Add boolean and one-dimensional list types to jsonToKv().
Add varToJson() and kvToJson() to convert Variants and KeyValues to JSON.
Contributed by Cynthia Shang.
The only change required was to remove the filter that prevented S3 storage from being used. The archive-get command did not require any modification which demonstrates that the storage interface is working as intended.
The mock/archive/3 integration test was modified to run S3 storage locally to provide coverage for the new code paths while mock/stanza/3 was modified to run S3 storage remotely to provide coverage for the existing code paths. This caused some churn in the expect logs but there was no change in behavior.
TlsClient introduced a non-blocking read which is required to read protocol messages that are linefeed-terminated rather than a known size. However, in many cases the expected number of bytes is known in advance so in that case it is more efficient to have tlsClientRead() block until all the bytes are read.
Add block parameter to all read functions and use it when a blocking read is required. For most read functions this is a noop, i.e. if the read function never blocks then it can ignore the parameter.
In passing, set the log level of storageNew*() functions to debug to expose more high-level I/O operations.
A robust HTTP client with pipelining support and automatic retries.
Using a single object to make multiple requests is more efficient because requests are pipelined whenever possible. Requests are automatically retried when the connection has been closed by the server. Any 5xx response is also retried.
Only the HTTPS protocol is currently supported.
A simple, secure TLS client intended to allow access to services that are exposed via HTTPS. We call it TLS instead of SSL because SSL methods are disabled so only TLS connections are allowed.
This object is intended to be used for multiple TLS connections against a service so tlsClientOpen() can be called each time a new connection is needed. By default, an open connection will be reused for pipelining so the user must be prepared to retry their transaction on a read/write error if the server closes the connection before it can be reused. If this behavior is not desirable then tlsClientClose() may be used to ensure that the next call to tlsClientOpen() will create a new TLS session.
Note that tlsClientRead() is non-blocking unless there are *zero* bytes to be read from the session in which case it will raise an error after the defined timeout. In any case the tlsClientRead()/tlsClientWrite()/tlsClientEof() functions should not generally be called directly. Instead use the read/write interfaces available from tlsClientIoRead()/tlsClientIoWrite().
Test certificates were generated dynamically but there are advantages to using static certificates. For example, it possible to use the same certificate between container versions. Mostly, it is easier to document the certificates if they are not buried deep in the container code.
The new test certificates are initially intended to be used with the C unit tests but they will eventually be used for integration tests as well.
Two new certificates have been defined. See test/certificate/README.md for details.
The old dynamic certificates will be retained until they are replaced.
The embedded semicolon led to inconsistent semicolons when using the macro and is not our general convention.
Remove embedded semicolons from the macros and add semicolons in usage where they were not present.
Add XmlDocument, XmlNode, and XmlNodeList objects as a thin interface layer on libxml2.
This interface is not intended to be comprehensive. Only a few libxml2 capabilities are exposed but more can be added as needed.
S3 key options (repo1-s3-key/repo1-s3-key-secret) were not required which meant that users got an ugly assertion when they were missing rather than a tidy configuration error.
Only the local/remote commands need them to be optional. This is because local/remote commands get all their options from the command line but secrets cannot be passed on the command line. Instead, secrets are passed to the local/remote commands via the protocol for any operation that needs them.
The configuration system allows required to be set per command so use that to improve the error messages while not breaking the local/remote commands.
This allows a C unit test to access data in the code repository that might be useful for testing.
Add testRepoPathSet() to set the repository path.
In passing remove extra whitespace in the TEST_RESULT_VOID() macro.
Bug Fixes:
* Fix issue with archive-push-queue-max not being honored on connection error. (Reported by Lardière Sébastien.)
* Fix static WAL segment size used to determine if archive-push-queue-max has been exceeded.
* Fix error after log file open failure when processing should continue. (Reported by vthriller.)
Features:
* Automatically enable backup checksum delta when anomalies (e.g. timeline switch) are detected. (Contributed by Cynthia Shang.)
Improvements:
* Retry all S3 5xx errors rather than just 500 internal errors. (Suggested by Craig A. James.)
These interfaces previously used the memory context of the object they were associated with and did not have their own destructors.
There are times when it is useful to free the interface without also freeing the underlying object so give IoRead and IoWrite their own memory contexts and destructors.
In passing fix a comment type in bufferRead.c.
By default the IoWrite object does not write until the output buffer is full but this is a problem for protocol messages that must be sent in order to get a response.
ioWriteFlush() is not called internally by IoWrite but can be used at any time to immediately write all bytes from the output buffer without closing the IoWrite object.
Documentation block syntax requires that at least one var be specified.
This limitation should be removed but for now add a comment to describe why a bogus var is defined.
The prior message stated that there had been a buffer overrun which is not true since the code prevents that.
In fact, this message means the parameter buffer filled while building the parameter list. Rather than display a partial list we output this message instead.
Also remove !!! which by convention we use as a marker for code that needs attention before it can be committed to master.
These macros provide a convenient way to output debug information in tests.
They are not intended to be left in test code when it is committed to master.
ioReadLine() calls ioRead(), which aggressively tries to fill the output buffer, but this doesn't play well with blocking reads.
Give ioReadLine() an option that tells it to read only what is available. That doesn't mean the function will never block but at least it won't do so by reading too far.
The report HTML generated by lcov is overly verbose and cumbersome to navigate. Since we maintain 100% coverage it's far more interesting to look at what is not covered than what is.
The new report presents all missing coverage on a single page and excludes code that is covered for brevity.
Add HTML tags for table elements.
The strExtra parameter allows adhoc tags to be added to an element for features that can't be implemented with CSS, e.g. colspan.
There are many places (and the number is growing) where a zero-terminated string constant must be transformed into a String object to be usable. This pattern wastes time and memory, especially since the created string is generally used in a read-only fashion.
Define macros to create constant String objects that are initialized at compile time rather than at run time.
The storageList() command accepts a regular expression as a filter. This works fine for local filesystems where it is relatively cheap to get a complete list of files and filter them in code. However, for remote filesystems like S3 it can be expensive to fetch a complete list of files only to discard the bulk of them locally.
S3 does not filter on regular expressions but it can accept a static prefix so this function extracts a prefix from a regular expression when possible.
Even a few characters can drastically reduce the amount of data that must be fetched remotely so the function does not try to be too clever. It requires a ^ anchor and stops scanning when the first special character is found.
Allow buffers to report a lower size than their allocated size. This means a larger buffer can be used to do the work of a smaller buffer without having to create a new buffer and concatenate.
This is useful for blocking I/O where the buffer may be too large for the amount of data that is available to read.
The Wait object accepted a double in the constructor for wait time but used TimeMSec internally. This was done for compatibility with the Perl code.
Instead, use TimeMSec in the Wait constructor and make changes as needed to calling code.
Note that Perl still uses a double for its Wait object so translation is needed in some places. There are no plans to update the Perl code as it will become obsolete.
