asterisk users - May 2005 - asterisk take 99% of CPU resources

Hi, 

I've connected a two T100P  from digium with a 2 Rhino channelBank.

Everything is working as expected. but I have occasional Falls,
asterisk take 99% of CPU resources, with the following report

May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**
May 24 14:52:41 WARNING[11441]: We're Zap/1-1, not H8?F??n**

Any ideas are welcome!

Thanks in advance.
Jorge

These are my config files:

/etc/zaptel.conf

span=1,1,0,esf,b8zs
fxols=1-24
span=2,1,0,esf,b8zs
fxols=25-37

fxsls=49-52
defaultzone=us
loadzone=us

etc/asterisk/zapata.conf


[channels]
language = es
signalling = fxo_ls
usecallerid = yes
echocancel = yes
echotraining = yes
echocancelwhenbridged = yes
rxgain = 0
txgain = 0
musiconhold = Default
busydetect = yes
busycount = 4
jitterbuffers = 8
callwaiting = no
callprogress = no
threewaycalling = yes
transfer = yes
cancallforward = yes
callreturn = yes
group = 2
callgroup = 3
pickupgroup = 3
context = default
channel => 1-37


language = es
context = cotel
signalling = fxs_ls
callerid = cotel
usecallerid = yes
callwaitingcallerid = yes
echocancel = yes
echotraining = yes
echocancelwhenbridged = yes
rxgain = 0
txgain = 0
musiconhold = Default
busydetect = yes
busycount = 4
jitterbuffers = 8
relaxdtmf = yes
callwaiting = yes
usecallingpres = no
callprogress = no
threewaycalling = yes
transfer = yes
cancallforward = yes
callreturn = yes
group = 3
callgroup = 1
pickupgroup = 1
channel => 49-52








-- 
Jorge Verastegui <jorge@redcetus.com>
RedCetus