I have just finished installing a new asterisk box at my work. The box is
quite hefty, dual Zeon 2.8s with SCSI drives and 2Gb of memory. I have a 4
port Digium T1 card for channel bank and PRI access.
I activated a PRI from a local CLEC (DMS-500 based, National protocol).
This PRI is on slot 2 of the card and is set as the primary timing source.
It is ESF/B8ZS.
All the software is latest CVS HEAD as of 5/22/04.
My problem lies in random intermittent drops of calls. The entire PRI seems
to disappear, dropping all current established calls. I see occasional
printouts on an asterisk management console showing all 23 B channels
resetting with no reason given:
pbx1*CLI>
-- B-channel 1 successfully restarted on span 2
-- B-channel 2 successfully restarted on span 2
-- B-channel 3 successfully restarted on span 2
-- B-channel 4 successfully restarted on span 2
-- B-channel 5 successfully restarted on span 2
-- B-channel 6 successfully restarted on span 2
-- B-channel 7 successfully restarted on span 2
-- B-channel 8 successfully restarted on span 2
-- B-channel 9 successfully restarted on span 2
-- B-channel 10 successfully restarted on span 2
-- B-channel 11 successfully restarted on span 2
-- B-channel 12 successfully restarted on span 2
-- B-channel 13 successfully restarted on span 2
-- B-channel 14 successfully restarted on span 2
-- B-channel 15 successfully restarted on span 2
-- B-channel 16 successfully restarted on span 2
-- B-channel 17 successfully restarted on span 2
-- B-channel 18 successfully restarted on span 2
-- B-channel 19 successfully restarted on span 2
-- B-channel 20 successfully restarted on span 2
-- B-channel 21 successfully restarted on span 2
-- B-channel 22 successfully restarted on span 2
-- B-channel 23 successfully restarted on span 2
pbx1*CLI>
(This is from an asterisk console started with -rvvvv.)
Has anyone encountered similar behavior in the past? If so, what was the
resolution?
Thanks,
Tim McKee
configs:
zaptel.conf
#
# span 1 is for channel bank (24 FXS)
span=1,2,2,esf,b8zs
# span 2 is for NuVox PRI (1-23B, 24D)
span=2,1,2,esf,b8zs
# span 3 is unused
span=3,0,0,esf,b8zs
# span 4 is unused
span=4,0,0,esf,b8zs
#
#
zapata.conf
[channels]
;
; Default language
;
language=en
;
; Default context
;
context=dialnational
;
signalling=fxo_ks
use_callerid=yes
callwaiting=no
restrictcid=no
usecallingpres=yes
callwaitingcallerid=no
threewaycalling=yes
transfer=yes
cancallforward=yes
callreturn=yes
echocancel=yes
echocancelwhenbridged=no
echotraining=yes
relaxdtmf=no
rxgain=0.0
txgain=0.0
callgroup=1
pickupgroup=1
immediate=no
amaflags=billing
<channels 1-24 omitted>
;
; NuVox PRI
context=sdnpri
;
switchtype = national
pridialplan = unknown
signalling = pri_cpe
callerid=asreceived
group = 2
channel => 25-47
-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 2586 bytes
Desc: not available
Url :
http://lists.digium.com/pipermail/asterisk-users/attachments/20040523/5f81a917/winmail.bin
On Sun, 2004-05-23 at 21:52, Timothy R. McKee wrote:> My problem lies in random intermittent drops of calls. The entire PRI seems > to disappear, dropping all current established calls. I see occasional > printouts on an asterisk management console showing all 23 B channels > resetting with no reason given: > > pbx1*CLI> > -- B-channel 1 successfully restarted on span 2> (This is from an asterisk console started with -rvvvv.) > > Has anyone encountered similar behavior in the past? If so, what was the > resolution? >Is there a problem with your google access? http://tinyurl.com/2fjkb This was answered as late as last month and covered about every 4 months for people to lame to do their own simple google search. -- Steven Critchfield <critch@basesys.com>
