Over the past week, we have bundled and released new versions of 7.2, 7.3 and 7.4 to address a couple of potentially critical bugs that were recently identified.
For all three versions, these releases address the following:
* Prevent possible loss of committed transactions during crash
Due to insufficient interlocking between transaction commit and
checkpointing, it was possible for transactions committed just
before the most recent checkpoint to be lost, in whole or in part,
following a database crash and restart.Specific to 7.4, the following has also been fixed:
* Repair possible crash during concurrent btree index insertions
This patch fixes a rare case in which concurrent insertions into a
btree index could result in a server panic. No permanent damage
would result, but it's still worth a re-release. The bug does not
exist in pre-7.4 releases.Due to the seriousness of the first bug, as relates to data integrity, it is strongly recommended that administrators upgrade their database servers at their earliest convience.
None of these releases require a dump-n-reload in order to address the problems.
It should be noted that, for those looking to upgrade from a 7.4.0/7.4.1 server to 7.4.5, that there a dump-n-reload is *recommended* (but not required) to address a fix introduced between 7.4.1 and 7.4.2. Please refer to the 7.4.5 HISTORY file for information on how this can be accomplished without a dump-n-reload, as there are steps that can be followed for this.
Please refer to the following URL for a list of mirror sites where the new releases can be downloaded:
http://www.postgresql.org/mirror-ftp.html
Devrim has been working n SRPMs for v7.3.7 and v7.4.5, as well as David Fetter has made sure that v7.4.5 is available via BitTorrent:
http://bt.postgresql.org
As always, please report any bugs to [EMAIL PROTECTED]
Marc G. Fournier, Core Member PostgreSQL Global Development Group
---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
joining column's datatypes do not match
