On NVMe we can write really fast... fast enough that we may fill up before
we realize we are nearly full. It's a problem here only because our
devices are so small that the OSDs fill up proportionally quickly.
Signed-off-by: Sage Weil <sage@redhat.com>
debug rocksdb: 10
bluestore compression mode: aggressive
bluestore fsck on mount: true
+ # lower the full ratios since we can fill up a 100gb osd so quickly
+ mon osd full ratio: .9
+ mon osd backfillfull_ratio: .85
+ mon osd nearfull ratio: .8
+ osd failsafe full ratio: .95
# this doesn't work with failures bc the log writes are not atomic across the two backends
# bluestore bluefs env mirror: true
debug bluefs: 20
debug rocksdb: 10
bluestore fsck on mount: true
-
+ # lower the full ratios since we can fill up a 100gb osd so quickly
+ mon osd full ratio: .9
+ mon osd backfillfull_ratio: .85
+ mon osd nearfull ratio: .8
+ osd failsafe full ratio: .95
# this doesn't work with failures bc the log writes are not atomic across the two backends
# bluestore bluefs env mirror: true