If an object free() method was called manually when a callback was set then the callback would call free() again. This meant that each free() method had to protect against a subsequent call.
Instead, clear the callback (if present) before calling memContextFree(). This is faster (since there is no unecessary callback) and removes the need for semaphores to protect against a double free().
Code generation saved files even when they had not changed, which often caused code generation cascades. So, don't save files unless they have changed.
Use rsync to determine which files have changed since the last test run. The manifest of changed files is saved and not removed until all code generation and builds have completed. If an error occurs the work will be redone on the next run.
The eventual goal is to do all the builds from the test/repo directory created by rsync but for now it is only used to track changes.
The contents were already preserved between tests in a single test.pl run but for a separate execution the entire project had to be built from scratch, which was getting slower as we added code.
Save the important build flags in a file so the new execution knows whether the build contents can be reused.
Mounting/unmounting tmpfs on /home/[user]/test takes time, forces at least 3GB of memory to be available for tests, and makes it harder to preserve data between tests.
Instead, move mounting of tmpfs to the Vagrantfile and add it to fstab so it survives reboots.
There are a number of cases where a checksum delta is more appropriate than the default time-based delta:
* Timeline has switched since the prior backup
* File timestamp is older than recorded in the prior backup
* File size changed but timestamp did not
* File timestamp is in the future compared to the start of the backup
* Online option has changed since the prior backup
A practical example is that checksum delta will be enabled after a failover to standby due to the timeline switch. In this case, timestamps can't be trusted and our recommendation has been to run a full backup, which can impact the retention schedule and requires manual intervention.
Now, a checksum delta will be performed if the backup type is incr/diff. This means more CPU will be used during the backup but the backup size will be smaller and the retention schedule will not be impacted.
Contributed by Cynthia Shang.
We were already retrying 500 errors but 503 (rate-limiting) errors were not being retried and would cause an instant failure which aborted the command.
There are only two 5xx errors currently implemented by S3 but instead of adding 503 simply retry all 5xx errors. This is consistent with the http definition of this error class, "the server failed to fulfill an apparently valid request."
Suggested by Craig A. James.
This calculation was missed when the WAL segment size was made dynamic in preparation for PostgreSQL 11.
Fix the calculation by checking the actual WAL file sizes instead of using an estimate based on WAL segment size. This is more accurate because it takes into account .history and .backup files, which are smaller. Since the calculation is done in the async process the additional processing time should not adversely affect performance.
Remove the PG_WAL_SIZE constant and instead use local constants where the old value is still required. This is only the case for some tests and PostgreSQL 8.3 which does not provide a way to get the WAL segment size from pg_control.
If an error occurred while acquiring a lock on a remote server the error would be reported correctly, but the queue max detection code was not reached. The tests failed to detect this because they fixed the connection before queue max, allowing the ccde to be reached.
Move the queue max code before the lock so it will run even when remote connections are not working. This means that no attempt will be made to transfer WAL once queue max has been exceeded, but it makes it much more likely that the code will be reach without error.
Update tests to continue errors up to the point where queue max is exceeded.
Reported by Lardière Sébastien.
The C code was warning on failure and continuing but the Perl logging code was never updated with the same feature.
Rather than add the feature to Perl, just disable file logging if the log file cannot be opened. Log files are always opened by C first, so this will eliminate the error in Perl.
Reported by vthriller.
The existing tests were not adequate to ensure the history was being added in the correct order when some entries were loaded from a file and others added with infoPgAdd().
Contributed by Cynthia Shang.
The InfoPg object was partially modified in 960ad732 to place the current history item in position 0, but infoPgDataCurrent() didn't get updated correctly.
Remove this->indexCurrent and make the current position always equal 0. Use the new lstInsert() function when adding new history items via infoPgAdd(), but continue to use lstAdd() when loading from a file for efficiency.
This does not appear to be a live bug because infoPgDataCurrent() and infoPgAdd() are not yet used in any production code. The archive-get command is the only C code using InfoPG and it always looks at the entire list of items rather than just the current item.
Suggested by Cynthia Shang.
Bug Fixes:
* Fix missing missing URI encoding in S3 driver. (Reported by Dan Farrell.)
* Fix incorrect error message for duplicate options in configuration files. (Reported by Jesper St John.)
* Fix incorrectly reported error return in info logging. A return code of 1 from the archive-get was being logged as an error message at info level but otherwise worked correctly.
Features:
* Add checksum delta for incremental backups which uses checksums rather than timestamps to determine if files have changed. (Contributed by Cynthia Shang.)
* PostgreSQL 11 support, including configurable WAL segment size.
Improvements:
* Ignore all files in a linked tablespace directory except the subdirectory for the current version of PostgreSQL. Previously an error would be generated if other files were present and not owned by the PostgreSQL user.
* Improve info command to display the stanza cipher type. (Contributed by Cynthia Shang. Suggested by Douglas J Hunley.)
* Improve support for special characters in filenames.
* Allow delta option to be specified in the pgBackRest configuration file. (Contributed by Cynthia Shang.)
PostgreSQL 11 RC1 support was tested in 9ae3d8c46 when the u18 container was rebuilt. Nothing substantive changed after RC1 so pgBackRest is ready for PostgreSQL 11 GA.
The standard npm packages on Ubuntu 18.04 suddenly required libssl1.0 which broke the pgbackrest package builds. Installing nodejs from deb.nodesource.com seems to work fine with standard libssl.
This package is required by ScalityS3 which is used for local S3 testing.
When the filter interface internals were split out into a new header file the documentation was not moved as it should have been. Additionally some functions which should have been moved were left behind.
Move the documentation and functions to filter.internal.h and add more documentation. Filters are a tricky subject so the more documentation the better.
Also add documentation for the user-facing filter functions in filter.h.
Allow a single linefeed-terminated line to be read or written. This is useful for various protocol implementations, including HTTP and pgBackRest's protocol.
On read the maximum line size is limited to buffer-size to prevent runaway memory usage in case a linefeed is not found. This seems fine for HTTP but we may need to revisit this decision when implementing the pgBackRest protocol. Another option would be to increase the minimum buffer size (currently 16KB).
This test has been flapping since 9b9396c7. It seems to be some kind of timing issue since all integration tests pass and this unit passes on all other VMs. It only happens on Travis and is not reproducible in any development environment that we have tried.
For now, disable the test since the constant flapping is causing major delays in testing and quite a bit of time has been spent trying to identify the root cause. We are actively developing these tests and hope the issue will be identified during the course of normal development.
A number of improvements were made to the tests while searching for this issue. While none of them helped, it makes sense to keep the improvements.
Duplicating a non-multi-value option was not throwing the correct message when the option was a boolean.
The reason was that the option was being validated as a boolean before the multi-value check was being done. The validation code assumed it was operating on a string but was instead operating on a string list causing an assertion to fail.
Since it's not safe to do the multi-value check so late, move it up to the command-line and configuration file parse phases instead.
Reported by Jesper St John.
