Add ETIMEDOUT to ALL_CONNECTION_FAILURE_ERRNOS' list of "errnos that
identify hard failure of a previously-established network connection".
While one could imagine that this is sometimes recoverable, the same
could be said of other entries such as ENETDOWN.
In support of this, handle ETIMEDOUT on par with other socket errors
in relevant infrastructure, such as TranslateSocketError().
(I made a couple of cosmetic adjustments in TranslateSocketError(),
too.) The code now assumes that ETIMEDOUT is defined everywhere,
which it should be given that POSIX has required it since SUSv2.
Perhaps this should be back-patched, but I'm hesitant to do so given
the lack of previous complaints, and the hazard that there's a small
ABI break on Windows from redefining the symbol. Even if we decide
to do that, it'd be prudent to let this bake awhile in HEAD first.
Jelte Fennema
Discussion: https://postgr.es/m/AM5PR83MB01782BFF2978505F6D6C559AF7AA9@AM5PR83MB0178.EURPRD83.prod.outlook.com
First, as pointed out by Tom Lane and Michael Paquier, I failed to
realize that Windows' PostgresNode needs an extra pg_hba.conf line
(added by PostgresNode->set_replication_conf, called internally by
->init() when 'allows_streaming=>1' is given -- but I purposefully
omitted that). I think a good fix should be to have nodes with only
'has_archiving=>1' set up for replication too, but that's a bigger
discussion. Fix it by calling ->set_replication_conf, which is not
unprecedented, as pointed out by Andrew Dunstan.
I also forgot to uncomment a ->finish() call for a pumpable IPC::Run
file descriptor. Apparently this is innocuous in almost all platforms.
Backpatch to 14. The older branches were added this file too, but not
this particular part of the test.
Discussion: https://postgr.es/m/3000074.1632947632@sss.pgh.pa.us
Discussion: https://postgr.es/m/YVT7qwhR8JmC2kfz@paquier.xyz
Physical replication always ships WAL segment files to replicas once
they are complete. This is a problem if one WAL record is split across
a segment boundary and the primary server crashes before writing down
the segment with the next portion of the WAL record: WAL writing after
crash recovery would happily resume at the point where the broken record
started, overwriting that record ... but any standby or backup may have
already received a copy of that segment, and they are not rewinding.
This causes standbys to stop following the primary after the latter
crashes:
LOG: invalid contrecord length 7262 at A8/D9FFFBC8
because the standby is still trying to read the continuation record
(contrecord) for the original long WAL record, but it is not there and
it will never be. A workaround is to stop the replica, delete the WAL
file, and restart it -- at which point a fresh copy is brought over from
the primary. But that's pretty labor intensive, and I bet many users
would just give up and re-clone the standby instead.
A fix for this problem was already attempted in commit 515e3d84a0, but
it only addressed the case for the scenario of WAL archiving, so
streaming replication would still be a problem (as well as other things
such as taking a filesystem-level backup while the server is down after
having crashed), and it had performance scalability problems too; so it
had to be reverted.
This commit fixes the problem using an approach suggested by Andres
Freund, whereby the initial portion(s) of the split-up WAL record are
kept, and a special type of WAL record is written where the contrecord
was lost, so that WAL replay in the replica knows to skip the broken
parts. With this approach, we can continue to stream/archive segment
files as soon as they are complete, and replay of the broken records
will proceed across the crash point without a hitch.
Because a new type of WAL record is added, users should be careful to
upgrade standbys first, primaries later. Otherwise they risk the standby
being unable to start if the primary happens to write such a record.
A new TAP test that exercises this is added, but the portability of it
is yet to be seen.
This has been wrong since the introduction of physical replication, so
backpatch all the way back. In stable branches, keep the new
XLogReaderState members at the end of the struct, to avoid an ABI
break.
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Reviewed-by: Nathan Bossart <bossartn@amazon.com>
Discussion: https://postgr.es/m/202108232252.dh7uxf6oxwcy@alvherre.pgsql
Previously socket errors such as invalid socket or socket wait method failures
during benchmark caused pgbench to exit with status 0. Instead, errors during
the run should result in exit status 2.
Back-patch to v12 where pgbench started reporting exit status.
Original complaint and patch by Hayato Kuroda.
Author: Yugo Nagata, Fabien COELHO
Reviewed-by: Kyotaro Horiguchi, Fujii Masao
Discussion: https://postgr.es/m/TYCPR01MB5870057375ACA8A73099C649F5349@TYCPR01MB5870.jpnprd01.prod.outlook.com
The failure of socket wait method like "select()" doesn't terminate pgbench.
