aboutsummaryrefslogtreecommitdiff
path: root/src/backend/executor/nodeHash.c
diff options
context:
space:
mode:
authorTom Lane <tgl@sss.pgh.pa.us>2015-07-28 18:42:59 -0400
committerTom Lane <tgl@sss.pgh.pa.us>2015-07-28 18:42:59 -0400
commit0f272b8b4cf2a2217de64cb72e12530efb9c9e3b (patch)
tree419872512c0134ad32d7d6dbb4df6d471a7f09c2 /src/backend/executor/nodeHash.c
parent082d4283b0e96ad58d7a4bcc196e3bcd9584fb78 (diff)
downloadpostgresql-0f272b8b4cf2a2217de64cb72e12530efb9c9e3b.tar.gz
postgresql-0f272b8b4cf2a2217de64cb72e12530efb9c9e3b.zip
Update our documentation concerning where to create data directories.
Although initdb has long discouraged use of a filesystem mount-point directory as a PG data directory, this point was covered nowhere in the user-facing documentation. Also, with the popularity of pg_upgrade, we really need to recommend that the PG user own not only the data directory but its parent directory too. (Without a writable parent directory, operations such as "mv data data.old" fail immediately. pg_upgrade itself doesn't do that, but wrapper scripts for it often do.) Hence, adjust the "Creating a Database Cluster" section to address these points. I also took the liberty of wordsmithing the discussion of NFS a bit. These considerations aren't by any means new, so back-patch to all supported branches.
Diffstat (limited to 'src/backend/executor/nodeHash.c')
0 files changed, 0 insertions, 0 deletions