Previously this was done in two separate places by checking if an option was type hash or list.
Bad enough that it was in two places, but an upcoming bug fix will add another instance so make it a function.
There doesn't seem to be any need to implement this as a filter since current use cases (S3 authentication) work on small datasets.
So, use the single function method provided by OpenSSL for simplicity.
This constructor creates a Buffer object directly from a zero-terminated string. The old way was to create a String object first, then convert that to a Buffer using bufNewStr().
Updated in all places that used the old pattern.
PostgreSQL 11 introduces configurable WAL segment sizes, from 1MB to 1GB.
There are two areas that needed to be updated to support this: building the archive-get queue and checking that WAL has been archived after a backup. Both operations require the WAL segment size to properly build a list.
Checking the archive after a backup is still implemented in Perl and has an active database connection, so just get the WAL segment size from the database.
The archive-get command does not have a connection to the database, so get the WAL segment size from pg_control instead. This requires a deeper inspection of pg_control than has been done in the past, so it seemed best to copy the relevant data structures from each version of PostgreSQL and build a generic interface layer to address them. While this approach is a bit verbose, it has the advantage of being relatively simple, and can easily be updated for new versions of PostgreSQL.
Since the integration tests generate pg_control files for testing, teach Perl how to generate files with the correct offsets for both 32-bit and 64-bit architectures.
Unsecured, passwordless SSH can be a scary thing. If an attacker gains access to one system they can easily hop to other systems.
Add documentation on how to use the command parameter in authorized_keys to limit ssh to running a single command, pgbackrest. There is more that could be done for security but this likely addresses most needs.
Also change references to "trusted ssh" to "passwordless ssh" since this seems more correct.
Suggested by Stephen Frost, Magnus Hagander.
Use checksums rather than timestamps to determine if files have changed. This is useful in cases where the timestamps may not be trustworthy, e.g. when performing an incremental after failing over to a standby.
If checksum delta is enabled then checksums will be used for verification of resumed backups, even if they are full. Resumes have always used checksums to verify the files in the repository, enabling delta performs checksums on the database files as well.
Note that the user must manually enable this feature in cases were it would be useful or just keep in enabled all the time. A future commit will address automatically enabling the feature in cases where it seems likely to be useful.
Contributed by Cynthia Shang.
This option was previously allowed on the command-line only for no particular reason that we could determine.
Being able to specify it in the config file seems like a good idea and won't change current usage.
Contributed by Cynthia Shang.
Apparently we never needed to run this function remotely.
It will be needed by the backup checksum delta feature, so implement it now.
Contributed by Cynthia Shang.
The test to make sure that some files (e.g. pg_control) do not get removed during the backup was lost during the storage refactor committed at de7fc37f.
This did not impact the integrity of the backups, but bring it back since it is a nice sanity check.
Contributed by Cynthia Shang.
As we add storage drivers it's important to keep the tests for each completely separate. Rather than have three tests for each driver, standardize on having a single test unit for each driver.
This is a workaround for inefficient handling of many setjmps in gcc >= 4.9. Setjmp is used in all error handling, but in the unit tests each test macro contains an error handling block so they add up pretty quickly for large unit tests.
Enabling -ftree-coalesce-vars in affected versions reduces build time and memory requirements by nearly an order of magnitude. Even so, compiles are much slower than gcc <= 4.8.
We submitted a bug for this at: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87316
Which was marked as a duplicate of: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63155
For read-only repositories the Posix and CIFS drivers behave exactly the same. Since that's all we support in C right now it's valid to treat them as the same thing. An assertion has been added to remind us to add the CIFS driver before allowing the repository to be writable.
Mostly we want to make sure that the C code does not blow up when the repository type is CIFS.
Previously it was the responsibility of the individual tests to clean up after themselves. Now the test harness now does the cleanup automatically.
This means that some paths/files need to be recreated with each run but that doesn't happen very often.
An attempt has been made to remove all redundant cleanup code but it's hard to know if everything has been caught. No issues will be caused by anything that was missed, but they will continue to chew up time in the tests.
Storing the expect log (created by common/harnessLog) in the regular test directory was not ideal. It showed up in tests and made it difficult to clear the test directory between each run.
Move the expect log to a purpose-built directory one level up so it does not interfere with regular testing.
These are separated the same way in the Perl code where the remote storage driver is located in the Protocol module. However, in the C code the intention is to implement the remote storage driver as a regular driver in the storage layer rather than making a special case out of it.
So, merge the storage helpers. This also has the benefit of making the code a bit simpler.
Also separate storageSpool() and storageSpoolWrite() to make it clearer which operations require write access and to maintain consistency with the other storage helper functions.
If the total bytes read from the expect log file was 0 then the last byte of whatever was in memory before harnessLogBuffer would be set to 0.
On 32-bit systems this expressed as the high order byte of a pointer being cleared and wackiness (in the form of segfaults) ensued.
Fixed parameter constructors made adding new interface functions a burden, so we switched to using structs to define interfaces in the storage module at c49eaec7.
While propagating this pattern to the IO interfaces it became obvious that the existing variable parameter function pattern (begun in the storage module) was more succinct and consistent with the existing code.
So, use variable parameter functions to define all interfaces. This assumes that the non-interface parameters will be fixed, which seems reasonable for low-level code.
C or Perl coverage tests can now be run on any VM provided a recent enough version of Devel::Cover or lcov is available.
For now, leave u18 as the only VM to run coverage tests due to some issues with older versions of lcov.
The external storage interfaces (Storage, StorageFileRead, etc.) have been stable for a while, but internally they were calling the posix driver functions directly.
Create driver interfaces for storage, fileRead, and fileWrite and remove all references to the posix driver outside storage/driver/posix (with the exception of a direct call to pathRemove() in Perl LibC).
Posix is still the only available driver so more adjustment may be needed, but this should represent the bulk of the changes.
The posix driver was developed over time and the naming is not very consistent.
Rename the files and functions to work well with other drivers and generally favor longer names since the driver functions are seldom (eventually never) used outside the driver itself.
The Storage object represents some some optional parameters as negated if the default is true. This allows sensible defaults without having to specify most optional parameters.
However, there's no need to propagate this down to functions that require all parameters to be passed -- it makes the code and logging more confusing. Rename the parameters and update logic to remove negations.
Previously, debug log functions had to handle NULLs and truncate output to the available buffer size. This was verbose for both coding and testing.
Instead, create a function/macro combination that allows log functions to return a simple String object. The wrapper function takes care of the memory context, handles NULLs, and truncates the log string based on the available buffer size.
The archive-get command will only be executed in C if the repository is local, unencrypted, and type posix or cifs. Admittedly a limited use case, but this is just the first step in migrating the archive-get command entirely into C.
This is a direct migration from the Perl code (including messages) to integrate as seamlessly with the remaining Perl code as possible. It should not be possible to determine if the C version is running unless debug-level logging is enabled.
The lock is now released before the fork and reacquired after the fork so the parent process no longer needs to worry about clearing the lock.
