Hi, We saw a very strange case in one of our production cluster. A couple regions cannot get their deleted rows or delete marker removed even after major compaction. However when the region triggered split (we set 100G for auto split), the deletion worked. The 100G region becomes two 10G daughter regions, and all the delete marker are gone.
Also, the same region in the slave cluster (through replication) have normal size at about 20+G. BTW, the delete marker in the regions are mostly deleteFamily if it matters. This is really weird. Anyone has any clue for this strange behavior? Thanks Tian-Ying A snippet of the HFile generated by the major compaction: : \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459808114380/DeleteFamily/vlen=0/ts=2292870047 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459808114011/DeleteFamily/vlen=0/ts=2292869794 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459805381104/DeleteFamily/vlen=0/ts=2291072240 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459805380673/DeleteFamily/vlen=0/ts=2291071997 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459802643449/DeleteFamily/vlen=0/ts=2290248886 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459802643246/DeleteFamily/vlen=0/ts=2290248786 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459799913003/DeleteFamily/vlen=0/ts=2289446916 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459797181831/DeleteFamily/vlen=0/ts=2288670451 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459794447388/DeleteFamily/vlen=0/ts=2287911443 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459791708803/DeleteFamily/vlen=0/ts=2287213792 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459788978387/DeleteFamily/vlen=0/ts=2286488738 V: K: \xA0\x00\x00L\x1A@\x1CBe\x00\x00\x08m\x03\x1A@\x10\x00?PF/d:/1459786243642/DeleteFamily/vlen=0/ts=2285778927 V:
