There is a known issue affecting ESXi 5.x running on HP hardware with one of the following HP-AMS (HP Agentless Management Service) versions:
- hp-ams 500.9.6.0-12.434156
- hp-ams 500.10.0.0-18.434156
If you run into this bug, you may notice various symptoms (one or more) like:
- it is not possible to perform a vMotion to and from the ESXi host
- error “VMK_NO_MEMORY” when you try to power on a VM
- error “A general system error occurred: command /bin/sh failed”
- error message “can’t fork” when using the console
- message: “WARNING: Heap: 2677: Heap globalCartel-1 already at its maximum size. Cannot expand.”
- ESXi host disconnects from vCenter
- ESXi host unmanageable
As written above, this is a known issue. To resolve, just upgrade your hp-ams to version 10.0.1 using the latest HP Agentless Management Offline Bundle:
If you already have the problem and your host is unmanageable or you cannot perform a vMotion, you will have to power down all VMs on the host.
Then reboot the host and upgrade to the latest version immediately.
You want to read more about this issue – check this out:
Advisory: (Revision) HP ProLiant Servers – HP Agentless Management Service (hp-ams) Version 9.6 or 10.0 Cannot Initiate VMotion or Enable Services and May Report a HEAP Global Error When Attempting to Acquire Memory