This is the same locking mechanism that will be used once archive-get-async is exec'd as a separate command, so introduce it now to simplify testing.
The info messages were spread around and logged differently based on the execution path and in some cases logged nothing at all.
Temporarily track the async server status with a flag so that info messages are not output in the async process. The async process will be refactored as a separate command to be exec'd in a future commit.
% characters caused issues in backup/restore due to filenames being appended directly into a format string.
Reserved XML characters (<>&') caused issues in the S3 driver due to improper escaping.
Add a file with all common special characters to regression testing.
File names with uncommon characters (e.g. @) caused authentication failures due to S3 encoding them correctly while the S3 driver did not.
Reported by Dan Farrell.
By default Valgrind does not exit with an error code when a non-fatal error is detected, e.g. unfreed memory. Use the --error-exitcode option to enabled this behavior.
Update some minor issues discovered in the tests as a result. Luckily, no issues were missed in the core code.
Basic functions to detect the presence of stanza or all stop files and error when they are present.
The functionality to detect stop files without error was not migrated. This functionality is only used by stanza-delete and will be migrated with that command.
Implement rules for generating paths within the archive part of the repository. Add a helper function, storageRepo(), to create the repository storage based on configuration settings.
The repository storage helper is located in the protocol module because it will support remote file systems in the future, just as the Perl version does.
Also, improve the existing helper functions a bit using string functions that were not available when they were written.
Use JSON code now that it is available and remove temporary hacks used to get things working initially.
Use passed storage objects rather than using storageLocal(). All storage objects in C are still local but this won't always be the case.
Also, move Postgres version conversion functions to postgres/info.c since they have no dependency on the info objects and will likely be useful elsewhere.
A return code of 1 from the archive-get was being logged as an error message at info level but otherwise worked correctly.
Also improve info messages when an archive segment is or is not found.
The Perl functions do so and the integration tests rely on checking for these errors. This has been exposed as more functionality is moved into C.
Passing the errors types is now a bit complicated so instead use a flag to determine which errors to throw.
Previously an error would be generated if other files were present and not owned by the PostgreSQL user. This hasn't been a big deal in practice but it could cause issues.
Also add tests to make sure the same logic applies with links to files, i.e. all other files in the directory should be ignored. This was actually working correctly, but there were no tests for it before.
Bug Fixes:
* Fix issue where relative links in $PGDATA could be stored in the backup with the wrong path. This issue did not affect absolute links and relative tablespace links were caught by other checks. (Reported by Cynthia Shang.)
* Remove incompletely implemented online option from the check command. Offline operation runs counter to the purpose of this command, which is to check if archiving and backups are working correctly. (Reported by Jason O'Donnell.)
* Fix issue where errors raised in C were not logged when called from Perl. pgBackRest properly terminated with the correct error code but lacked an error message to aid in debugging. (Reported by Douglas J Hunley.)
* Fix issue when a boolean option (e.g. delta) was specified more than once. (Reported by Yogesh Sharma.)
Features:
* Allow any option to be set in an environment variable. This includes options that previously could only be specified on the command line, e.g. stanza, and secret options that could not be specified on the command-line, e.g. repo1-s3-key-secret.
* Exclude temporary and unlogged relation (table/index) files from backup. Implemented using the same logic as the patches adding this feature to PostgreSQL, 8694cc96 and 920a5e50. Temporary relation exclusion is enabled in PostgreSQL ≥ 9.0. Unlogged relation exclusion is enabled in PostgreSQL ≥ 9.1, where the feature was introduced. (Contributed by Cynthia Shang.)
* Allow arbitrary directories and/or files to be excluded from a backup. Misuse of this feature can lead to inconsistent backups so read the --exclude documentation carefully before using. (Reviewed by Cynthia Shang.)
* Add log-subprocess option to allow file logging for local and remote subprocesses.
* PostgreSQL 11 Beta 3 support.
Improvements:
* Allow zero-size files in backup manifest to reference a prior manifest regardless of timestamp delta. (Contributed by Cynthia Shang.)
* Improve asynchronous archive-get/archive-push performance by directly checking status files. (Contributed by Stephen Frost.)
* Improve error message when a command is missing the stanza option. (Suggested by Sarah Conway.)
Relative link paths were being combined with the paths of previous links (relative or absolute) due to the $strPath variable being modified in the current iteration rather than simply being passed to the next level of recursion.
This issue did not affect absolute links and relative tablespace links were caught by other checks, though the error was confusing.
Reported by Cynthia Shang.
Prior to this commit, an expression was used to search the spool directory for ok/error files for a specific WAL segment. This involved setting up a regular expression and using opendir/readdir.
Instead, directly probe for the status files, checking directly if a '.ok' or '.error' file exists, avoiding the regular expression and eliminating the directory scan.
Only the two files now probed for could have ever matched the regular expression which had been provided and it's unlikely that many more additional files will be added, so this is a good improvement, and optimization, with little downside.
Contributed by Stephen Frost.
Offline operation runs counter to the purpose of this command, which is to check if archiving and backups are working correctly.
Reported by Jason O'Donnell.
Contributor names have always been presented in the release notes exactly as given, but we tried to assign internal IDs based on last/first name which can be hard to determine and ultimately doesn't make sense.
Inspired by Christophe Pettus' PostgresOpen 2017 talk, "Human Beings Do Not Have a Primary Key".
Implemented using the same logic as the patches adding this feature to PostgreSQL, 8694cc96 and 920a5e50. Temporary relation exclusion is enabled in PostgreSQL ≥ 9.0. Unlogged relation exclusion is enabled in PostgreSQL ≥ 9.1, where the feature was introduced.
Contributed by Cynthia Shang.
This includes PostgreSQL installation which had previously been included in the documentation. This way produces faster builds and there is no need for us to document PostgreSQL installation.
This allows setting the test log level independently from the general test harness setting, but current only works for the C tests. It is useful for seeing log output from functions on the console while a test is running.
common/harnessLog was not ideally suited for general testing and made all the tests quite awkward. Instead, move all code used to test the common/log module into the logTest module and repurpose common/harnessLog to do log expect testing for all other tests in a cleaner way.
Add a few exceptions for config testing since the log levels are reset by default in config/parse.
This is more efficient overall and allows the caller to specify how many bytes will be read on each call. Reads are appended if the buffer already contains data but the buffer size will never increase.
Allow Buffer object "used size" to be different than "allocated size". Add functions to manage used size and remaining size and update automatically when possible.
Use strtoll() instead of sprintf() for conversion. Also use available integer min/max constants rather than hard-coded values.
Reviewed by Stephen Frost.
Suggested by Stephen Frost.
IMPORTANT NOTE: This release fixes a critical bug in the backup resume feature. All resumed backups prior to this release should be considered inconsistent. A backup will be resumed after a prior backup fails, unless resume=n has been specified. A resumed backup can be identified by checking the backup log for the message "aborted backup of same type exists, will be cleaned to remove invalid files and resumed". If the message exists, do not use this backup or any backup in the same set for a restore and check the restore logs to see if a resumed backup was restored. If so, there may be inconsistent data in the cluster.
