversion.h"
#include "catalog/pg_control.h"
@@ -6174,12 +6173,6 @@ StartupXLOG(void)
UpdateControlFile();
LWLockRelease(ControlFileLock);
/*
* Wake up all waiters for replay LSN. They need to report an error that
* recovery was ended before reaching the target LSN.
*/
WaitLSNWakeup(InvalidXLogRecPtr);
/*
* Shutdown the recovery environment. This must occur after
* RecoverPreparedTransactions() (see notes in lock_twophase_recover())

116
src/backend/access/transam/xlogfuncs.c
View File

@@ -22,19 +22,17 @@
#include "access/xlog_internal.h"
#include "access/xlogbackup.h"
#include "access/xlogrecovery.h"
#include "access/xlogwait.h"
#include "catalog/pg_type.h"
#include "funcapi.h"
#include "miscadmin.h"
#include "pgstat.h"
#include "replication/walreceiver.h"
#include "storage/fd.h"
#include "storage/proc.h"
#include "storage/latch.h"
#include "storage/standby.h"
#include "utils/builtins.h"
#include "utils/memutils.h"
#include "utils/pg_lsn.h"
#include "utils/snapmgr.h"
#include "utils/timestamp.h"
/*
@@ -750,115 +748,3 @@ pg_promote(PG_FUNCTION_ARGS)
wait_seconds)));
PG_RETURN_BOOL(false);
}
static WaitLSNResult lastWaitLSNResult = WAIT_LSN_RESULT_SUCCESS;
/*
* Waits until recovery replays the target LSN with optional timeout. Unless
* 'no_error' provided throws an error on failure
*/
Datum
pg_wal_replay_wait(PG_FUNCTION_ARGS)
{
XLogRecPtr target_lsn = PG_GETARG_LSN(0);
int64 timeout = PG_GETARG_INT64(1);
bool no_error = PG_GETARG_BOOL(2);
if (timeout < 0)
ereport(ERROR,
(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
errmsg("\"timeout\" must not be negative")));
/*
* We are going to wait for the LSN replay. We should first care that we
* don't hold a snapshot and correspondingly our MyProc->xmin is invalid.
* Otherwise, our snapshot could prevent the replay of WAL records
* implying a kind of self-deadlock. This is the reason why
* pg_wal_replay_wait() is a procedure, not a function.
*
* At first, we should check there is no active snapshot. According to
* PlannedStmtRequiresSnapshot(), even in an atomic context, CallStmt is
* processed with a snapshot. Thankfully, we can pop this snapshot,
* because PortalRunUtility() can tolerate this.
*/
if (ActiveSnapshotSet())
PopActiveSnapshot();
/*
* At second, invalidate a catalog snapshot if any. And we should be done
* with the preparation.
*/
InvalidateCatalogSnapshot();
/* Give up if there is still an active or registered snapshot. */
if (GetOldestSnapshot())
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("pg_wal_replay_wait() must be only called without an active or registered snapshot"),
errdetail("Make sure pg_wal_replay_wait() isn't called within a transaction with an isolation level higher than READ COMMITTED, another procedure, or a function.")));
/*
* As the result we should hold no snapshot, and correspondingly our xmin
* should be unset.
*/
Assert(MyProc->xmin == InvalidTransactionId);
lastWaitLSNResult = WaitForLSNReplay(target_lsn, timeout);
if (no_error)
PG_RETURN_VOID();
/*
* Process the result of WaitForLSNReplay(). Throw appropriate error if
* needed.
*/
switch (lastWaitLSNResult)
{
case WAIT_LSN_RESULT_SUCCESS:
/* Nothing to do on success */
break;
case WAIT_LSN_RESULT_TIMEOUT:
ereport(ERROR,
(errcode(ERRCODE_QUERY_CANCELED),
errmsg("timed out while waiting for target LSN %X/%X to be replayed; current replay LSN %X/%X",
LSN_FORMAT_ARGS(target_lsn),
LSN_FORMAT_ARGS(GetXLogReplayRecPtr(NULL)))));
break;
case WAIT_LSN_RESULT_NOT_IN_RECOVERY:
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("recovery is not in progress"),
errdetail("Recovery ended before replaying target LSN %X/%X; last replay LSN %X/%X.",
LSN_FORMAT_ARGS(target_lsn),
LSN_FORMAT_ARGS(GetXLogReplayRecPtr(NULL)))));
break;
}
PG_RETURN_VOID();
}
Datum
pg_wal_replay_wait_status(PG_FUNCTION_ARGS)
{
const char *result_string = "";
/* Process the result of WaitForLSNReplay(). */
switch (lastWaitLSNResult)
{
case WAIT_LSN_RESULT_SUCCESS:
result_string = "success";
break;
case WAIT_LSN_RESULT_TIMEOUT:
result_string = "timeout";
break;
case WAIT_LSN_RESULT_NOT_IN_RECOVERY:
result_string = "not in recovery";
break;
}
PG_RETURN_TEXT_P(cstring_to_text(result_string));
}

