Most of the replay functions for WAL record types that modify more than
one page failed to ensure that those pages were locked correctly to ensure
that concurrent queries could not see inconsistent page states. This is
a hangover from coding decisions made long before Hot Standby was added,
when it was hardly necessary to acquire buffer locks during WAL replay
at all, let alone hold them for carefully-chosen periods.
The key problem was that RestoreBkpBlocks was written to hold lock on each
page restored from a full-page image for only as long as it took to update
that page. This was guaranteed to break any WAL replay function in which
there was any update-ordering constraint between pages, because even if the
nominal order of the pages is the right one, any mixture of full-page and
non-full-page updates in the same record would result in out-of-order
updates. Moreover, it wouldn't work for situations where there's a
requirement to maintain lock on one page while updating another. Failure
to honor an update ordering constraint in this way is thought to be the
cause of bug #7648 from Daniel Farina: what seems to have happened there
is that a btree page being split was rewritten from a full-page image
before the new right sibling page was written, and because lock on the
original page was not maintained it was possible for hot standby queries to
try to traverse the page's right-link to the not-yet-existing sibling page.
To fix, get rid of RestoreBkpBlocks as such, and instead create a new
function RestoreBackupBlock that restores just one full-page image at a
time. This function can be invoked by WAL replay functions at the points
where they would otherwise perform non-full-page updates; in this way, the
physical order of page updates remains the same no matter which pages are
replaced by full-page images. We can then further adjust the logic in
individual replay functions if it is necessary to hold buffer locks
for overlapping periods. A side benefit is that we can simplify the
handling of concurrency conflict resolution by moving that code into the
record-type-specfic functions; there's no more need to contort the code
layout to keep conflict resolution in front of the RestoreBkpBlocks call.
In connection with that, standardize on zero-based numbering rather than
one-based numbering for referencing the full-page images. In HEAD, I
removed the macros XLR_BKP_BLOCK_1 through XLR_BKP_BLOCK_4. They are
still there in the header files in previous branches, but are no longer
used by the code.
In addition, fix some other bugs identified in the course of making these
changes:
spgRedoAddNode could fail to update the parent downlink at all, if the
parent tuple is in the same page as either the old or new split tuple and
we're not doing a full-page image: it would get fooled by the LSN having
been advanced already. This would result in permanent index corruption,
not just transient failure of concurrent queries.
Also, ginHeapTupleFastInsert's "merge lists" case failed to mark the old
tail page as a candidate for a full-page image; in the worst case this
could result in torn-page corruption.
heap_xlog_freeze() was inconsistent about using a cleanup lock or plain
exclusive lock: it did the former in the normal path but the latter for a
full-page image. A plain exclusive lock seems sufficient, so change to
that.
Also, remove gistRedoPageDeleteRecord(), which has been dead code since
VACUUM FULL was rewritten.
Back-patch to 9.0, where hot standby was introduced. Note however that 9.0
had a significantly different WAL-logging scheme for GIST index updates,
and it doesn't appear possible to make that scheme safe for concurrent hot
standby queries, because it can leave inconsistent states in the index even
between WAL records. Given the lack of complaints from the field, we won't
work too hard on fixing that branch.
Give the correct name of the GUC parameter being complained of.
Also, emit a more suitable SQLSTATE (INVALID_PARAMETER_VALUE,
not the default INTERNAL_ERROR).
Gurjeet Singh, errcode adjustment by me
This reverts commit 090e8a984c.
Since WalSndWakeup does not exist in 9.0, it's clear that this patch
wasn't even compiled in this branch. Perhaps some variant of it is
appropriate in 9.0, but for the moment I'm just going to un-break
the buildfarm.
WALSender now woken up after each background flush by WALwriter, avoiding
multi-second replication delay for an all-async commit workload.
Replication delay reduced from 7s with default settings to 200ms, allowing
significantly reduced data loss at failover.
Andres Freund and Simon Riggs
Initialise ckptXidEpoch from starting checkpoint and maintain the correct
value as we roll forwards. This allows GetNextXidAndEpoch() to return the
correct epoch when executed during recovery. Backpatch to 9.0 when the
problem is first observable by a user.
Bug report from Daniel Farina
Previously we used ReadRecPtr rather than EndRecPtr, which was
not a serious error but caused pg_stat_replication to report
incorrect replay_location until at least one WAL record is replayed.