Bug Fixes:
* Fix critical bug in resume that resulted in inconsistent backups. A regression in v0.82 removed the timestamp comparison when deciding which files from the aborted backup to keep on resume. See note above for more details. (Reported by David Youatt, Yogesh Sharma, Stephen Frost.)
* Fix error in selective restore when only one user database exists in the cluster. (Fixed by Cynthia Shang. Reported by Nj Baliyan.)
* Fix non-compliant ISO-8601 timestamp format in S3 authorization headers. AWS and some gateways were tolerant of space rather than zero-padded hours while others were not. (Fixed by Andrew Schwartz.)
Features:
* PostgreSQL 11 Beta 2 support.
Improvements:
* Improve the HTTP client to set content-length to 0 when not specified by the server. S3 (and gateways) always set content-length or transfer-encoding but HTTP 1.1 does not require it and proxies (e.g. HAProxy) may not include either. (Suggested by Adam K. Sumner.)
* Set search_path = 'pg_catalog' on PostgreSQL connections. (Suggested by Stephen Frost.)
A regression in v0.82 removed the timestamp comparison when deciding which files from the aborted backup to keep on resume. All resumed backups should be considered inconsistent. A resumed backup can be identified by checking the log for the message "aborted backup of same type exists, will be cleaned to remove invalid files and resumed".
Reported by David Youatt, Yogesh Sharma, Stephen Frost.
S3 (and gateways) always set content-length or transfer-encoding but HTTP 1.1 does not require it and proxies (e.g. HAProxy) may not include either.
Suggested by Adam K. Sumner.
* Build containers from scratch for more accurate testing.
* Allow environment load to be skipped.
* Allow bash wrapping to be skipped.
* Allow forcing a command to run as a user without sudo.
Bug Fixes:
* Fix potential buffer overrun in error message handling. (Reported by Lætitia.)
* Fix archive write lock being taken for the synchronous archive-get command. (Reported by Uspen.)
Improvements:
* Embed exported C functions and Perl modules directly into the pgBackRest executable.
* Use time_t instead of __time_t for better portability. (Suggested by Nick Floersch.)
* Print total runtime in milliseconds at command end.
Low-level functions only include stack trace in test builds while higher-level functions ship with stack trace built-in. Stack traces include all parameters passed to the function but production builds only create the parameter list when the log level is set high enough, i.e. debug or trace depending on the function.
* Allow more than one test to provide coverage for the same module.
* Add option to disable valgrind.
* Add option to disabled coverage.
* Add option to disable debug build.
* Add option to disable compiler optimization.
* Add --dev-test mode.
Bug Fixes:
* Fix directory syncs running recursively when only the specified directory should be synced. (Reported by Craig A. James.)
* Fix archive-copy throwing "path not found" error for incr/diff backups. (Reported by yummyliu, Vitaliy Kukharik.)
* Fix failure in manifest build when two or more files in PGDATA are linked to the same directory. (Reported by Vitaliy Kukharik.)
* Fix delta restore failing when a linked file is missing.
* Fix rendering of key/value and list options in help. (Reported by Clinton Adams.)
Features:
* Add asynchronous, parallel archive-get. This feature maintains a queue of WAL segments to help reduce latency when PostgreSQL requests a WAL segment with restore_command.
* Add support for additional pgBackRest configuration files in the directory specified by the --config-include-path option. Add --config-path option for overriding the default base path of the --config and --config-include-path option. (Contributed by Cynthia Shang.)
* Add repo-s3-token option to allow temporary credentials tokens to be configured. pgBackRest currently has no way to request new credentials so the entire command (e.g. backup, restore) must complete before the credentials expire. (Contributed by Yogesh Sharma.)
Improvements:
* Update the archive-push-queue-max, manifest-save-threshold, and buffer-size options to accept values in KB, MB, GB, TB, or PB where the multiplier is a power of 1024. (Contributed by Cynthia Shang.)
* Make backup/restore path sync more efficient. Scanning the entire directory can be very expensive if there are a lot of small tables. The backup manifest contains the path list so use it to perform syncs instead of scanning the backup/restore path.
* Show command parameters as well as command options in initial info log message.
* Rename archive-queue-max option to archive-push-queue-max to avoid confusion with the new archive-get-queue-max option. The old option name will continue to be accepted.
pgBackRest currently has no way to request new credentials so the entire command (e.g. backup, restore) must complete before the credentials expire.
Contributed by Yogesh Sharma.
Many options that were set per test can instead be inferred from the types, i.e. container, c, expect, and individual.
Also finish renaming Perl unit tests with the -perl suffix.
* Add storageCopy(), storageMove(), and storagePathSync().
* Separate StorageFile object into separate read and write objects.
* Abstract out Posix file read/write objects.
Configuration files are loaded from the directory specified by the --config-include-path option.
Add --config-path option for overriding the default base path of the --config and --config-include-path option.
Contributed by Cynthia Shang.
Mainly this helps with unit tests that need to do log expect testing. Add harnessCfgLoad() test function, which allows a new config to be loaded for unit testing without resetting log functions, opening a log file, or taking locks.
The Perl process was exiting directly when called but that interfered with proper locking for the forked async process. Now Perl returns results to the C process which handles all errors, including signals.
Now only two types of locks can be taken: archive and backup. Most commands use one or the other but the stanza-* commands acquire both locks. This provides better protection than the old command-based locking scheme.
This implementation should be faster because it does not stat each file. It simply assumes that most directory entries are files so attempts an unlink() first. If the entry is reported by error codes to be a directory then it attempts an rmdir().
This makes it easier to create objects and then copy them to another context when they are complete without having to worry about freeing them on error. Update List, StringList, and Buffer to allow moves. Update Ini and Storage to take advantage of moves.
Scanning the entire backup directory can be very expensive if there are a lot of small tables. The backup manifest contains the backup directory list so use it to perform syncs instead of scanning the backup directory.
* Perform apt-get update to ensure packages are up to date before installing.
* Add -p to the repository mkdir so it won't fail if the directory already exists, handy for testing packages.
Switch from Devel::Cover because it would not report on branch coverage for reports converted from gcov.
Branch coverage is not complete, so for the time being errors will only be generated when statement coverage is not complete. Coverage of unit tests is not displayed in the report unless they are incomplete for either statement or branch coverage.
Bug Fixes:
* Fix --target-action and --recovery-option options being reported as invalid when restoring with --type=immediate. (Reported by Brad Nicholson.)
* Immediately error when a secure option (e.g. repo1-s3-key) is passed on the command line. Since pgBackRest would not pass secure options on to sub-processes an obscure error was thrown. The new error is much clearer and provides hints about how to fix the problem. Update command documentation to omit secure options that cannot be specified on the command-line. (Reported by Brad Nicholson.)
