Hi, Here is the result from host: mount | grep /dev/shm => tmpfs on /dev/shm type tmpfs (rw,nosuid,nodev) du -hs /dev/shm => 0 /dev/shm df /dev/shm => Filesystem 1K-blocks Used Available Use% Mounted on tmpfs 2023252 0 2023252 0% /dev/shm ---------------------------------------------------------------------- And here is the result from postgres container: mount | grep /dev/shm => shm on /dev/shm type tmpfs (rw,nosuid,nodev,noexec,relatime,size=65536k) du -hs /dev/shm => 8.0K /dev/shm df /dev/shm => Filesystem 1K-blocks Used Available Use% Mounted on shm 65536 8 65528 1% /dev/shm On Wed, Jan 3, 2018 at 12:05 PM, Thomas Munro <thomas.munro@xxxxxxxxxxxxxxxx> wrote: > On Wed, Jan 3, 2018 at 5:39 PM, Tom Lane <tgl@xxxxxxxxxxxxx> wrote: >> Thomas Munro <thomas.munro@xxxxxxxxxxxxxxxx> writes: >>> So you have 16GB of RAM and here we're failing to posix_fallocate() >>> 50MB (actually we can't tell if it's the ftruncate() or >>> posix_fallocate() call that failed, but the latter seems more likely >>> since the former just creates a big hole in the underlying tmpfs >>> file). Can you share the query plan (EXPLAIN SELECT ...)? >> >> I wonder if OP is running with a tmpfs size setting that's less than >> the traditional Linux default of half of physical RAM size. > > Hmm. Canh, can you please share the output of the following commands? > > mount | grep /dev/shm > du -hs /dev/shm > df /dev/shm > > -- > Thomas Munro > http://www.enterprisedb.com