So the log level of error message when that failure happens should be ERROR.
But previously FATAL was used in that case.
Back-patch to v13 where pgbench started using common logging API.
Author: Yugo Nagata, Fabien COELHO
Reviewed-by: Kyotaro Horiguchi, Fujii Masao
Discussion: https://postgr.es/m/20210617005934.8bd37bf72efd5f1b38e6f482@sraoss.co.jp
The code inconsistently used "statistic object" or "statistics" where
the correct term, as discussed, is actually "statistics object". This
improves the state of the code to be more consistent.
While on it, fix an incorrect error message introduced in a4d75c8. This
error should never happen, as the code states, but it would be
misleading.
Author: Justin Pryzby
Reviewed-by: Álvaro Herrera, Michael Paquier
Discussion: https://postgr.es/m/20210924215827.GS831@telsasoft.com
Backpatch-through: 14
A forked logging collector in EXEC_BACKEND builds passes down file
descriptors (or HANDLEs in WIN32) through a command for files to be
reopened (for stderr and csvlog). Some of its logic was duplicated, and
this commit refactors the code with some wrapper routines for file
reopening after forking and fd grabbing when building the command for
the fork.
While on it, this simplifies a use of "long" in the code, introduced by
ab0ba6e to take care of a warning related to MinGW-W64 when mapping a
intptr_t to a printed value. "long" is 32-bit long on Windows, and
interoperability of Win32 and Win64 ensures that handles are always
32-bit significant, so we can just use "int" for the same result. This
also makes the new routines more symmetric.
This change makes easier the introduction of new log destinations in the
logging collector, and this is not the only piece of refactoring
planned. I have tested this change with EXEC_BACKEND on linux, macos,
and of course MSVC (both Win32 and Win64), but not MinGW so the
buildfarm may have something to say here.
Author: Sehrope Sarkuni, Michael Paquier
Discussion: https://postgr.es/m/CAH7T-aqswBM6JWe4pDehi1uOiufqe06DJWaU5=X7dDLyqUExHg@mail.gmail.com
It turns out that the instability complained of in commit d3c09b9b1
has an embarrassingly simple explanation. The test script waits for
the standby to flush incoming WAL to disk, but it should wait for
the WAL to be replayed, since we are testing for the effects of that
to be visible.
While at it, use wait_for_catchup instead of reinventing that logic,
and adjust $Test::Builder::Level to improve future error reports.
Back-patch to v12 where the necessary infrastructure came in
(cf. aforesaid commit). Also back-patch 7d1aa6bf1 so that the
test will actually get run.
Discussion: https://postgr.es/m/2854602.1632852664@sss.pgh.pa.us
Sequences were left out of the list of relation kinds that
verify_heapam knew how to check, though it is fairly trivial to allow
them. Doing that, and while at it, updating pg_amcheck to include
sequences in relations matched by table and relation patterns.
Author: Mark Dilger <mark.dilger@enterprisedb.com>
Discussion: https://www.postgresql.org/message-id/flat/81ad4757-92c1-4aa3-7bee-f609544837e3%40enterprisedb.com
These tests were disabled back in 2018 (commit d3c09b9b1) because of
failures observed in the buildfarm. I've not been able to reproduce
any failure on longfin's host, though, so I'm curious whether or to
what extent we've fixed the problem. Let's re-enable it (in HEAD
only) and see what blows up.
Discussion: https://postgr.es/m/2769443.1632773967@sss.pgh.pa.us
Several mistakes have piled in the code comments over the time,
including incorrect grammar, function names and simple typos. This
commit takes care of a portion of these.
No backpatch is done as this is only cosmetic.
Author: Justin Pryzby
Discussion: https://postgr.es/m/20210924215827.GS831@telsasoft.com
Discussing the low level issue of nbtree VACUUM and recovery conflicts
in btvacuumpage() now seems inappropriate. The same issue is discussed
in nbtxlog.h, as well as in a comment block above _bt_delitems_vacuum().
The comment block made more sense when it was part of a broader
discussion of nbtree VACUUM "pin scans". These were removed by commit
9f83468b.
Splitting the time field into days and microseconds is pretty
useless when we're just going to recombine those values.
It's unclear if anyone will notice the speedup in real-world
cases, but a cycle shaved is a cycle earned.
Discussion: https://postgr.es/m/2629129.1632675713@sss.pgh.pa.us
_bt_delitems_delete() is no longer the high-level entry point used by
index tuple deletion driven by index tuples whose LP_DEAD bits are set
(now called "simple index tuple deletion"). It became a lower level
routine that's only called by _bt_delitems_delete_check() following
commit d168b66682.
