asterisk users - Aug 2014 - load testing and pattern testing sangoma A116 card

Hi,

I am trying to validate a setup with a sangoma A116 card (16PRI in one
card). I currently have two machines set up, each with a sangoma A116
card. Those are interconnected with crossed PRI cables. One of them is
in NT mode, the other in TE.

The ports are configured in E1 mode, asterisk is all set up and I can
make calls between them, filling up all 480 (16 * 30) channels.

Now, I learned about the pattest and patgen tools that come with
dahdi. As far as I understood I should just run patgen on a channel on
the first machine, and then run pattest on the same channel on the
second machine (assuming port 1 on machine 1 is connected to port 1 on
machine 2) . So:
patgen /dev/dahdi/1 #on machine 1
pattest /dev/dahdi/1 # on machine 2

This however gives a never ending flood of output on the pattest on
machine 2. Am I correct in assuming that this is not a good thing?

Does anyone have any other recommendations on how to validate this setup?

I am running all this on ubuntu lucid, kernel 2.6.32.
Asterisk version 11.5.0
Dahdi 2.6.1
Wanpipe 7.0.3

Here's a few config snippets:
wanpipe1.conf (the other wanpipe.conf files are similar)
[devices]
wanpipe1 = WAN_AFT_TE1, Comment

[interfaces]
w1g1 = wanpipe1, , TDM_VOICE, Comment

[wanpipe1]
CARD_TYPE       = AFT
S514CPU         = A
CommPort        = PRI
AUTO_PCISLOT    = NO
PCISLOT         = 4
PCIBUS          = 5
FE_MEDIA        = E1
FE_LCODE        = HDB3
FE_FRAME        = NCRC4
FE_LINE         = 1
TE_CLOCK        = NORMAL
TE_REF_CLOCK    = 0
TE_SIG_MODE     = CCS
TE_HIGHIMPEDANCE        = NO
TE_RX_SLEVEL    =  430
LBO             =  120OH
FE_TXTRISTATE   = NO
MTU             = 1500
UDPPORT         = 9000
TTL             = 255
IGNORE_FRONT_END        = NO
TDMV_SPAN               = 1
TDMV_DCHAN              = 0
TDMV_HW_DTMF            = NO            # YES: receive dtmf events from hardware
TDMV_HW_FAX_DETECT      = NO            # YES: receive fax 1100hz
events from hardware

[w1g1]
ACTIVE_CH       = ALL
TDMV_HWEC       = YES
MTU             = 8

/etc/dahdi/system.conf:
defaultzone=be
loadzone=be

span=1,1,0,ccs,hdb3
dchan=16
bchan=1-15,17-31
echocanceller=mg2,1-15,17-31

span=2,2,0,ccs,hdb3
dchan=47
bchan=32-46,48-62
echocanceller=mg2,32-46,48-62

span=3,3,0,ccs,hdb3
dchan=78
bchan=63-77,79-93
echocanceller=mg2,63-77,79-93

span=4,4,0,ccs,hdb3
dchan=109
bchan=94-108,110-124
echocanceller=mg2,94-108,110-124

span=5,5,0,ccs,hdb3
dchan=140
bchan=125-139,141-155
echocanceller=mg2,125-139,141-155

span=6,6,0,ccs,hdb3
dchan=171
bchan=156-170,172-186
echocanceller=mg2,156-170,172-186

span=7,7,0,ccs,hdb3
dchan=202
bchan=187-201,203-217
echocanceller=mg2,187-201,203-217

span=8,8,0,ccs,hdb3
dchan=233
bchan=218-232,234-248
echocanceller=mg2,218-232,234-248

span=9,9,0,ccs,hdb3
dchan=264
bchan=249-263,265-279
echocanceller=mg2,249-263,265-279

span=10,10,0,ccs,hdb3
dchan=295
bchan=280-294,296-310
echocanceller=mg2,280-294,296-310

span=11,11,0,ccs,hdb3
dchan=326
bchan=311-325,327-341
echocanceller=mg2,311-325,327-341

span=12,12,0,ccs,hdb3
dchan=357
bchan=342-356,358-372
echocanceller=mg2,342-356,358-372

span=13,13,0,ccs,hdb3
dchan=388
bchan=373-387,389-403
echocanceller=mg2,373-387,389-403

span=14,14,0,ccs,hdb3
dchan=419
bchan=404-418,420-434
echocanceller=mg2,404-418,420-434

span=15,15,0,ccs,hdb3
dchan=450
bchan=435-449,451-465
echocanceller=mg2,435-449,451-465

span=16,16,0,ccs,hdb3
dchan=481
bchan=466-480,482-496
echocanceller=mg2,466-480,482-496

/etc/asterisk/chan_dahdi.conf:
switchtype=EuroISDN
context=NoRestrict
group=0
echocancel=yes
echocancelwhenbridged=no
echotraining=no
pridialplan=unknown
prilocaldialplan=unknown
usecallerid=yes
usecallingpres=yes
signalling=pri_cpe
channel=1-15,17-46,48-77,79-108,110-139,141-170,172-201,203-232,234-263,265-294,296-325,327-356,358-387,389-418,420-449,451-480,482-496

Cheers,

Frederic

> This however gives a never ending flood of output on the pattest on
> machine 2. Am I correct in assuming that this is not a good thing?
The errors I was seeing were being caused by the echo cancellation
modules on the A116 card kicking in. All I had to do was disable the
module before starting the tests like this for all interfaces:
wan_ec_client wanpipe1 mpd all
> Does anyone have any other recommendations on how to validate this setup?
Sangoma has a wiki page that helped me in finding out why my test was
failing: http://wiki.sangoma.com/Wanpipe-Self-Test
There are also additional tests that can be done documented on that
page. May be useful to someone at some point.