https://tracker.ceph.com/https://tracker.ceph.com/favicon.ico2016-05-17T18:44:54ZCeph Ceph - Bug #15659: Bluestore OSD crash after high amount of writes and object movement/deletionhttps://tracker.ceph.com/issues/15659?journal_id=709422016-05-17T18:44:54ZEvgeniy Firsovevgeniy.firsov@sandisk.com
<ul></ul><p>For me it always reproducible after using the following test case:<br /><pre>
./vstart.sh -n -k --bluestore -l -X --osd_num 1 --mon_num 1 mon osd
./rbd create --size $((7168 * 1024)) test
fio/fio --ioengine=rbd --clientname=admin --pool=rbd --rbdname=test --thread --output=out.fio.txt --direct=1 --rw=write -refill_buffers --randrepeat=0 --bs=1M --rwmixread=0 -iodepth=64 --group_reporting --name=4ktest --fill_device=1 --filesize=6T #load
fio/fio --ioengine=rbd --clientname=admin --pool=rbd --rbdname=test --thread --output=out.fio.txt --direct=1 --rw=randrw -refill_buffers --randrepeat=0 --bs=4k --rwmixread=0 -iodepth=64 --group_reporting --name=4ktest --thread --numjobs=$THREAD --runtime=86400 --filesize=6T --time_based --status-interval=10 # 100%write
</pre></p> Ceph - Bug #15659: Bluestore OSD crash after high amount of writes and object movement/deletionhttps://tracker.ceph.com/issues/15659?journal_id=709542016-05-17T21:25:38ZSage Weilsage@newdream.net
<ul><li><strong>Priority</strong> changed from <i>Normal</i> to <i>High</i></li></ul> Ceph - Bug #15659: Bluestore OSD crash after high amount of writes and object movement/deletionhttps://tracker.ceph.com/issues/15659?journal_id=891602017-04-12T21:16:31ZSage Weilsage@newdream.net
<ul><li><strong>Status</strong> changed from <i>New</i> to <i>Won't Fix</i></li></ul><p>closing this as it was the jewel prototype</p>