Remove obsolete reference to lazy_vacuum()'s onecall argument. The
function argument was removed by commit 3499df0dee.
Also remove adjoining comment block that introduces the wraparound
failsafe concept. Talking about the failsafe here no longer makes
sense, since lazy_vacuum() (and related functions) are no longer the
only place where the failsafe might be triggered. This has been the
case since commit c242baa4a8 taught VACUUM to consider triggering the
failsafe mechanism during its initial heap scan.
Be a little more vocal about the risks of remote collations not
matching local ones. Actually fixing these risks seems hard,
and I've given up on the idea that it might be back-patchable.
So the best we can do for the back branches is add documentation.
Per discussion of bug #16583 from Jiří Fejfar.
Discussion: https://postgr.es/m/2438715.1632510693@sss.pgh.pa.us
Index vacuums may happen multiple times depending on the number of dead
tuples stored, as of maintenance_work_mem for a manual VACUUM. For
autovacuum, this is controlled by autovacuum_work_mem instead, if set.
The documentation mentioned the former, but not the latter in the
context of autovacuum.
Reported-by: Nikolai Berkoff
Author: Laurenz Albe, Euler Taveira
Discussion: https://postgr.es/m/161545365522.10134.12195402324485546870@wrigleys.postgresql.org
Backpatch-through: 9.6
Point out that index tuple deletion generally needs a latestRemovedXid
value for the deletion operation's WAL record. This is bound to be the
most expensive part of the whole deletion operation now that it takes
place up front, during original execution.
This was arguably an oversight in commit 558a9165e0, which moved the
work required to generate these values from index deletion REDO routines
to original execution of index deletion operations.
unicode_east_asian_fw_table.h should not be compiled standalone, similarly
to unicode_combining_table.h, but cpluspluscheck did not get the memo.
Oversight in bab982161.
Per report from Tom Lane
That allows to include just visibilitymapdefs.h from file.c, and in turn,
remove include of postgres.h from relcache.h.
Reported-by: Andres Freund
Discussion: https://postgr.es/m/20210913232614.czafiubr435l6egi%40alap3.anarazel.de
Author: Alexander Korotkov
Reviewed-by: Andres Freund, Tom Lane, Alvaro Herrera
Backpatch-through: 13
Calculating degree of a functional dependency may allocate a lot of
memory - we have released mot of the explicitly allocated memory, but
e.g. detoasted varlena values were left behind. That may be an issue,
because we consider a lot of dependencies (all combinations), and the
detoasting may happen for each one again.
Fixed by calling dependency_degree() in a dedicated context, and
resetting it after each call. We only need the calculated dependency
degree, so we don't need to copy anything.
Backpatch to PostgreSQL 10, where extended statistics were introduced.
Backpatch-through: 10
Discussion: https://www.postgresql.org/message-id/20210915200928.GP831%40telsasoft.com
Until now, all extended statistics on a given relation were built in the
same memory context, without resetting. Some of the memory was released
explicitly, but not all of it - for example memory allocated while
detoasting values is hard to free. This is how it worked since extended
statistics were introduced in PostgreSQL 10, but adding support for
extended stats on expressions made the issue somewhat worse as it
increases the number of statistics to build.
Fixed by adding a memory context which gets reset after building each
statistics object (all the statistics kinds included in it). Resetting
it after building each statistics kind would be even better, but it
would require more invasive changes and copying of results, making it
harder to backpatch.
Backpatch to PostgreSQL 10, where extended statistics were introduced.
Author: Justin Pryzby
Reported-by: Justin Pryzby
Reviewed-by: Tomas Vondra
Backpatch-through: 10
Discussion: https://www.postgresql.org/message-id/20210915200928.GP831%40telsasoft.com
Checking that an offset number isn't past the end of a heap page's line
pointer array was just a defensive sanity check for HOT-chain traversal
code before commit 3c3b8a4b. It's etrictly necessary now, though. Add
comments that reference the issue to code in heapam that needs to get it
right.
Per suggestion from Alexander Lakhin.
Discussion: https://postgr.es/m/f76a292c-9170-1aef-91a0-59d9443b99a3@gmail.com
In postgres_fdw, pgfdw_xact_callback() and pgfdw_subxact_callback()
callback functions do almost the same thing to rollback remote toplevel-
and sub-transaction. But previously their such rollback logics were
implemented separately in each function and in different way. Which
could decrease the readability and maintainability of the code.
