CPU Hog Detected by Timer on vpn slave server

One vpn master and 65 slaves on Novell BorderManager 3.9 support pack 1.
When one or more slaves are removed from the S2S configuration, few slaves abend at the same time:

Novell Open Enterprise Server, NetWare 6.5
PVER: 6.50.07

Server VPN_slave halted Wednesday,  5 November 2008  23.26.23,566 
Abend 1 on P00: Server-5.70.07-1937: CPU Hog Detected by Timer

Registers:
    CS = 0008 DS = 0010 ES = 0010 FS = 0010 GS = 0023 SS = 0010
    EAX = FBF17BA3 EBX = 983B1880 ECX = 00000282 EDX = 985DA170
    ESI = 00000000 EDI = 00000000 EBP = 00000003 ESP = 985DA124
    EIP = 00000000 FLAGS = 00000082 


Running process: VPNINF_THREAD     Process
Thread Owned by NLM: VPSLAVE.NLM
Stack pointer: 985D9EE8
OS Stack limit: 985D6740
Scheduling priority: 67371008
Wait state: 3030070  Yielded CPU
Stack: --FBF17BA3  ?
       00013D71  (LOADER.NLM|SetFlags+5)
       003057F5  (SERVER.NLM|GetCurrentClock+1B9)
       --00000282  (LOADER.NLM|KernelAddressSpace+282)
       --0000CF78  (LOADER.NLM|KernelAddressSpace+CF78)
       --00000000  (LOADER.NLM|KernelAddressSpace+0)
       --00000282  (LOADER.NLM|KernelAddressSpace+282)
       --00000207  (LOADER.NLM|KernelAddressSpace+207)
       --8307D4E7  ?
       --7A6C5051  ?
       --00000003  (LOADER.NLM|KernelAddressSpace+3)
       --7A6D0C0D  ?
       --985DA1B0  ?

Fixed on NBM39SP2 support pack

As workaround, stop the vpn services on the master previous to remove the slave. The remove the slave from iManager and after that restart th vpn services on the master. Master will reconfigure the slaves without abending.