From: Sage Weil Date: Thu, 14 Nov 2019 15:37:44 +0000 (-0600) Subject: mgr/pg_autoscaler: default to pg_num[_min] = 16 X-Git-Tag: v14.2.8~20^2~15^2 X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=e5135ee5b0143870f7f91069192b5faa91285b58;p=ceph.git mgr/pg_autoscaler: default to pg_num[_min] = 16 4 or 8 PGs doesn't provide much parallelism at baseline. Start with 16 and set the floor there; that's a more reasonable number of OSDs that will be put to work on a single pool. Note that there is no magic number here. At some point someone has to tell Ceph if an empty pool should get lots of PGs across lots of devices to get the full throughput of the cluster. But this will be a bit less painful/surprising for users. Fixes: https://tracker.ceph.com/issues/42509 Signed-off-by: Sage Weil (cherry picked from commit 78bf92448002eece7501da01b67f900a84207e70) --- diff --git a/doc/rados/configuration/pool-pg-config-ref.rst b/doc/rados/configuration/pool-pg-config-ref.rst index aaead074672c..d006f540b713 100644 --- a/doc/rados/configuration/pool-pg-config-ref.rst +++ b/doc/rados/configuration/pool-pg-config-ref.rst @@ -207,7 +207,7 @@ Ceph configuration file. value is the same as ``pg_num`` with ``mkpool``. :Type: 32-bit Integer -:Default: ``8`` +:Default: ``16`` ``osd pool default pgp num`` diff --git a/src/common/options.cc b/src/common/options.cc index cc10924dc511..1a2c8870c94f 100644 --- a/src/common/options.cc +++ b/src/common/options.cc @@ -2565,7 +2565,7 @@ std::vector