11
src/backend/access/transam/xlogrecovery.c
View File

@@ -40,7 +40,6 @@
#include "access/xlogreader.h"
#include "access/xlogrecovery.h"
#include "access/xlogutils.h"
#include "access/xlogwait.h"
#include "backup/basebackup.h"
#include "catalog/pg_control.h"
#include "commands/tablespace.h"
@@ -1829,16 +1828,6 @@ PerformWalRecovery(void)
break;
}
/*
* If we replayed an LSN that someone was waiting for then walk
* over the shared memory array and set latches to notify the
* waiters.
*/
if (waitLSNState &&
(XLogRecoveryCtl->lastReplayedEndRecPtr >=
pg_atomic_read_u64(&waitLSNState->minWaitedLSN)))
WaitLSNWakeup(XLogRecoveryCtl->lastReplayedEndRecPtr);
/* Else, try to fetch the next WAL record */
record = ReadRecord(xlogprefetcher, LOG, false, replayTLI);
} while (record != NULL);

337
src/backend/access/transam/xlogwait.c
View File

@@ -1,337 +0,0 @@
/*-------------------------------------------------------------------------
*
* xlogwait.c
* Implements waiting for the given replay LSN, which is used in
* CALL pg_wal_replay_wait(target_lsn pg_lsn,
* timeout float8, no_error bool).
*
* Copyright (c) 2024, PostgreSQL Global Development Group
*
* IDENTIFICATION
* src/backend/access/transam/xlogwait.c
*
*-------------------------------------------------------------------------
*/
#include "postgres.h"
#include <float.h>
#include <math.h>
#include "pgstat.h"
#include "access/xlog.h"
#include "access/xlogrecovery.h"
#include "access/xlogwait.h"
#include "miscadmin.h"
#include "storage/latch.h"
#include "storage/proc.h"
#include "storage/shmem.h"
#include "utils/fmgrprotos.h"
#include "utils/pg_lsn.h"
#include "utils/snapmgr.h"
static int waitlsn_cmp(const pairingheap_node *a, const pairingheap_node *b,
void *arg);
struct WaitLSNState *waitLSNState = NULL;
/* Report the amount of shared memory space needed for WaitLSNState. */
Size
WaitLSNShmemSize(void)
{
Size size;
size = offsetof(WaitLSNState, procInfos);
size = add_size(size, mul_size(MaxBackends, sizeof(WaitLSNProcInfo)));
return size;
}
/* Initialize the WaitLSNState in the shared memory. */
void
WaitLSNShmemInit(void)
{
bool found;
waitLSNState = (WaitLSNState *) ShmemInitStruct("WaitLSNState",
WaitLSNShmemSize(),
&found);
if (!found)
{
pg_atomic_init_u64(&waitLSNState->minWaitedLSN, PG_UINT64_MAX);
pairingheap_initialize(&waitLSNState->waitersHeap, waitlsn_cmp, NULL);
memset(&waitLSNState->procInfos, 0, MaxBackends * sizeof(WaitLSNProcInfo));
}
}
/*
* Comparison function for waitLSN->waitersHeap heap. Waiting processes are
* ordered by lsn, so that the waiter with smallest lsn is at the top.
*/
static int
waitlsn_cmp(const pairingheap_node *a, const pairingheap_node *b, void *arg)
{
const WaitLSNProcInfo *aproc = pairingheap_const_container(WaitLSNProcInfo, phNode, a);
const WaitLSNProcInfo *bproc = pairingheap_const_container(WaitLSNProcInfo, phNode, b);
if (aproc->waitLSN < bproc->waitLSN)
return 1;
else if (aproc->waitLSN > bproc->waitLSN)
return -1;
else
return 0;
}
/*
* Update waitLSN->minWaitedLSN according to the current state of
* waitLSN->waitersHeap.
*/
static void
