E.6.1. Migration to Version 9.6.19

A dump/restore is not required for those running 9.6.X.

However, if you are upgrading from a version earlier than 9.6.16, see Section E.9.

E.6.2. Changes

• Make contrib modules' installation scripts more secure (Tom Lane)

  Attacks similar to those described in CVE-2018-1058 could be carried out against an extension installation script, if the attacker can create objects in either the extension's target schema or the schema of some prerequisite extension. Since extensions often require superuser privilege to install, this can open a path to obtaining superuser privilege. To mitigate this risk, be more careful about the search_path used to run an installation script; disable check_function_bodies within the script; and fix catalog-adjustment queries used in some contrib modules to ensure they are secure. Also provide documentation to help third-party extension authors make their installation scripts secure. This is not a complete solution; extensions that depend on other extensions can still be at risk if installed carelessly. (CVE-2020-14350)

• In logical replication walsender, fix failure to send feedback messages after sending a keepalive message (Álvaro Herrera)

  This is a relatively minor problem when using built-in logical replication, because the built-in walreceiver will send a feedback reply (which clears the incorrect state) fairly frequently anyway. But with some other replication systems, such as pglogical, it causes significant performance issues.

• Fix slow execution of ts_headline() (Tom Lane)

  The phrase-search fix added in our previous set of minor releases could cause ts_headline() to take unreasonable amounts of time for long documents; to make matters worse, the query was not cancellable within the troublesome loop.

• Ensure the repeat() function can be interrupted by query cancel (Joe Conway)

• Fix mis-handling of NaN inputs during parallel aggregation on numeric-type columns (Tom Lane)

  If some partial aggregation workers found only NaNs while others found only non-NaNs, the results were combined incorrectly, possibly leading to the wrong overall result (i.e., not NaN when it should be).

• Undo double-quoting of index names in EXPLAIN's non-text output formats (Tom Lane, Euler Taveira)

• Fix timing of constraint revalidation in ALTER TABLE (David Rowley)

  If ALTER TABLE needs to fully rewrite the table's contents (for example, due to change of a column's data type) and also needs to scan the table to re-validate foreign keys or CHECK constraints, it sometimes did things in the wrong order, leading to odd errors such as "could not read block 0 in file "base/nnnnn/nnnnn": read only 0 of 8192 bytes".

• Cope with LATERAL references in restriction clauses attached to an un-flattened sub-SELECT in the FROM clause (Tom Lane)

  This oversight could result in assertion failures or crashes at query execution.

• Avoid believing that a never-analyzed foreign table has zero tuples (Tom Lane)

  This primarily affected the planner's estimate of the number of groups that would be obtained by GROUP BY.

• Improve error handling in the server's buffile module (Thomas Munro)

  Fix some cases where I/O errors were indistinguishable from reaching EOF, or were not reported at all. Also add details such as block numbers and byte counts where appropriate.

• Fix conflict-checking anomalies in SERIALIZABLE isolation mode (Peter Geoghegan)

  If a concurrently-inserted tuple was updated by a different concurrent transaction, and neither tuple version was visible to the current transaction's snapshot, serialization conflict checking could draw the wrong conclusions about whether the tuple was relevant to the results of the current transaction. This could allow a serializable transaction to commit when it should have failed with a serialization error.

• Avoid repeated marking of dead btree index entries as dead (Masahiko Sawada)

  While functionally harmless, this led to useless WAL traffic when checksums are enabled or wal_log_hints is on.

• Fix failure of some code paths to acquire the correct lock before modifying pg_control (Nathan Bossart, Fujii Masao)

  This oversight could allow pg_control to be written out with an inconsistent checksum, possibly causing trouble later, including inability to restart the database if it crashed before the next pg_control update.

• Fix errors in currtid() and currtid2() (Michael Paquier)

  These functions (which are undocumented and used only by ancient versions of the ODBC driver) contained coding errors that could result in crashes, or in confusing error messages such as "could not open file" when applied to a relation having no storage.

• Avoid calling elog() or palloc() while holding a spinlock (Michael Paquier, Tom Lane)

  Logic associated with replication slots had several violations of this coding rule. While the odds of trouble are quite low, an error in the called function would lead to a stuck spinlock.

• Report out-of-disk-space errors properly in pg_dump and pg_basebackup (Justin Pryzby, Tom Lane, Álvaro Herrera)

  Some code paths could produce silly reports like "could not write file: Success".

• Fix parallel restore of tables having both table-level privileges and per-column privileges (Tom Lane)

  The table-level privilege grants have to be applied first, but a parallel restore did not reliably order them that way; this could lead to "tuple concurrently updated" errors, or to disappearance of some per-column privilege grants. The fix for this is to include dependency links between such entries in the archive file, meaning that a new dump has to be taken with a corrected pg_dump to ensure that the problem will not recur.

• Ensure that pg_upgrade runs with vacuum_defer_cleanup_age set to zero in the target cluster (Bruce Momjian)

  If the target cluster's configuration has been modified to set vacuum_defer_cleanup_age to a nonzero value, that prevented freezing of the system catalogs from working properly, which caused the upgrade to fail in confusing ways. Ensure that any such setting is overridden for the duration of the upgrade.

• Fix pg_recvlogical to drain pending messages before exiting (Noah Misch)

  Without this, the replication sender might detect a send failure and exit without making the expected final update to the replication slot's LSN position. That led to re-transmitting data after the next connection. It was also possible to miss error messages sent after the last data that pg_recvlogical wants to consume.

• Fix pg_rewind's handling of just-deleted files in the source data directory (Justin Pryzby, Michael Paquier)

  When working with an on-line source database, concurrent file deletions are possible, but pg_rewind would get confused if deletion happened between seeing a file's directory entry and examining it with stat().

• Make pg_test_fsync use binary I/O mode on Windows (Michael Paquier)

  Previously it wrote the test file in text mode, which is not an accurate reflection of PostgreSQL's actual usage.

• Fix failure to initialize local state correctly in contrib/dblink (Joe Conway)

  With the right combination of circumstances, this could lead to dblink_close() issuing an unexpected remote COMMIT.

• Fix contrib/pgcrypto's misuse of deflate() (Tom Lane)

  The pgp_sym_encrypt functions could produce incorrect compressed data due to mishandling of zlib's API requirements. We have no reports of this error manifesting with stock zlib, but it can be seen when using IBM's zlibNX implementation.

• Fix corner case in decompression logic in contrib/pgcrypto's pgp_sym_decrypt functions (Kyotaro Horiguchi, Michael Paquier)

  A compressed stream can validly end with an empty packet, but the decompressor failed to handle this and would complain about corrupt data.

• Use POSIX-standard strsignal() in place of the BSD-ish sys_siglist[] (Tom Lane)

  This avoids build failures with very recent versions of glibc.

• Support building our NLS code with Microsoft Visual Studio 2015 or later (Juan José Santamaría Flecha, Davinder Singh, Amit Kapila)

• Avoid possible failure of our MSVC install script when there is a file named configure several levels above the source code tree (Arnold Müller)

  This could confuse some logic that looked for configure to identify the top level of the source tree.