To fix the issue, this commit creates the common function to rollback
remote transactions, and makes those callback functions use it. Which
allows us to avoid unnecessary code duplication.
Author: Fujii Masao
Reviewed-by: Zhihong Yu, Bharath Rupireddy
Discussion: https://postgr.es/m/62fbb63a-d46c-fb47-a56d-f6be1909aa44@oss.nttdata.com
Updates/Deletes on a partition were allowed even without replica identity
after the parent table was added to a publication. This would later lead
to an error on subscribers. The reason was that we were not invalidating
the partition's relcache and the publication information for partitions
was not getting rebuilt. Similarly, we were not invalidating the
partitions' relcache after dropping a partitioned table from a publication
which will prohibit Updates/Deletes on its partition without replica
identity even without any publication.
Reported-by: Haiying Tang
Author: Hou Zhijie and Vignesh C
Reviewed-by: Vignesh C and Amit Kapila
Backpatch-through: 13
Discussion: https://postgr.es/m/OS0PR01MB6113D77F583C922F1CEAA1C3FBD29@OS0PR01MB6113.jpnprd01.prod.outlook.com
It is not appropriate for deduplication to apply single value strategy
when triggered by a bottom-up index deletion pass. This wastes cycles
because later bottom-up deletion passes will overinterpret older
duplicate tuples that deduplication actually just skipped over "by
design". It also makes bottom-up deletion much less effective for low
cardinality indexes that happen to cross a meaningless "index has single
key value per leaf page" threshold.
To fix, slightly narrow the conditions under which deduplication's
single value strategy is considered. We already avoided the strategy
for a unique index, since our high level goal must just be to buy time
for VACUUM to run (not to buy space). We'll now also avoid it when we
just had a bottom-up pass that reported failure. The two cases share
the same high level goal, and already overlapped significantly, so this
approach is quite natural.
Oversight in commit d168b666, which added bottom-up index deletion.
Author: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-WznaOvM+Gyj-JQ0X=JxoMDxctDTYjiEuETdAGbF5EUc3MA@mail.gmail.com
Backpatch: 14-, where bottom-up deletion was introduced.
This addresses two issues with the tests added in 0c39c292 for runtime
GUCs:
- Re-enable the test on Msys. The test could fail because of \r\n
generated by Msys perl. 0d91c52a has taken care of this issue.
- Allow the test to run in the context of a privileged account. CIs
running under privileged accounts would fail on permission failures, as
reported by Andres Freund. This issue is fixed by wrapping the postgres
command within pg_ctl as the latter will take care of any permissions
needed. The test checking a failure of postgres -C for a runtime
parameter with an instance running is removed, as pg_ctl produces an
unstable error code (no need for a CI to reproduce that).
Discussion: https://postgr.es/m/20210921032040.lyl4lcax37aedx2x@alap3.anarazel.de
Contrary to the output of native perl, Msys perl generates outputs with
CRLFs characters. There are already places in the TAP code where CRLFs
(\r\n) are automatically converted to LF (\n) on Msys, but we missed a
couple of places when running commands and using their output for
comparison, that would lead to failures.
This problem has been found thanks to the test added in 5adb067 using
TestLib::command_checks_all(), but after a closer look more code paths
were missing a filter.
This is backpatched all the way down to prevent any surprises if a new
test is introduced in stable branches.
Reviewed-by: Andrew Dunstan, Álvaro Herrera
Discussion: https://postgr.es/m/1252480.1631829409@sss.pgh.pa.us
Backpatch-through: 9.6
Before commit 84f5c2908, a STABLE function in a plpgsql CALL
statement's argument list would see an up-to-date snapshot,
because exec_stmt_call would push a new snapshot. I got rid of
that because the possibility of the snapshot disappearing within
COMMIT made it too hard to manage a snapshot across the CALL
statement. That's fine so far as the procedure itself goes,
but I forgot to think about the possibility of STABLE functions
within the CALL argument list. As things now stand, those'll
be executed with the Portal's snapshot as ActiveSnapshot,
keeping them from seeing updates more recent than Portal startup.
(VOLATILE functions don't have a problem because they take their
own snapshots; which indeed is also why the procedure itself
doesn't have a problem. There are no STABLE procedures.)
We can fix this by pushing a new snapshot transiently within
ExecuteCallStmt itself. Popping the snapshot before we get
into the procedure proper eliminates the management problem.
The possibly-useless extra snapshot-grab is slightly annoying,
but it's no worse than what happened before 84f5c2908.