updateMinWaitedLSN(void)
{
XLogRecPtr minWaitedLSN = PG_UINT64_MAX;
if (!pairingheap_is_empty(&waitLSNState->waitersHeap))
{
pairingheap_node *node = pairingheap_first(&waitLSNState->waitersHeap);
minWaitedLSN = pairingheap_container(WaitLSNProcInfo, phNode, node)->waitLSN;
}
pg_atomic_write_u64(&waitLSNState->minWaitedLSN, minWaitedLSN);
}
/*
* Put the current process into the heap of LSN waiters.
*/
static void
addLSNWaiter(XLogRecPtr lsn)
{
WaitLSNProcInfo *procInfo = &waitLSNState->procInfos[MyProcNumber];
LWLockAcquire(WaitLSNLock, LW_EXCLUSIVE);
Assert(!procInfo->inHeap);
procInfo->procno = MyProcNumber;
procInfo->waitLSN = lsn;
pairingheap_add(&waitLSNState->waitersHeap, &procInfo->phNode);
procInfo->inHeap = true;
updateMinWaitedLSN();
LWLockRelease(WaitLSNLock);
}
/*
* Remove the current process from the heap of LSN waiters if it's there.
*/
static void
deleteLSNWaiter(void)
{
WaitLSNProcInfo *procInfo = &waitLSNState->procInfos[MyProcNumber];
LWLockAcquire(WaitLSNLock, LW_EXCLUSIVE);
if (!procInfo->inHeap)
{
LWLockRelease(WaitLSNLock);
return;
}
pairingheap_remove(&waitLSNState->waitersHeap, &procInfo->phNode);
procInfo->inHeap = false;
updateMinWaitedLSN();
LWLockRelease(WaitLSNLock);
}
/*
* Remove waiters whose LSN has been replayed from the heap and set their
* latches. If InvalidXLogRecPtr is given, remove all waiters from the heap
* and set latches for all waiters.
*/
void
WaitLSNWakeup(XLogRecPtr currentLSN)
{
int i;
ProcNumber *wakeUpProcs;
int numWakeUpProcs = 0;
wakeUpProcs = palloc(sizeof(ProcNumber) * MaxBackends);
LWLockAcquire(WaitLSNLock, LW_EXCLUSIVE);
/*
* Iterate the pairing heap of waiting processes till we find LSN not yet
* replayed. Record the process numbers to wake up, but to avoid holding
* the lock for too long, send the wakeups only after releasing the lock.
*/
while (!pairingheap_is_empty(&waitLSNState->waitersHeap))
{
pairingheap_node *node = pairingheap_first(&waitLSNState->waitersHeap);
WaitLSNProcInfo *procInfo = pairingheap_container(WaitLSNProcInfo, phNode, node);
if (!XLogRecPtrIsInvalid(currentLSN) &&
procInfo->waitLSN > currentLSN)
break;
wakeUpProcs[numWakeUpProcs++] = procInfo->procno;
(void) pairingheap_remove_first(&waitLSNState->waitersHeap);
procInfo->inHeap = false;
}
updateMinWaitedLSN();
LWLockRelease(WaitLSNLock);
/*
* Set latches for processes, whose waited LSNs are already replayed. As
* the time consuming operations, we do it this outside of WaitLSNLock.
* This is actually fine because procLatch isn't ever freed, so we just
* can potentially set the wrong process' (or no process') latch.
*/
for (i = 0; i < numWakeUpProcs; i++)
{
SetLatch(&GetPGProcByNumber(wakeUpProcs[i])->procLatch);
}
pfree(wakeUpProcs);
}
/*
* Delete our item from shmem array if any.
*/
void
WaitLSNCleanup(void)
{
/*
* We do a fast-path check of the 'inHeap' flag without the lock. This
* flag is set to true only by the process itself. So, it's only possible
* to get a false positive. But that will be eliminated by a recheck
* inside deleteLSNWaiter().
*/
if (waitLSNState->procInfos[MyProcNumber].inHeap)
