Hi,
Im running mysql-5.6 on a solaris 10 zone with 16GB of capped memory. The database is configured with innodb_buffer_pool_size of 10G, which is still less than recommended, but the moment I start my database server, my /tmp filesystem (swap) is bloated with data. Mysql does allocate all that memory, which is correct, but I dont understand why it takes everything from swap. When I install the database in the global zone the maximum amount of available swap is reduced, which is ok I guess, but the behavior in the zone is different. For a better understanding here's some output from those zones
global zone, first df without mysql, second with mysql running
# df -h /tmp/.
Filesystem size used avail capacity Mounted on
swap 51G 240K 51G 1% /tmp
# df -h /tmp/.
Filesystem size used avail capacity Mounted on
swap 42G 248K 42G 1% /tmp
local zone, first df without mysql, second with mysql running
# df -h /tmp/.
Filesystem size used avail capacity Mounted on
swap 16G 329M 16G 3% /tmp
# df -h /tmp/.
Filesystem size used avail capacity Mounted on
swap 16G 11G 5.1G 68% /tmp
Any hints or suggestions appreciated, cheers
Felix