Per bug #17199 from Alexander Nawratil. Back-patch to v11,
like the previous patch.
Discussion: https://postgr.es/m/17199-1ab2561f0d94af92@postgresql.org
I noticed that commit 0bead9af48 left this flag undocumented in
XLogSetRecordFlags, which led me to discover that the flag doesn't
actually do what the one comment on it said it does. Improve the
situation by adding some more comments.
Backpatch to 14, where the aforementioned commit appears.
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Discussion: https://postgr.es/m/202109212119.c3nhfp64t2ql@alvherre.pgsql
This runtime-computed GUC shows the number of huge pages required
for the server's main shared memory area, taking advantage of the
work done in 0c39c29 and 0bd305e. This is useful for users to estimate
the amount of huge pages required for a server as it becomes possible to
do an estimation without having to start the server and potentially
allocate a large chunk of shared memory.
The number of huge pages is calculated based on the existing GUC
huge_page_size if set, or by using the system's default by looking at
/proc/meminfo on Linux. There is nothing new here as this commit reuses
the existing calculation methods, and just exposes this information
directly to the user. The routine calculating the huge page size is
refactored to limit the number of files with platform-specific flags.
This new GUC's name was the most popular choice based on the discussion
done. This is only supported on Linux.
I have taken the time to test the change on Linux, Windows and MacOS,
though for the last two ones large pages are not supported. The first
one calculates correctly the number of pages depending on the existing
GUC huge_page_size or the system's default.
Thanks to Andres Freund, Robert Haas, Kyotaro Horiguchi, Tom Lane,
Justin Pryzby (and anybody forgotten here) for the discussion.
Author: Nathan Bossart
Discussion: https://postgr.es/m/F2772387-CE0F-46BF-B5F1-CC55516EB885@amazon.com
A broken HOT chain is not an unexpected condition, even when the offset
number points past the end of the page's line pointer array.
heap_prune_chain() does not (and never has) treated this condition as
unexpected, so derivative code in heap_index_delete_tuples() shouldn't
do so either.
Oversight in commit 4228817449.
The assertion can probably only fail on Postgres 14 and master. Earlier
releases don't have commit 3c3b8a4b, which taught VACUUM to truncate the
line pointer array of heap pages. Backpatch all the same, just to be
consistent.
Author: Peter Geoghegan <pg@bowt.ie>
Reported-By: Alexander Lakhin <exclusion@gmail.com>
Discussion: https://postgr.es/m/17197-9438f31f46705182@postgresql.org
Backpatch: 12-, just like commit 4228817449.
The upcoming shared memory stats patch drops stats for dropped objects in a
transactional manner, rather than removing them later as part of vacuum. This
means that stats for DROP inside a transaction needs to handle aborted
(sub-)transactions similar to TRUNCATE: The stats up to the DROP should be
restored.
Rename the existing infrastructure in preparation.
Author: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/20210405092914.mmxqe7j56lsjfsej@alap3.anarazel.de
An upcoming patch will add additional work to these functions. To avoid the
functions getting too complicated / doing too many things at once, split out
sub-tasks into their own functions.
Author: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/20210405092914.mmxqe7j56lsjfsej@alap3.anarazel.de
Since introduction of extended statistics, we've disallowed references
to system columns. So for example
CREATE STATISTICS s ON ctid FROM t;
would fail. But with extended statistics on expressions, it was possible
to work around this limitation quite easily
CREATE STATISTICS s ON (ctid::text) FROM t;
This is an oversight in a4d75c86bf, fixed by adding a simple check.
Backpatch to PostgreSQL 14, where support for extended statistics on
expressions was introduced.
Backpatch-through: 14
Discussion: https://postgr.es/m/20210816013255.GS10479%40telsasoft.com
It was harder than necessary to understand PostgresMain() because the code for
a normal backend was interspersed with single-user mode specific code. Split
most of the single-user mode code into its own function
PostgresSingleUserMain(), that does all the necessary setup for single-user
mode, and then hands off after that to PostgresMain().
There still is some single-user mode code in InitPostgres(), and it'd likely
be worth moving at least some of it out. But that's for later.
Reviewed-By: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Author: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/20210802164124.ufo5buo4apl6yuvs@alap3.anarazel.de
Tom Lane
Alvaro Herrera
Peter Eisentraut
Fujii Masao
Michael Paquier
Magnus Hagander
Peter Geoghegan
Thomas Munro
John Naylor
Alexander Korotkov
Tomas Vondra
Amit Kapila
Andres Freund