* Fix issue passing --no-config to embedded Perl. (Reported by Ibrahim Edib Kokdemir.)
* Fix issue where specifying log-level-stderr > warn would cause a local/remote process to error on exit due to output found on stderr when none was expected. The max value for a local/remote process is now error since there is no reason for these processes to emit warnings. (Reported by Clinton Adams.)
* Fix manifest test in the check command when tablespaces are present. (Fixed by Cynthia Shang. Reported by Thomas Flatley.)
Improvements:
* Error when multiple arguments are set in the config file for an option that does not accept multiple arguments. (Contributed by Cynthia Shang.)
* Remove extraneous sudo commands from src/Makefile. (Contributed by Adrian Vondendriesch.)
Since pgBackRest would not pass secure options on to sub-processes an obscure error was thrown. The new error is much clearer and provides hints about how to fix the problem.
Update command documentation to omit secure options that cannot be specified on the command-line.
Reported by Brad Nicholson.
* Replace remaining NDEBUG blocks with the more granular DEBUG_UNIT.
* Remove some debug memset() calls in MemContext since valgrind is more useful for these checks.
Move command begin to C except when it must be called after another command in Perl (e.g. expire after backup). Command begin logs correctly for complex data types like hash and list. Specify which commands will log to file immediately and set the default log level for log messages that are common to all commands. File logging is initiated from C.
Features:
* The archive-push command is now partially coded in C which allows the PostgreSQL archive_command to run significantly faster when processing status messages from the asynchronous archive process. (Reviewed by Cynthia Shang.)
Improvements:
* Improve check command to verify that the backup manifest can be built. (Contributed by Cynthia Shang.)
* Improve performance of HTTPS client. Buffering now takes the pending bytes on the socket into account (when present) rather than relying entirely on select(). In some instances the final bytes would not be flushed until the connection was closed.
* Improve S3 delete performance. The constant S3_BATCH_MAX had been replaced with a hard-coded value of 2, probably during testing.
* Allow any non-command-line option to be reset to default on the command-line. This allows options in pgbackrest.conf to be reset to default which reduces the need to write new configuration files for specific needs.
* The C library is now required. This eliminates conditional loading and eases development of new library features.
* The pgbackrest executable is now a C binary instead of Perl. This allows certain time-critical commands (like async archive-push) to run more quickly.
* Rename db-* options to pg-* and backup-* options to repo-* to improve consistency. repo-* options are now indexed although currently only one is allowed.
It would be better if the hostnames were also pg1 and pg2 to illustrate that primaries and standbys can change hosts, but at this time the configuration ends up being confusing since pg1, pg2, etc. are also used in the option naming. So, for now leave the names as pg-primary and pg-standby to avoid confusion.
The host-*-ip variable is created from the id param so the name param can be changed without affecting the host-*-ip variable. If id is not specified then it is copied from name.
Buffering now takes the pending bytes on the socket into account (when present) rather than relying entirely on select(). In some instances the final bytes would not be flushed until the connection was closed.
* Check both doc-path and bin-path for logo.
* Allow PDF to be output to a location other than the output directory.
* Use PDF-specific version variable for more flexible formatting.
1) Error when the same option is defined multiple times in a section using alternate names.
2) Fix logging of invalid command error.
3) Warn when a stanza-only option is in a global section.
Also, make a note to add validation of section names to the check command.
Per review by Cynthia Shang.
The log-level-console option should not be changed in the parent process. Even though it is harmless at the moment, that may not always be true.
Per review by Cynthia Shang.
This provides correct matching in the event there are system-id and db-version duplicates (e.g. after reverting a pg_upgrade).
Fixed by Cynthia Shang.
Reported by Adam K. Sumner.
This allows specific options in pgbackrest.conf to be ignored (and set to default) which reduces the need to write new configuration files for specific needs.
Note that boolean, non-command-line options are already negatable.
When a backup host is present, backups should only be allowed on the backup host and restores should only be allowed on the database host unless an alternate configuration is created that ignores the remote host.
Reported by Lardière Sébastien.
Required to test restores on the backup server, a fairly common scenario.
Improve the restore function to accept optional parameters rather than a long list of parameters. In passing, clean up extraneous use of strType and strComment variables.
Bug Fixes:
* Fixed an issue that could cause copying large manifests to fail during restore. (Reported by Craig A. James.)
* Fixed incorrect WAL offset for 32-bit architectures. (Fixed by Javier Wilson.)
* Fixed an issue retrieving WAL for old database versions. After a stanza-upgrade it should still be possible to restore backups from the previous version and perform recovery with archive-get. However, archive-get only checked the most recent db version/id and failed. Also clean up some issues when the same db version/id appears multiple times in the history. (Fixed by Cynthia Shang. Reported by Clinton Adams.)
* Fixed an issue with invalid backup groups being set correctly on restore. If the backup cannot map a group to a name it stores the group in the manifest as false then uses either the owner of $PGDATA to set the group during restore or failing that the group of the current user. This logic was not working correctly because the selected group was overwriting the user on restore leaving the group undefined and the user incorrectly set to the group. (Reported by Jeff McCormick.)
* Fixed an issue passing parameters to remotes. When more than one db was specified the path, port, and socket path would for db1 were passed no matter which db was actually being addressed. (Reported by Uspen.)
Features:
* Repository encryption support. (Contributed by Cynthia Shang, David Steele.)
When more than one db was specified the path, port, and socket path would for db1 were passed no matter which db was actually being addressed.
Reported by Uspen.
If the backup cannot map a group to a name it stores the group in the manifest as false then uses either the owner of $PGDATA to set the group during restore or failing that the group of the current user. This logic was not working correctly because the selected group was overwriting the user on restore leaving the group undefined and the user incorrectly set to the group. (Reported by Jeff McCormick.)
Different encoded strings could be generated based on compiler optimizations. Even though decoding was still successful the encoded strings did not match the standard.
The existing static files would not work with 32-bit or big-endian systems so create functions to generate these files dynamically rather than creating a bunch of new static files.
Running coverage testing on multiple distros takes time but doesn't add significant value. Also ensure that the distro designated to run coverage tests is one of the default test distros.
After a stanza-upgrade it should still be possible to restore backups from the previous version and perform recovery with archive-get. However, archive-get only checked the most recent db version/id and failed.
Also clean up some issues when the same db version/id appears multiple times in the history.
Fixed by Cynthia Shang.
Reported by Clinton Adams.
db-path was the only option with a hint so the feature seemed wasteful. All missing stanza options now output the same hint without needing configuration.
Bug Fixes:
* Fix custom settings for compress-level option being ignored. (Reported by Jens Wilke.)
* Remove error when overlapping timelines are detected. Overlapping timelines are valid in many Point-in-Time-Recovery (PITR) scenarios. (Reported by blogh.)
* Fix instances where database-id was not rendered as an integer in JSON info output. (Fixed by Cynthia Shang. Reported by Jason O'Donnell.)
