Hi Maish,
- Is this happenning to all the VM's or only some (perhaps one(/more)specific host?We package the guest OS and application into an OVA for distribution.And this is happening on all the deployed VMs. Not isolated to 1 or 2.
- What are the resources allocated to the VM - CPU / RAM?
4 vCPUs and 12 GB RAM - Are there any limits assigned to the VM's?
No Limits, only a reservation of 10GB of RAM for each VM - Is the Host oversubscribed?
No, the host has 12 physical cores and we are maintaining a 1:1 ratio of core to vCPU.
Also output from the esxtop for the host with the VM will be helpful.
esxtop also suggests that the load on the CPU is around 60%, which is what we expect.
Below is the output.
~ # esxtop
5:26:06pm up 13 days 22:54, 318 worlds, 3 VMs, 12 vCPUs; CPU load average: 0.65, 0.65, 0.65
PCPU USED(%): 47 39 42 40 42 40 43 38 41 38 39 41 27 0.3 27 1.0 0.1 35 0.5 0.4 0.1 67 48 0.2 AVG: 29
PCPU UTIL(%): 57 49 52 50 50 47 50 45 45 42 44 45 26 0.5 25 0.9 0.1 34 0.7 0.6 0.2 64 46 0.5 AVG: 32
CORE UTIL(%): 79 77 77 76 73 74 27 26 34 1.1 64 46 AVG: 55
~ # esxtop
5:27:19pm up 13 days 22:55, 318 worlds, 3 VMs, 12 vCPUs; CPU load average: 0.64, 0.65, 0.65
GID VMNAME VDEVNAME NVDISK CMDS/s READS/s WRITES/s MBREAD/s MBWRTN/s LAT/rd LAT/wr
1486 VSM7-MS001 - 5 151.18 0.40 150.78 0.00 37.16 185.45 91.41
1515 VSM7-MS003 - 3 185.51 7.83 177.68 0.05 33.23 107.30 73.05
1516 VSM7-MS002 - 5 165.64 0.20 165.44 0.00 35.19 660.02 82.52