ESXI 6.5 host increase memory and not getting stable even no activity in VM client

[Rickyboy2018]

Hi I'm using ESXI 6.5 update 01 in Dell power edge 720, with a memory of 127 GB and with 3 VM running, one of my VM is running apache server and other was only File server,

The problem is that if there is no activity in each Guest, but the Main Host does not neutralised, and it keeps increasing everytime there is an activity, i already check each VM if the vmtool is running using ESXtop and it seems everything is ok, i already disable Delayed ACK, and it seems it didnt work, any advice would gratefully help.Thanks

                  Memory

VM1   -      20 GB   Win 7

VM2   -      10 GB   Win2012

VM3-         15 GB   Win2008

[hussainbte]

is this host connected to vCenter? are you facing any kind of connectivity issues with vCenter or other VMs?

could you also share output for esxtop .. a screenshot for the memory information for now.

[Rickyboy2018]

   No.  this is not yet connected to VCenter, Once it reach 50% memory utilization it will start to get slow and cannot access the host itself, so i force to lessen the memory per VM.  for your reference see attached.

[hussainbte]

Could you share the memory allocation setting page for each VM.

make sure the unlimited allocation is checked. this makes sure the VM can use all that has been allocated.

with host having 128 GB, and total for all VMs is less than 50GB, ideally there should be no ballooning.. but it seems like there is memory Ballooning happening.

check this KB

VMware Knowledge Base

[khaliqamar]

Is there no memory reservation?

no hardware errors ?

vmtools are working or compatible  .. recheck

mem ballowing Status .. check

[Rickyboy2018]

Could you share the memory allocation setting page for each VM.?

Can you guide me on how to do this?

make sure the unlimited allocation is checked. this makes sure the VM can use all that has been allocated.

This is my default in ESXI6.5, dont know how to calculate the memory i need.

with host having 128 GB, and total for all VMs is less than 50GB, ideally there should be no ballooning.. but it seems like there is memory Ballooning happening.

this is my sched.mem.max and i don't know where to start the configuration, can you guide me on this to prevent memory Ballooning.

[Rickyboy2018]

2018-02-11T01:31:38.721Z cpu5:66248)WARNING: ScsiDeviceIO: 1188: Device naa.6848f690e9fdb00021ae370b16e736b4 performance has deteriorated. I/O latency increased from average value of 1892 microseconds to 57482 microseconds.

I found this in my warning logs and read some notes about if, still not sure where to start with this, i'm planning to update the drivers if this will solve this issue.

[hussainbte]

What troubleshooting has been done so far..? have you rebooted the VMs ..? have rebooted the ESXi also..?

[Rickyboy2018]

So far the only remedy is to reboot the VM then it will neutralized, but once there is a higher activity in each VM then is will rise again and does not normalized or the memory retain in the ESXi host, for my Hardware error it resolve but rebooting the ESXI host. Can you teach me on how to do the reservation of memory if really needed to avoid high memory usage in ESXI host.

[hussainbte]

another option to isolate this is to stop memory ballooning ..

can be done in 2 ways.

1) configure memory reservations for the 3 VMs..

Just reserve whatever memory has been allocated for all three VMs.

2) reinstall VM tools on the VMs..

I have not found any known issues on the internet so far.. so one option is to uninstall vmtools from the vms completely and reinstall them.

if there is an issues with the ballooning it might get addressed.

You will need a VM reboot in either of the cases