| Current Path : /proc/2/root/usr/share/doc/db4-devel-4.7.25/ref/changelog/ |
| Current File : //proc/2/root/usr/share/doc/db4-devel-4.7.25/ref/changelog/4.5.20.html |
<!--$Id: 4.5.20.html,v 1.25 2007/05/17 18:17:18 bostic Exp $-->
<html>
<head>
<title>The Berkeley DB Package: DB 4.5.20 Change Log</title>
<meta name="description" content="Berkeley DB: A database programmatic toolkit.">
<meta name="keywords" content="embedded,database,programmatic,toolkit,b+tree,btree,hash,hashing,transaction,transactions,locking,logging,access method,access methods">
</head>
<body bgcolor=white>
<h3 align=center>Berkeley DB 4.5.20 Change Log</h3>
<h3>Database or Log File On-Disk Format Changes:</h3>
<ol>
<li>The on-disk log format has changed.
</ol>
<h3>New Features:</h3>
<ol>
<li>Multi-Version Concurrency Control for the Btree/Recno access
methods.
<li>A new replication framework with a default TCP/IP setup.
<li>Online replication upgrades for high availability replicated 24/7
systems.
<li>A new event-style notification.
<li>Several enhancements to the Java Collections API including the
implementation of the size() method.
</ol>
<h3>Database Environment Changes:</h3>
<ol>
<li>Update the DB_ENV->failchk method to garbage collect per-process
mutexes stranded after unexpected process failure. [#13964]
<li>Fix a bug that could cause memory used to track threads for
DB_ENV->failchk to not be reused when a thread no longer exists.
[#14425]
<li>Add set_event_notify behavior as part of new event notification in
Berkeley DB. [#14534]
<li>Fix a bug so that we no longer panic on DB_ENV->close() if a
previous environment close failed to log. This condition will now return
an error. [#14693]
<li>Created os_getenv, removed clib/getenv, implemented Windows specific
behavior. [#14942]
<li>Fix a bug where it was possible to corrupt the DB_REGISTER
information file, making it impossible for Berkeley DB applications to
join database environments. [#14998]
</ol>
<h3>Concurrent Data Store Changes:</h3>
<ol>
<li>Fix a bug where renaming a subdatabase in a Concurrent Data Store
environment could fail. [#14185]
</ol>
<h3>General Access Method Changes:</h3>
<ol>
<li>Fix a bug that could leave extra unallocated pages at the end of a
database file. [#14031]
<li>Optimize secondary updates when overwriting primary records.
[#14075]
<li>Fix a bug to prevent a trap when creating a named in-memory database
and there are already temporary files open. [#14133]
<li>Fix a bug which caused a trap if the key parameter to DBC->c_get was
omitted with DB_CURRENT. [#14143]
<li>Fix a bug with secondary cursors when the secondary has off-page
duplicates. This bug resulted in incorrect primary data being returned.
[#14240]
<li>Improve performance when removing a subdatabase by not locking every
page. [#14366]
<li>Fix a bug that would not properly upgrade database files from
releases 3.2.9 (and earlier) to releases 4.0 (and greater). [#14461]
<li>Fix a bug that could cause a DB_READ_UNCOMMITTED get through a
secondary index to return DB_SECONDARY_CORRUPT. [#14487]
<li>Fix a bug so that non-transactional cursor updates of a
transactional database will generate an error. [#14519]
<li>Add a message when the system panics due to a page in the wrong
state at its time of allocation. [#14527]
<li>Fix a remove failure when attempting to remove a file that is open
in another thread of control. [#14780]
<li>Fix a bug where the key was not ignored when doing a cursor put with
the DB_CURRENT flag. [#14988]
</ol>
<h3>Btree Access Method Changes:</h3>
<ol>
<li>Changed the implementation of internal nodes in btrees so that they
no longer share references to overflow pages with leaf nodes. [#10717]
<li>Fix a bug that could cause a diagnostic assertion by setting the
deleted bit on a record in an internal node. [#13944]
<li>Fix three problems in BTREE compaction: [#14238]
<ol type="a">
<li>When deleting a page don't check the next key in the parent if we
are going to delete the parent too.
<li>Need to check that the tree has not collapsed between dropping a
read lock and getting the write lock. If it has collapsed we will fetch
the root of the tree.