Features:
* Improve performance of list requests on S3. Any beginning literal portion of a filter expression is used to generate a search prefix which often helps keep the request small enough to avoid rate limiting. (Suggested by Mihail Shvein.)
Previously, functions with sensitive options had to be logged at trace level to avoid exposing them. Trace level logging may still expose secrets so use with caution.
Any beginning literal portion of a filter expression is used to generate a search prefix which often helps keep the request small enough to avoid rate limiting.
Suggested by Mihail Shvein.
Bug Fixes:
* Fixed an issue where warnings were being emitted in place of lower priority log messages during backup from standby initialization. (Reported by Uspen.)
* Fixed an issue where some db-* options (e.g. db-port) were not being passed to remotes. (Reported by Uspen.)
Features:
* Exclude contents of pg_snapshots, pg_serial, pg_notify, and pg_dynshmem from backup since they are rebuilt on startup.
* Exclude pg_internal.init files from backup since they are rebuilt on startup.
* Exclude contents of pg_snapshots, pg_serial, pg_notify, and pg_dynshmem from backup since they are rebuilt on startup.
* Exclude pg_internal.init files from backup since they are rebuilt on startup.
Bug Fixes:
* Fixed an issue that could cause compression to abort on growing files. (Reported by Jesper St John, Aleksandr Rogozin.)
* Fixed an issue with keep-alives not being sent to the remote from the local process. (Reported by William Cox.)
Features:
* Up to seven standbys can be configured for backup from standby. (Contributed by Cynthia Shang.)
* PostgreSQL 10 support.
* Allow content-length (in addition to chunked encoding) when reading XML data to improve compatibility with third-party S3 gateways. (Suggested by Victor Gdalevich.)
Refactoring:
* Configuration rules are now pulled from the C library when present.
* Increase HTTP timeout for S3.
* Add HTTP retries to harden against transient S3 network errors.
Bug Fixes:
* The archive_status directory is now recreated on restore to support PostgreSQL 8.3 which does not recreate it automatically like more recent versions do. (Reported by Stephen Frost.)
* Fixed an issue that could cause the empty archive directory for an old PostgreSQL version to be left behind after a stanza-upgrade. (Fixed by Cynthia Shang.)
Features:
* Modified the info command (both text and JSON output) to display the archive ID and minimum/maximum WAL currently present in the archive for the current and prior, if any, database cluster version. (Contributed by Cynthia Shang.)
* Added --backup-ssh-port and --db-ssh-port options to support non-default SSH ports. (Contributed by Cynthia Shang.)
Refactoring:
* Retry when S3 returns an internal error (500).
* Add bIgnoreMissing parameter to Local->manifest().
* Move repository options into a separate section in command/command-line help. (Suggested by Stephen Frost.)
* Fix description of --online based on the command context.
* Remove vestigial repository options from backup command.
The archive_status directory is now recreated on restore to support PostgreSQL 8.3 which does not recreate it automatically like more recent versions do.
Also fixed log checking after PostgreSQL shuts down to include FATAL messages and disallow immediate shutdowns which can throw FATAL errors in the log.
Reported by Stephen Frost.
Modified the info command (both text and JSON output) to display the archive ID and minimum/maximum WAL currently present in the archive for the current and prior, if any, database cluster version.
Contributed by Cynthia Shang.
Bug Fixes:
* Fixed an issue that prevented tablespaces from being backed up on PostgreSQL ≤ 8.4.
* Fixed missing flag in C library build that resulted in a mismatched binary on 32-bit systems. (Reported by Adrian Vondendriesch.)
Features:
* Add s3-repo-ca-path and s3-repo-ca-file options to accommodate systems where CAs are not automatically found by IO::Socket::SSL, i.e. RHEL7, or to load custom CAs. (Suggested by Scott Frazer.)
Refactoring:
* Harden protocol handshake to handle race conditions.
* Fixed misleading error message when a file was opened for write in a missing directory.
* Change log level of hardlink logging to detail.
* Cast size in S3 manifest to integer.
* Rename Archive modules to remove redundancy.
* Improve S3 error reporting.
* Minor optimizations to package loads and ordering for archive-get and archive-push commands.
The integration tests that were supposed to prevent this regression did not work as intended. They verified the contents of a table in the (supposedly) restored tablespace, deleted the table, and then deleted the tablespace. All of this was deemed sufficient to prove that the tablespace had been restored correctly and was valid.
However, PostgreSQL will happily recreate a tablespace on the basis of a single full-page write, at least in the affected versions. Since writes to the test table were replayed from WAL with each recovery, all the tests passed even though the tablespace was missing after the restore.
The tests have been updated to include direct comparisons against the file system and a new table that is not replayed after a restore because it is created before the backup and never modified again.
Versions ≥ 9.0 were not affected due to numerous synthetic integration tests that verify backups and restores file by file.
* More optimized container suite that greatly improves build time.
* Added static Debian packages for Devel::Cover to reduce build time.
* Add deprecated state for containers. Deprecated containers may only be used to build packages.
* Remove Debian 8 from CI because it does not provide additional coverage over Ubuntu 14.04 and Ubuntu 16.04.
The options accommodate systems where CAs are not automatically found by IO::Socket::SSL, i.e. RHEL7, or to load custom CAs.
Suggested by Scott Frazer.
* Combine hardlink and non/compressed in synthetic tests to reduce test time and improve coverage.
* Change log level of hardlink logging to detail.
* Cast size in S3 manifest to integer.
Bug Fixes:
* Fixed the info command so the WAL archive min/max displayed is for the current database version. (Fixed by Cynthia Shang.)
* Fixed the backup command so the backup-standby option is reset (and the backup proceeds on the master) if the standby is not configured and/or reachable. (Fixed by Cynthia Shang.)
* Fixed config warnings raised from a remote process causing errors in the master process. (Fixed by Cynthia Shang.)
Features:
* Amazon S3 repository support. (Reviewed by Cynthia Shang.)
Refactoring:
* Refactor storage layer to allow for new repository filesystems using drivers. (Reviewed by Cynthia Shang.)
* Refactor IO layer to allow for new compression formats, checksum types, and other capabilities using filters. (Reviewed by Cynthia Shang.)
* Move modules in Protocol directory in subdirectories.
* Move backup modules into Backup directory.
Refactor storage layer to allow for new repository filesystems using drivers. (Reviewed by Cynthia Shang.)
Refactor IO layer to allow for new compression formats, checksum types, and other capabilities using filters. (Reviewed by Cynthia Shang.)
Bug Fixes:
* Fixed an issue where read-only operations that used local worker processes (i.e. restore) were creating write locks that could interfere with parallel archive-push. (Reported by Jens Wilke.)
Features:
* Added the stanza-upgrade command to provide a mechanism for upgrading a stanza after upgrading to a new major version of PostgreSQL. (Contributed by Cynthia Shang.)
* Added validation of pgbackrest.conf to display warnings if options are not valid or are not in the correct section. (Contributed by Cynthia Shang.)
