Similar issue, however we're running wyseP20
Another log to look at is C:\ProgramData\VMware\VDM\logs\pcoip_server_TIMESTAMP.txt
From debug-TIMESTAMP.txt found a string "[wsnm_desktop] Console session needs to be disconnected" at .373
filtering the pcoip_server log for close to that time frame shows what it seems the reason for the disconnection is a ping latency?
07/30/2013, 18:18:56.274> LVL:1 RC: -504 MGT_PCOIP_DATA : Unable to communicate with peer on PCoIP media channels (data manager ping timer expired)
Modifications done before finding the above:
Modified the pool to use 1 display and the lowest resolution available (1680x1050)
Recently added new host and its Date/Time and NTP were not configured; Sync the same as other hosts now.
Upgraded VconServer VM to 4GRam 4Cores from .5G 1Core... yeah I know
I believe this is where the cause is, logged in and noticed 80% usage where ws_TunnelSErvice was taking up 200M and there were pages being done...maybe just didn't have enough resources.
Keeping observation and running a day test to see if the issue persists.
Hope this information is useful for someone!