deleteLSNWaiter();
}
/*
* Wait using MyLatch till the given LSN is replayed, the postmaster dies or
* timeout happens.
*/
WaitLSNResult
WaitForLSNReplay(XLogRecPtr targetLSN, int64 timeout)
{
XLogRecPtr currentLSN;
TimestampTz endtime = 0;
int wake_events = WL_LATCH_SET | WL_POSTMASTER_DEATH;
/* Shouldn't be called when shmem isn't initialized */
Assert(waitLSNState);
/* Should have a valid proc number */
Assert(MyProcNumber >= 0 && MyProcNumber < MaxBackends);
if (!RecoveryInProgress())
{
/*
* Recovery is not in progress. Given that we detected this in the
* very first check, this procedure was mistakenly called on primary.
* However, it's possible that standby was promoted concurrently to
* the procedure call, while target LSN is replayed. So, we still
* check the last replay LSN before reporting an error.
*/
if (targetLSN <= GetXLogReplayRecPtr(NULL))
return WAIT_LSN_RESULT_SUCCESS;
return WAIT_LSN_RESULT_NOT_IN_RECOVERY;
}
else
{
/* If target LSN is already replayed, exit immediately */
if (targetLSN <= GetXLogReplayRecPtr(NULL))
return WAIT_LSN_RESULT_SUCCESS;
}
if (timeout > 0)
{
endtime = TimestampTzPlusMilliseconds(GetCurrentTimestamp(), timeout);
wake_events |= WL_TIMEOUT;
}
/*
* Add our process to the pairing heap of waiters. It might happen that
* target LSN gets replayed before we do. Another check at the beginning
* of the loop below prevents the race condition.
*/
addLSNWaiter(targetLSN);
for (;;)
{
int rc;
long delay_ms = 0;
/* Recheck that recovery is still in-progress */
if (!RecoveryInProgress())
{
/*
* Recovery was ended, but recheck if target LSN was already
* replayed. See the comment regarding deleteLSNWaiter() below.
*/
deleteLSNWaiter();
currentLSN = GetXLogReplayRecPtr(NULL);
if (targetLSN <= currentLSN)
return WAIT_LSN_RESULT_SUCCESS;
return WAIT_LSN_RESULT_NOT_IN_RECOVERY;
}
else
{
/* Check if the waited LSN has been replayed */
currentLSN = GetXLogReplayRecPtr(NULL);
if (targetLSN <= currentLSN)
break;
}
/*
* If the timeout value is specified, calculate the number of
* milliseconds before the timeout. Exit if the timeout is already
* reached.
*/
if (timeout > 0)
{
delay_ms = TimestampDifferenceMilliseconds(GetCurrentTimestamp(), endtime);
if (delay_ms <= 0)
break;
}
CHECK_FOR_INTERRUPTS();
rc = WaitLatch(MyLatch, wake_events, delay_ms,
WAIT_EVENT_WAIT_FOR_WAL_REPLAY);
/*
* Emergency bailout if postmaster has died. This is to avoid the
* necessity for manual cleanup of all postmaster children.
*/
if (rc & WL_POSTMASTER_DEATH)
ereport(FATAL,
(errcode(ERRCODE_ADMIN_SHUTDOWN),
errmsg("terminating connection due to unexpected postmaster exit"),
errcontext("while waiting for LSN replay")));
if (rc & WL_LATCH_SET)
ResetLatch(MyLatch);
}
/*
* Delete our process from the shared memory pairing heap. We might
* already be deleted by the startup process. The 'inHeap' flag prevents
* us from the double deletion.
*/
deleteLSNWaiter();
/*
* If we didn't reach the target LSN, we must be exited by timeout.
*/
if (targetLSN > currentLSN)
return WAIT_LSN_RESULT_TIMEOUT;
return WAIT_LSN_RESULT_SUCCESS;
}