Refactoring:
* Simplify locking scheme. Now, only the master process will hold write locks (for archive-push and backup commands) and not all local and remote worker processes as before.
* Refactor Ini.pm to facilitate testing.
* Do not set timestamps of files in the backup directories to match timestamps in the cluster directory. This was originally done to enable backup resume, but that process is now implemented with checksums.
* Improved error message when the restore command detects the presence of postmaster.pid. (Suggested by Yogesh Sharma.)
* Renumber return codes between 25 and 125 to avoid PostgreSQL interpreting some as fatal signal exceptions. (Suggested by Yogesh Sharma.)
* The backup and restore commands no longer copy via temp files. In both cases the files are checksummed on resume so there's no danger of partial copies.
* Allow functions to accept optional parameters as a hash.
* Refactor File->list() and fileList() to accept optional parameters.
* Refactor backupLabel() and add unit tests.
* Silence some perl critic warnings. (Contributed by Cynthia Shang.)
Making this dynamic in commit 5d2e792 broke doc builds from cache. The long-term solution is to create a special user for doc builds but that’s beyond the scope of this release.
* Refactor Ini.pm to facilitate testing.
* Complete statement/branch coverage for Ini.pm.
* Improved functions used to test/munge manifest and info files.
* Full coverage is verified when specified.
* Modules marked with partial coverage will error if they are actually fully covered.
* Simplified test representation is DefineTest.
* Added new representation for queries in DefineTest and added API functions.
* Update modules using DefineTest to use new API.
* Fixed an issue where read-only operations that used local worker processes (i.e. restore) were creating write locks that could interfere with parallel archive-push. (Reported by Jens Wilke.)
* Simplify locking scheme. Now, only the master process will hold write locks (archive-push, backup) and not all the local and remote worker processes as before.
The stanza-upgrade command provides a mechanism for upgrading a stanza after upgrading to a new major version of PostgreSQL.
Contributed by Cynthia Shang.
Bug Fixes:
* Fixed an issue where tables over 1GB would report page checksum warnings after the first segment. (Reported by Stephen Frost.)
* Fixed an issue where databases created with a non-default tablespace would raise bogus warnings about pg_filenode.map and pg_internal.init not being page aligned. (Reported by blogh.)
Refactoring:
* Improved the code and tests for fileManifest() to prevent a possible race condition when files are removed by the database while the manifest is being built.
* Automated builds of Debian packages for all supported distributions.
* Added --dev option to aggregate commonly used dev options.
* Added --no-package option to skip package builds.
* C library and packages are built by default, added -smart option to rebuild only when file changes are detected.
* The --libc-only option has been changed to --build-only now that packages builds have been added.
Bug Fixes:
* Fixed a regression introduced in v1.13 that could cause backups to fail if files were removed (e.g. tables dropped) while the manifest was being built. (Reported by Navid Golpayegani.)
Refactoring:
* Refactor FileCommon::fileManifest() and FileCommon::fileStat to be more modular to allow complete branch/statement level coverage testing.
Bug Fixes:
* Fixed an issue where an archive-push error would not be retried and would instead return errors to PostgreSQL indefinitely (unless the .error file was manually deleted). (Reported by Jens Wilke.)
* Fixed a race condition in parallel archiving where creation of new paths generated an error when multiple processes attempted to do so at the same time. (Reported by Jens Wilke.)
Refactoring:
* Improved performance of wal archive min/max provided by the info command. (Suggested by Jens Wilke.)
* Documentation can now be built with reusable blocks to reduce duplication.
* Added ability to pass options to containers within the documentation.
* Add proper tag to slightly emphasize proper nouns.
IMPORTANT NOTE: The new implementation of asynchronous archiving no longer copies WAL to a separate queue. If there is any WAL left over in the old queue after upgrading to 1.13, it will be abandoned and not pushed to the repository.
To prevent this outcome, stop archiving by setting archive_command = false. Next, drain the async queue by running pgbackrest --stanza=[stanza-name] archive-push and wait for the process to complete. Check that the queue in [spool-path]/archive/[stanza-name]/out is empty. Finally, install 1.13 and restore the original archive_command.
IMPORTANT NOTE: The stanza-create command is not longer optional and must be executed before backup or archiving can be performed on a new stanza. Pre-existing stanzas do not require stanza-create to be executed.
Bug Fixes:
* Fixed const assignment giving compiler warning in C library. (Fixed by Adrian Vondendriesch.)
* Fixed a few directory syncs that were missed for the --repo-sync option.
* Fixed an issue where a missing user/group on restore could cause an "uninitialized value" error in File->owner(). (Reported by Leonardo Avellar.)
* Fixed an issue where protocol mismatch errors did not output the expected value.
* Fixed a spurious archive-get log message that indicated an exit code of 1 was an abnormal termination.
Features:
* Improved, multi-process implementation of asynchronous archiving.
* Improved stanza-create command so that it can repair broken repositories in most cases and is robust enough to be made mandatory. (Contributed by Cynthia Shang.)
* Improved check command to run on a standby, though only basic checks are done because pg_switch_xlog() cannot be executed on a replica. (Contributed by Cynthia Shang.)
* Added archive and backup WAL ranges to the info command.
* Added warning to update pg_tablespace.spclocation when remapping tablespaces in PostgreSQL < 9.2. (Contributed by blogh.)
* Remove remote lock requirements for the archive-get, restore, info, and check commands since they are read-only operations. (Suggested by Michael Vitale.)
Refactoring:
* Refactor File and BackupCommon modules to improve test coverage.
* Return proper error code when unable to convert a relative path to an absolute path. (Suggested by Yogesh Sharma.)
* Log file banner is not output until the first log entry is written. (Suggested by Jens Wilke.)
* Moved File->manifest() into the FileCommon.pm module.
* Moved the Archive modules to the Archive directory and split the archive-get and archive-push commands into separate modules.
* Split the check command out of the Archive.pm module.
* Allow logging to be suppressed via logDisable() and logEnable().
* Allow for locks to be taken more than once in the same process without error.
* Lock directories can be created when more than one directory level is required.
* Clean up optionValid()/optionTest() logic in Lock.pm.
* Added Exception::exceptionCode() and Exception::exceptionMessage() to simplify error handling logic.
* Represent .gz extension with a constant.
* Allow empty files to be created with FileCommon::fileStringWrite() and use temp files to avoid partial reads.
* Refactor process IO and process master/minion code out from the common protocol code.
* Reduced the likelihood of torn pages causing a false positive in page checksums by filtering on start backup LSN.
* Remove Intel-specific optimization from C library build flags. (Contributed by Adrian Vondendriesch.)
* Removed --lock option. This option was introduced before the lock directory could be located outside the repository and is now obsolete.
* Added --log-timestamp option to allow timestamps to be suppressed in logging. This is primarily used to avoid filters in the automated documentation.
* Fixed alignment issues with multiline logging.
