Verifique o uso de memória vmmemctl. Eu tive um problema semelhante, consulte RedHat Linux: paginação do servidor, soma de buffers RES / RSS + em cache < TOTAL. Quem está usando minha memória?
No meu caso, tínhamos um servidor de 8 GB de RAM e não conseguimos encontrar qual processo estava sendo usado. Este é o nosso vmmemctl:
cat /proc/vmmemctl
target: 1000894 pages
current: 1000894 pages
rateNoSleepAlloc: 16384 pages/sec
rateSleepAlloc: 2048 pages/sec
rateFree: 16384 pages/sec
timer: 325664
start: 3 ( 0 failed)
guestType: 3 ( 0 failed)
lock: 3623088 ( 29 failed)
unlock: 623698 ( 0 failed)
target: 325664 ( 2 failed)
primNoSleepAlloc: 3620199 ( 11 failed)
primCanSleepAlloc: 2900 ( 0 failed)
primFree: 2622165
errAlloc: 28
errFree: 28
getconf PAGESIZE
4096
Então vmmemctl está usando 4GBs
É uma pena que o vmmemctl não use um método padrão para relatar a quantidade de memória que está usando, mas acho que é porque foi implementado.
A principal referência do vmware oferece muitos detalhes sobre o balonismo. Cito já que é relevante para o nosso problema original ('porque é que esta paginação do servidor se tem memória não usada'?):"Typically, the hypervisor inflates the virtual machine balloon when it is under memory pressure. By inflating the balloon, a virtual machine consumes less physical memory on the host, but more physical memory inside the guest. As a result, the hypervisor offloads some of its memory overload to the guest operating system while slightly loading the virtual machine. That is, the hypervisor transfers the memory pressure from the host to the virtual machine. Ballooning induces guest memory pressure. In response, the balloon driver allocates and pins guest physical memory. The guest operating system determines if it needs to page out guest physical memory to satisfy the balloon driver’s allocation requests. If the virtual machine has plenty of free guest physical memory, inflating the balloon will induce no paging and will not impact guest performance. In this case, as illustrated in Figure 6, the balloon driver allocates the free guest physical memory from the guest free list. Hence, guest-level paging is not necessary.
However, if the guest is already under memory pressure, the guest operating system decides which guest physical pages to be paged out to the virtual swap device in order to satisfy the balloon driver’s allocation requests. The genius of ballooning is that it allows the guest operating system to intelligently make the hard decision about which pages to be paged out without the hypervisor’s involvement."
"genius of ballooning" :)