diff options
author | Jonathan Brassow <jbrassow@redhat.com> | 2014-02-04 07:03:52 -0600 |
---|---|---|
committer | Jonathan Brassow <jbrassow@redhat.com> | 2014-02-04 07:03:52 -0600 |
commit | 324781953178ed9787257334a73d93e731add0c4 (patch) | |
tree | 0e2c0c06e62fce79ae1d35b68a6bc207a8ca0ab9 /conf/example.conf.in | |
parent | 131383963ff7cc9b101b9a80a8e495473ccec4cf (diff) | |
download | lvm2-324781953178ed9787257334a73d93e731add0c4.tar.gz |
pool: Make another thin pool fn generic for cache usage also
Make '_recalculate_thin_pool_chunk_size_with_dev_hints' so it can
be used for cache and thin pools.
Diffstat (limited to 'conf/example.conf.in')
-rw-r--r-- | conf/example.conf.in | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/conf/example.conf.in b/conf/example.conf.in index 7b98b4b8e..08ce87707 100644 --- a/conf/example.conf.in +++ b/conf/example.conf.in @@ -305,6 +305,17 @@ allocation { # placed on different PVs from the cache_pool data. cache_pool_metadata_require_separate_pvs = 0 + # Specify the minimal chunk size (in kiB) for cache pool volumes. + # Using a chunk_size that is too large can result in wasteful use of + # the cache, where small reads and writes can cause large sections of + # an LV to be mapped into the cache. However, choosing a chunk_size + # that is too small can result in more overhead trying to manage the + # numerous chunks that become mapped into the cache. The former is + # more of a problem than the latter in most cases, so we default to + # a value that is on the smaller end of the spectrum. Supported values + # range from 32(kiB) to 1048576 in multiples of 32. + # cache_pool_chunk_size = 64 + # Set to 1 to guarantee that thin pool metadata will always # be placed on different PVs from the pool data. thin_pool_metadata_require_separate_pvs = 0 |