8.3 - Database corrupted after powering off the host - VMPLayer

[fferrero]

Hi,

Every time I'm shutting down the host on which the Simulator is running, the database seems to be corrupted.

mycluster::cluster*> cluster ring show
Node UnitName Epoch DB Epoch DB Trnxs Master Online
--------- -------- -------- -------- -------- --------- ---------
mycluster-01
mgmt 3 3 42 mycluster-01
master

Error: rdb_ring_info: RDB ring state query of 127.0.0.1 for vldb failed on RPC connect: clnttcp_create: RPC: Program not registered
mycluster-01
vldb - - - - -

Error: rdb_ring_info: RDB ring state query of 127.0.0.1 for vifmgr failed on RPC connect: clnttcp_create: RPC: Program not registered
mycluster-01
vifmgr - - - - -

Error: rdb_ring_info: RDB ring state query of 127.0.0.1 for bcomd failed on RPC connect: clnttcp_create: RPC: Program not registered
mycluster-01
bcomd - - - - -

Error: rdb_ring_info: RDB ring state query of 127.0.0.1 for crs failed on RPC connect: clnttcp_create: RPC: Program not registered
mycluster-01
crs - - - - -
5 entries were displayed.

I made a clean power-off and a clean suspend operations. Same errors. I have to do a full re-install from scratch.

Is it a simulator issue, a VMPlayer issue or a windows issue (the host is Windows 7-64 bits with 8GB of RAM).

I'm afraid that the virtual disks or the virtual NVRAM are not correctly flushed when halting the simulator.

The simulator release is 8.3RC1. I don't remember to have such issues with the previous releases.

[shatfield]

I've got one exibiting similar behaviour.  the management ring comes up in a 'normal' time frame.  The rest of the rings show the same error.  But they do eventualy come up.  Its maybe 20 minutes before the rest of the rings come online.  

This particular sim is running on ESX, so I think you can rule out the hypervisor.

Its also running 8.3GA, so its not isolated to the RC.