Hi, On 05/10/2017 12:29 PM, zengran zhang wrote: > hi cephers: > recently,we found some pgs incomplete in our bluestore+ec > overwrite environment, we think it > is because the pg's acting set size below the min_size when some osd > failure out. we wonder know how to reslove the crush problem in EC > environment. thanks! > > PG_STAT OBJECTS MISSING_ON_PRIMARY DEGRADED MISPLACED UNFOUND BYTES > LOG DISK_LOG STATE STATE_STAMP VERSION REPORTED > UP UP_PRIMARY ACTING > ACTING_PRIMARY LAST_SCRUB SCRUB_STAMP LAST_DEEP_SCRUB > DEEP_SCRUB_STAMP > 4.4c 0 0 0 0 0 0 > 0 0 incomplete 2017-05-10 15:51:16.592083 0'0 966:3 > [2147483647,6,8,2147483647,1,9] 5 > [2147483647,6,8,2147483647,1,9] 5 829'6744 2017-05-10 > 04:50:07.114874 725'4190 2017-05-06 05:47:25.286826 > 4.67 0 0 0 0 0 0 > 3000 3000 incomplete 2017-05-10 15:51:16.512486 831'5439 966:4 > [6,2147483647,2147483647,5,8,4] 6 > [6,2147483647,2147483647,5,8,4] 6 825'5292 2017-05-09 > 16:53:26.029346 696'5082 2017-05-04 00:28:50.459986 > It is indeed likely because there are not enough OSDs or hosts (depending on the crush rule you're using) to provide the required (6) OSDs for each PG. Cheers > regards! > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- Loïc Dachary, Artisan Logiciel Libre -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