Fujii Masao
Fix a longstanding thinko in replay of NEXTOID and checkpoint records: we
tried to advance nextOid only if it was behind the value in the WAL record,
but the comparison would draw the wrong conclusion if OID wraparound had
occurred since the previous value. Better to just unconditionally assign
the new value, since OID assignment shouldn't be happening during replay
anyway.
The consequences of a failure to update nextOid would be pretty minimal,
since we have long had the code set up to obtain another OID and try again
if the generated value is already in use. But in the worst case there
could be significant performance glitches while such loops iterate through
many already-used OIDs before finding a free one.
The odds of a wraparound happening during WAL replay would be small in a
crash-recovery scenario, and the length of any ensuing OID-assignment stall
quite limited anyway. But neither of these statements hold true for a
replication slave that follows a WAL stream for a long period; its behavior
upon going live could be almost unboundedly bad. Hence it seems worth
back-patching this fix into all supported branches.
Already fixed in HEAD in commit c6d76d7c82.
RestoreBkpBlocks was in the habit of zeroing and refilling the target
buffer; which was perfectly safe when the code was written, but is unsafe
during Hot Standby operation. The reason is that we have coding rules
that allow backends to continue accessing a tuple in a heap relation while
holding only a pin on its buffer. Such a backend could see transiently
zeroed data, if WAL replay had occasion to change other data on the page.
This has been shown to be the cause of bug #6425 from Duncan Rance (who
deserves kudos for developing a sufficiently-reproducible test case) as
well as Bridget Frey's re-report of bug #6200. It most likely explains the
original report as well, though we don't yet have confirmation of that.
To fix, change the code so that only bytes that are supposed to change will
change, even transiently. This actually saves cycles in RestoreBkpBlocks,
since it's not writing the same bytes twice.
Also fix seq_redo, which has the same disease, though it has to work a bit
harder to meet the requirement.
So far as I can tell, no other WAL replay routines have this type of bug.
In particular, the index-related replay routines, which would certainly be
broken if they had to meet the same standard, are not at risk because we
do not have coding rules that allow access to an index page when not
holding a buffer lock on it.
Back-patch to 9.0 where Hot Standby was added.
we don't reach consistency before replaying all of the WAL. Rename the
variable to reachedConsistency, to make its intention clearer.
In master, that was an active bug because of the recent patch to
immediately PANIC if a reference to a missing page is found in WAL after
reaching consistency, as Tom Lane's test case demonstrated. In 9.1 and 9.0,
the only consequence was a misleading "consistent recovery state reached at
%X/%X" message in the log at the beginning of crash recovery (the database
is not consistent at that point yet). In 8.4, the log message was not
printed in crash recovery, even though there was a similar
reachedMinRecoveryPoint local variable that was also set early. So,
backpatch to 9.1 and 9.0.
There was a timing window between when oldestActiveXid was derived
and when it should have been derived that only shows itself under
heavy load. Move code around to ensure correct timing of derivation.
No change to StartupSUBTRANS() code, which is where this failed.
Bug report by Chris Redekop
ReadRecord's habit of using both direct references to tmpRecPtr and
references to *RecPtr (which is pointing at tmpRecPtr) triggers an
optimization bug in gcc 4.6.0, which apparently has forgotten about
aliasing rules. Avoid the compiler bug, and make the code more readable
to boot, by getting rid of the direct references. Improve the comments
while at it.
Back-patch to all supported versions, in case they get built with 4.6.0.
Tom Lane, with some cosmetic suggestions from Alex Hunsaker
crash recovery, and throw an error if not. hubert depesz lubaczewski pointed
out that that situation also happens in the crash recovery following a
system crash that happens during an online backup.
We might want to do something smarter in 9.1, like put the check back for
backups taken with pg_basebackup, but that's for another patch.
archive recovery.
It's possible to restore an online backup without recovery.conf, by simply
copying all the necessary WAL files to pg_xlog. "pg_basebackup -x" does that
too. That's the use case where this cross-check is useful.
Backpatch to 9.0. We used to do this in earlier versins, but in 9.0 the code
was inadvertently changed so that the check is only performed after archive
recovery.
Fujii Masao.
Startup process waited for cleanup lock but when hot_standby = off
the pid was not registered, so that the bgwriter would not wake
the waiting process as intended.
an old transaction running in the master, and a lot of transactions have
started and finished since, and a WAL-record is written in the gap between
the creating the running-xacts snapshot and WAL-logging it, recovery will fail
with "too many KnownAssignedXids" error. This bug was reported by
Joachim Wieland on Nov 19th.