5
src/backend/catalog/system_functions.sql
View File

@@ -414,11 +414,6 @@ CREATE OR REPLACE FUNCTION
json_populate_recordset(base anyelement, from_json json, use_json_as_text boolean DEFAULT false)
RETURNS SETOF anyelement LANGUAGE internal STABLE ROWS 100 AS 'json_populate_recordset' PARALLEL SAFE;
CREATE OR REPLACE PROCEDURE pg_wal_replay_wait(target_lsn pg_lsn,
timeout int8 DEFAULT 0,
no_error bool DEFAULT false)
LANGUAGE internal AS 'pg_wal_replay_wait';
CREATE OR REPLACE FUNCTION pg_logical_slot_get_changes(
IN slot_name name, IN upto_lsn pg_lsn, IN upto_nchanges int, VARIADIC options text[] DEFAULT '{}',
OUT lsn pg_lsn, OUT xid xid, OUT data text)

18
src/backend/lib/pairingheap.c
View File

@@ -44,26 +44,12 @@ pairingheap_allocate(pairingheap_comparator compare, void *arg)
pairingheap *heap;
heap = (pairingheap *) palloc(sizeof(pairingheap));
pairingheap_initialize(heap, compare, arg);
return heap;
}
/*
* pairingheap_initialize
*
* Same as pairingheap_allocate(), but initializes the pairing heap in-place
* rather than allocating a new chunk of memory. Useful to store the pairing
* heap in a shared memory.
*/
void
pairingheap_initialize(pairingheap *heap, pairingheap_comparator compare,
void *arg)
{
heap->ph_compare = compare;
heap->ph_arg = arg;
heap->ph_root = NULL;
return heap;
}
/*

3
src/backend/storage/ipc/ipci.c
View File

@@ -24,7 +24,6 @@
#include "access/twophase.h"
#include "access/xlogprefetcher.h"
#include "access/xlogrecovery.h"
#include "access/xlogwait.h"
#include "commands/async.h"
#include "miscadmin.h"
#include "pgstat.h"
@@ -149,7 +148,6 @@ CalculateShmemSize(int *num_semaphores)
size = add_size(size, WaitEventCustomShmemSize());
size = add_size(size, InjectionPointShmemSize());
size = add_size(size, SlotSyncShmemSize());
size = add_size(size, WaitLSNShmemSize());
/* include additional requested shmem from preload libraries */
size = add_size(size, total_addin_request);
@@ -342,7 +340,6 @@ CreateOrAttachShmemStructs(void)
StatsShmemInit();
WaitEventCustomShmemInit();
InjectionPointShmemInit();
WaitLSNShmemInit();
}
/*

6
src/backend/storage/lmgr/proc.c
View File

@@ -36,7 +36,6 @@
#include "access/transam.h"
#include "access/twophase.h"
#include "access/xlogutils.h"
#include "access/xlogwait.h"
#include "miscadmin.h"
#include "pgstat.h"
#include "postmaster/autovacuum.h"
@@ -894,11 +893,6 @@ ProcKill(int code, Datum arg)
*/
LWLockReleaseAll();
/*
* Cleanup waiting for LSN if any.
*/
WaitLSNCleanup();
/* Cancel any pending condition variable sleep, too */
ConditionVariableCancelSleep();

9
src/backend/tcop/pquery.c
View File

@@ -1168,11 +1168,10 @@ PortalRunUtility(Portal portal, PlannedStmt *pstmt,
MemoryContextSwitchTo(portal->portalContext);
/*
* Some utility commands (e.g., VACUUM, CALL pg_wal_replay_wait()) pop the
* ActiveSnapshot stack from under us, so don't complain if it's now
* empty. Otherwise, our snapshot should be the top one; pop it. Note
* that this could be a different snapshot from the one we made above; see
* EnsurePortalSnapshotExists.
* Some utility commands (e.g., VACUUM) pop the ActiveSnapshot stack from
* under us, so don't complain if it's now empty. Otherwise, our snapshot
* should be the top one; pop it. Note that this could be a different
* snapshot from the one we made above; see EnsurePortalSnapshotExists.
*/
if (portal->portalSnapshot != NULL && ActiveSnapshotSet())
{

2
src/backend/utils/activity/wait_event_names.txt
View File

@@ -87,7 +87,6 @@ LIBPQWALRECEIVER_CONNECT "Waiting in WAL receiver to establish connection to rem
LIBPQWALRECEIVER_RECEIVE "Waiting in WAL receiver to receive data from remote server."
SSL_OPEN_SERVER "Waiting for SSL while attempting connection."
WAIT_FOR_STANDBY_CONFIRMATION "Waiting for WAL to be received and flushed by the physical standby."
WAIT_FOR_WAL_REPLAY "Waiting for a replay of the particular WAL position on the physical standby."
WAL_SENDER_WAIT_FOR_WAL "Waiting for WAL to be flushed in WAL sender process."
WAL_SENDER_WRITE_DATA "Waiting for any activity when processing replies from WAL receiver in WAL sender process."
@@ -346,7 +345,6 @@ WALSummarizer "Waiting to read or update WAL summarization state."
DSMRegistry "Waiting to read or update the dynamic shared memory registry."
InjectionPoint "Waiting to read or update information related to injection points."
SerialControl "Waiting to read or update shared <filename>pg_serial</filename> state."
WaitLSN "Waiting to read or update shared Wait-for-LSN state."
#
# END OF PREDEFINED LWLOCKS (DO NOT CHANGE THIS LINE)