<li>Fix a case where we fail to lock the next page before reading it.
</ol>
<li>Fix a bug that could cause the compaction of a Btree with sorted
duplicates to fail when attempting to compact an off page duplicate tree
if a key could not fit in an internal node. [#14771]
<li>Fix a bug that causes a loop if an empty Btree was compacted.
[#14493]
</ol>
<h3>Hash Access Method Changes:</h3>
<ol>
<li>Fix a bug to allow creation of hash pages during truncate recovery.
[#14247]
</ol>
<h3>Queue Access Method Changes:</h3>
<ol>
<li>Fix a bug where reads of data items outside the range of the queue
were not kept locked to the end of the transaction, breaking
serializability. [#13719]
<li>Fix a bug that could cause corruption in queue extent files if
multiple processes tried to open the same extent at the same time.
[#14438]
<li>Improve concurrency for in-place updates in queue databases.
[#14918]
</ol>
<h3>Recno Access Method Changes:</h3>
None.
<h3>C++-specific API Changes:</h3>
<ol>
<li>C++ applications that check the error code in exceptions should note
that DbMemoryException has been changed to have the error code
DB_BUFFER_SMALL rather than ENOMEM, to match the error returned by the
C API. DbMemoryException will be thrown when a Dbt is too small to
contain data returned by Berkeley DB. When a call to malloc fails, or
some other resource is exhausted, a plain DbException will be thrown
with error code set to ENOMEM. [#13939]
</ol>
<h3>Java-specific API Changes:</h3>
<ol>
<li>Database.verify may now be called. This method is now static and
takes a DatabaseConfig parameter. [#13971]
<li>Add DB_ENV->{fileid_reset, lsn_reset} to the public API. [#14076]
</ol>
<h3>Java collections and bind API Changes:</h3>
<ol>
<li>The com.sleepycat.collections package is now fully compatible with
the Java Collections framework. [#14732]
</ol>
<h3>Tcl-specific API Changes:</h3>
<ol>
<li>Fix a conflicting variable, sysscript.tcl. [#15051]
</ol>
<h3>RPC-specific Client/Server Changes:</h3>
None.
<h3>Replication Changes:</h3>
<ol>
<li>Fix a bug when running with DEBUG_ROP or DEBUG_WOP. [#13394]
<li>Add live replication upgrade support [#13670]
<li>Fix a bug so that client databases are removed at the start of
internal initialization. [#14147]
<li>Fix a bug in replication internal initialization so that data_dir
will be handled correctly. Make internal initialization resilient to
multiple data_dir calls with the same directory. [#14489]
<li>Fix a bug in the 4.2 sync-up algorithm that could result in no open
files. [#14552]
<li>Fix a bug when clients decide to re-request. [#14642]
<li>Fix a bug where a PERM bulk buffer could have a zero LSN passed to
the application callback. [#14675]
<li>Change names of some existing replication API methods as described
in "Replication Method Naming" page of the "Upgrading Berkeley DB
Applications to Release 4.5" section of Berkeley DB Reference Guide.
[#14723]
<li>Fix a bug which could cause an election to succeed only after
waiting for the timeout to expire, even when all sites responded in a
timely manner. The bug was most easily visible in an election between 2
sites. [#14752]
<li>Fix a bug where a process could have an old file handle to a log
file. [#14797]
<li>Fix a bug where a "log_more" message could be on a log file
boundary. [#15034]
<li>Fix a bug that could cause log corruption if a database open
operation were attempted during a call to rep_start in another thread.
[#15035]
<li>Fix a bug during elections where a vote2 arrives before its vote1.
[#15055]
<li>Fix a bug to make sure we are a client if sending a REP_REREQUEST.
[#15066]
</ol>
<h3>XA Resource Manager Changes:</h3>
None.
<h3>Locking Subsystem Changes:</h3>
<ol>
<li>Fix a bug that could cause a write to hang if DB_READ_UNCOMMITTED
is enabled and it tries to reacquire a write lock. [#14919]
</ol>
<h3>Logging Subsystem Changes:</h3>
<ol>
<li>Fix a bug so that log headers are now included in the checksum. This
avoids a possible race in doing hot backups. [#11636].
