From: Peter Eisentraut Date: Wed, 16 Oct 2002 22:06:33 +0000 (+0000) Subject: Put the disk information all in one chapter. X-Git-Tag: REL7_3~273 X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=841e4b20895630a41762e7b282be80af44967639;p=postgresql Put the disk information all in one chapter. --- diff --git a/doc/src/sgml/admin.sgml b/doc/src/sgml/admin.sgml index 75227bf182..5917434506 100644 --- a/doc/src/sgml/admin.sgml +++ b/doc/src/sgml/admin.sgml @@ -1,5 +1,5 @@ @@ -34,7 +34,6 @@ $Header: /cvsroot/pgsql/doc/src/sgml/Attic/admin.sgml,v 1.38 2002/09/25 21:16:10 &monitoring; &diskusage; &wal; - &recovery; ®ress; &release; diff --git a/doc/src/sgml/diskusage.sgml b/doc/src/sgml/diskusage.sgml index d7e5cb2794..96507dd093 100644 --- a/doc/src/sgml/diskusage.sgml +++ b/doc/src/sgml/diskusage.sgml @@ -1,12 +1,20 @@ Monitoring Disk Usage + + This chapter discusses how to monitor the disk usage of a PostgreSQL + database system. In the current release, the database administrator + does not have much control over the on-disk storage layout, so this + chapter is mostly informative and can give you some ideas how to + manage the disk usage with operating system tools. + + - Monitoring Disk Usage + Determining Disk Usage disk usage @@ -96,6 +104,40 @@ play-# ORDER BY relpages DESC; shows disk usage for each database. + + + Disk Full Failure + + + The most important disk monitoring task of a database administrator + is to make sure the disk doesn't grow full. A filled data disk may + result in subsequent corruption of database indexes, but not of the + fundamental data tables. If the WAL files are on the same disk (as + is the case for a default configuration) then a filled disk during + database initialization may result in corrupted or incomplete WAL + files. This failure condition is detected and the database server + will refuse to start up. + + + + If you cannot free up additional space on the disk by deleting + other things you can move some of the database files to other file + systems and create a symlink from the original location. But + note that pg_dump cannot save the location layout + information of such a setup; a restore would put everything back in + one place. To avoid running out of disk space, you can place the + WAL files or individual databases in other locations while creating + them. See the initdb documentation and for more information. + + + + + Some file systems perform badly when they are almost full, so do + not wait until the disk is full to take action. + + + + @@ -49,7 +49,6 @@ - diff --git a/doc/src/sgml/recovery.sgml b/doc/src/sgml/recovery.sgml deleted file mode 100644 index aa10b8cf64..0000000000 --- a/doc/src/sgml/recovery.sgml +++ /dev/null @@ -1,104 +0,0 @@ - - Database Failures - - - Database failures (or the possibility of such) must be assumed to be - lurking, ready to strike at some time in the future. A prudent - database administrator will plan for the inevitability of failures - of all possible kinds, and will have appropriate plans and - procedures in place before the failure occurs. - - - - Database recovery is necessary in the event of hardware or software - failure. There are several categories of failures; some of these - require relatively minor adjustments to the database, while others - may depend on the existence of previously prepared database dumps - and other recovery data sets. It should be emphasized that if your - data is important and/or difficult to regenerate, then you should - have considered and prepared for various failure scenarios. - - - - Disk Filled - - - A filled data disk may result in subsequent corruption of database - indexes, but not of the fundamental data tables. If the WAL files - are on the same disk (as is the case for a default configuration) - then a filled disk during database initialization may result in - corrupted or incomplete WAL files. This failure condition is - detected and the database will refuse to start up. You must free - up additional space on the disk (or move the WAL area to another - disk; see ) and then restart the - postmaster to recover from this condition. - - - - - Disk Failed - - - Failure of any disk (or of a logical storage device such as a RAID - subsystem) involved with an active database will require - that the database be recovered from a previously prepared database - dump. This dump must be prepared using - pg_dumpall, and updates to the database - occurring after the database installation was dumped will be lost. - - - - - - - -