In the same scenario, when fewer transactions have started so that all the
xids fit in KnownAssignedXids despite the first bug, a more serious bug
arises. We incorrectly initialize the clog code with the oldest still running
transaction, and when we see the WAL record belonging to a transaction with
an XID larger than one that committed already before the checkpoint we're
recovering from, we zero the clog page containing the already committed
transaction, leading to data loss.
In hindsight, trying to track xids in the known-assigned-xids array before
seeing the running-xacts record was too complicated. To fix that, hold
XidGenLock while the running-xacts snapshot is taken and WAL-logged. That
ensures that no transaction can begin or end in that gap, so that in recvoery
we know that the snapshot contains all transactions running at that point in
WAL.
location read from backup label file can be found: wasShutdown was set
incorrectly when a backup label file was found.
Jeff Davis, with a little tweaking by me.
rather than 0/0, so that we can safely use 0/0 as an invalid value. This is a
more future-proof fix for the corner-case bug in streaming replication that
was fixed yesterday. We had a similar corner-case bug with log/seg 0/0 back in
February as well. Avoiding 0/0 as a valid value should prevent bugs like that
in the future. Per Tom Lane's idea.
Back-patch to 9.0. Since this only affects bootstrapping, it makes no
difference to existing installations. We don't need to worry about the
bug in existing installations, because if you've managed to get past the
initial base backup already, you won't hit the bug in the future either.
streaming replication. We used log/seg 0/0 to indicate that no WAL segments
have been removed since startup, but 0/0 is a valid value for the very first
WAL segment after initdb. To make that disambiguous, store
(latest removed WAL segment + 1) in the global variable.
Per report from Matt Chesler, also reproduced by Greg Smith.
that WAL file containing the checkpoint redo-location can be found. This
avoids making the cluster irrecoverable if the redo location is in an earlie
WAL file than the checkpoint record.
Report, analysis and patch by Jeff Davis, with small changes by me.
idea from the start since the variable is only meant to track commit/abort
events. This patch reverts the logic around the variable to what it was in
8.4, except that the value is now kept in shared memory rather than a static
variable, so that it can be reported correctly by CreateRestartPoint (which is
executed in the bgwriter).
max_standby_streaming_delay, and revise the implementation to avoid assuming
that timestamps found in WAL records can meaningfully be compared to clock
time on the standby server. Instead, the delay limits are compared to the
elapsed time since we last obtained a new WAL segment from archive or since
we were last "caught up" to WAL data arriving via streaming replication.
This avoids problems with clock skew between primary and standby, as well
as other corner cases that the original coding would misbehave in, such
as the primary server having significant idle time between transactions.
Per my complaint some time ago and considerable ensuing discussion.
Do some desultory editing on the hot standby documentation, too.
master. Otherwise a subsequent crash could cause the master to lose WAL that
has already been applied on the slave, resulting in the slave being out of
sync and soon corrupt. Per recent discussion and an example from Robert Haas.
Fujii Masao
and retry. If the record is genuinely corrupt in the master database,
there's little hope of recovering, but it's better than simply retrying
to apply the corrupt WAL record in a tight loop without even trying to
retransmit it, which is what we used to do.
pg_last_xlog_replay_location(). Per Robert Haas's suggestion, after
Itagaki Takahiro pointed out an issue in the docs. Also, some wording
changes in the docs by me.
checkpoint_timeout to trigger restartpoints. We used to deliberately only
do time-based restartpoints, because if checkpoint_segments is small we
would spend time doing restartpoints more often than really necessary.
But now that restartpoints are done in bgwriter, they're not as
disruptive as they used to be. Secondly, because streaming replication
stores the streamed WAL files in pg_xlog, we want to clean it up more
often to avoid running out of disk space when checkpoint_timeout is large
and checkpoint_segments small.
Patch by Fujii Masao, with some minor changes by me.
the current one. Not doing this would leave the walwriter with a handle to a
deleted file if there was nothing for it to do for a long period of time,
preventing the file from being completely removed.
Reported by Tollef Fog Heen, and thanks to Heikki for some hand-holding with
the patch.
of requirements and documentation on LogStandbySnapshot(). Fixes
two minor bugs reported by Tom Lane that would lead to an incorrect
snapshot after transaction wraparound. Also fix two other problems
discovered that would give incorrect snapshots in certain cases.
ProcArrayApplyRecoveryInfo() substantially rewritten. Some minor
refactoring of xact_redo_apply() and ExpireTreeKnownAssignedTransactionIds().