<li>Add a check so that some log sequence errors are diagnosed at run
time rather than during recovery. [#13231]
<li>Fix a bug where recovery fails if there is no disk space for the
forced checkpoint that occurs at the end of processing the log. [#13986]
<li>Fix a bug which could cause a page to be missing from the end of a
database file if the page at the end of the file was freed while it
contained data and the system was restarted before the log record for
that free was flushed to disk. [#14090]
<li>Fix a bug that could cause log files to be incorrectly removed by
log_archive if it was run immediately after recovery. [#14874]
</ol>
<h3>Memory Pool Subsystem Changes:</h3>
<ol>
<li>Fix a bug that could cause corruption to the buffer pool cache if a
race condition was hit while using DB->compact. [#14360]
<li>Fix a bug where cache pages could be leaked in applications creating
temporary files for which the DB_MPOOL_NOFILE flag was set. [#14544]
</ol>
<h3>Transaction Subsystem Changes:</h3>
<ol>
<li>Fix a bug that could cause extra empty pages to appear in a database
file after recovery. [#11118]
<li>Fix a bug triggered when running recovery with a feedback function
that could cause a NULL pointer dereference. [#13834]
<li>Fix a bug where running recovery could create duplicate entries in
the data directory list. [#13884]
<li>Fix a bug to not trade locks if a write lock is already owned.
[#13917]
<li>Fix a bug that could cause traps or hangs if the DB_TXN->set_name
function is used in a multithreaded application. [#14033]
<li>Fix a bug so that a transaction can no longer be committed after it
had deadlocked. [#14037]
<li>Fix a bug that could cause a trap during recovery if multiple
operations that could remove the same extent are recovered. [#14061]
<li>Fix a bug that could cause an extent file to be deleted after the
last record in the extent was consumed but the consuming transaction was
aborted. [#14179]
<li>Fix a bug where the parent database would not use
DB_READ_UNCOMMITTED in certain cases when calling DBC->c_pget. [#14361]
<li>Fix a bug so that it is no longer possible to do a non-transactional
cursor update on a database that is opened transactionally. [#14519]
<li>Fix a bug that causes a sequence to ignore the DB_AUTO_COMMIT
settings. [#14582]
<li>Fix a bug, change txn_recover so that multiple processes will
recover prepared transactions without requiring that the first process
stay active. [#14707]
<li>Fix a bug that could cause the wrong record to be deleted if a
transaction had a cursor on a record with a pending delete and then
replaced a record that contained overflow data or replaced a record with
overflow data and that replace failed. [#14834]
</ol>
<h3>Utility Changes:</h3>
<ol>
<li>Fix a bug that caused db_verify to not check the order on leaf pages
which were the leftmost children of an internal node. [#13004]
<li>Fix a bug that caused db_hotbackup to not backup queue extent files.
[#13848]
<li>Fix a bug so that db_verify no longer reports that an unused hash
page is not fully zeroed. [#14030]
<li>Fix a bug where db_stat ignored the -f option to return "fast
statistics". [#14283]
<li>Fix a bug that prevented the db_stat utility from opening database
files with write permission so that meta data statistics would be
updated. [#14755]
<li>Fix a bug in db_hotbackup related to windows. Sub-directories are
now ignored. [#14757]
</ol>
<h3>Configuration, Documentation, Portability and Build Changes:</h3>
<ol>
<li>The Berkeley DB 4.3 and 4.4 releases disallowed using the
--with-uniquename configuration option with the C++, Java, or RPC
--enable-XXX options. The 4.5 release returns to the 4.2 release
behavior, allowing those combinations of configuration options. [#14067]
<li>Fix build issues when CONFIG_TEST is not enabled for Tcl. [#14507]
<li>There are updated build instructions for Berkeley DB PHP module on
Linux. [#14249]
<li>Use libtool's "standard" environment variable names so that you can
set "AR" to "ar -X64" for example, and modify both libtool and the
Makefile commands. Remove the install-strip target from the Makefile,
it is no longer used. [#14726]
<li>Fix a bug where, when a database is opened with the DB_THREAD flag
(the default in Java), and an operation in one thread causes the
database to be truncated (typically when the last page in the database
is freed) concurrently with a read or write in another thread, there can
be arbitrary data loss, as Windows zeros out pages from the read/write
location to the end of the file. [#15063]
</ol>
</body>
</html>
Copyright 2K16 - 2K18 Indonesian Hacker Rulez
;?>)