Hi, I''m seeing messages like this [ 3194.928153] btrfs allocation failed flags 1, wanted 65536 [ 3194.934874] space_info 1 has 147456 free, is full [ 3194.941205] space_info total=1903427584, used=1903280128, pinned=0, reserved=0, may_use=65536, readonly=0 [ 3194.941209] block group 12582912 has 8388608 bytes, 8376320 used 0 pinned 0 reserved [ 3194.941212] entry offset 20959232, bytes 12288, bitmap no [ 3194.941213] block group has cluster?: no [ 3194.941215] 0 blocks of free space at or bigger than bytes is [ 3194.941218] block group 136708096 has 218103808 bytes, 218095616 used 0 pinned 0 reserved [ 3194.941221] entry offset 354803712, bytes 8192, bitmap no [ 3194.941222] block group has cluster?: no mkfs is default, mount with noatime, happens consistently at xfstests/275 (on first run of the whole testsuite) the test uses a 2g scratch partition. Has been observed for like 2 weeks when testing next/master. The test description says # The posix write test. When write size is larger than disk free size, # should write as much as possible until ENOSPC. so enospc messages can be expected, although I don''t know if they''re just debugging or point to real bugs. Full log attached. david