Tim, I would double check the timing. It seems odd that you would supply clock rather than the switch, and if you get clock slips, that could certainly account for what you are seeing. Feel free to contact me off-list if you need more info or have any questions. Bruce Komito High Sierra Networks, Inc. www.servers-r-us.com (775) 284-5800 ext 115 On Sun, 23 May 2004, Timothy R. McKee wrote:> I have just finished installing a new asterisk box at my work. The box is > quite hefty, dual Zeon 2.8s with SCSI drives and 2Gb of memory. I have a 4 > port Digium T1 card for channel bank and PRI access. > > I activated a PRI from a local CLEC (DMS-500 based, National protocol). > This PRI is on slot 2 of the card and is set as the primary timing source. > It is ESF/B8ZS. > > All the software is latest CVS HEAD as of 5/22/04. > > My problem lies in random intermittent drops of calls. The entire PRI seems > to disappear, dropping all current established calls. I see occasional > printouts on an asterisk management console showing all 23 B channels > resetting with no reason given: > > pbx1*CLI> > -- B-channel 1 successfully restarted on span 2 > -- B-channel 2 successfully restarted on span 2 > -- B-channel 3 successfully restarted on span 2 > -- B-channel 4 successfully restarted on span 2 > -- B-channel 5 successfully restarted on span 2 > -- B-channel 6 successfully restarted on span 2 > -- B-channel 7 successfully restarted on span 2 > -- B-channel 8 successfully restarted on span 2 > -- B-channel 9 successfully restarted on span 2 > -- B-channel 10 successfully restarted on span 2 > -- B-channel 11 successfully restarted on span 2 > -- B-channel 12 successfully restarted on span 2 > -- B-channel 13 successfully restarted on span 2 > -- B-channel 14 successfully restarted on span 2 > -- B-channel 15 successfully restarted on span 2 > -- B-channel 16 successfully restarted on span 2 > -- B-channel 17 successfully restarted on span 2 > -- B-channel 18 successfully restarted on span 2 > -- B-channel 19 successfully restarted on span 2 > -- B-channel 20 successfully restarted on span 2 > -- B-channel 21 successfully restarted on span 2 > -- B-channel 22 successfully restarted on span 2 > -- B-channel 23 successfully restarted on span 2 > pbx1*CLI> > > (This is from an asterisk console started with -rvvvv.) > > Has anyone encountered similar behavior in the past? If so, what was the > resolution? > > Thanks, > > Tim McKee > > configs: > > zaptel.conf > # > # span 1 is for channel bank (24 FXS) > span=1,2,2,esf,b8zs > # span 2 is for NuVox PRI (1-23B, 24D) > span=2,1,2,esf,b8zs > # span 3 is unused > span=3,0,0,esf,b8zs > # span 4 is unused > span=4,0,0,esf,b8zs > # > # > > > zapata.conf > > [channels] > ; > ; Default language > ; > language=en > ; > ; Default context > ; > context=dialnational > ; > signalling=fxo_ks > use_callerid=yes > callwaiting=no > restrictcid=no > usecallingpres=yes > callwaitingcallerid=no > threewaycalling=yes > transfer=yes > cancallforward=yes > callreturn=yes > echocancel=yes > echocancelwhenbridged=no > echotraining=yes > relaxdtmf=no > rxgain=0.0 > txgain=0.0 > callgroup=1 > pickupgroup=1 > immediate=no > amaflags=billing > > <channels 1-24 omitted> > > ; > ; NuVox PRI > context=sdnpri > ; > switchtype = national > pridialplan = unknown > signalling = pri_cpe > callerid=asreceived > group = 2 > channel => 25-47 > > >
I thought the timing priority setting was for which incoming timing signal was used as the primary clock source, so I set the PRI as the highest priority clock source. In the telco world this is that way it normally works. Does the priority setting mean something different in zaptel.conf? My apologies to the group re the B channel restarts issue. My searches must have been too specific. ===================================================================Timothy R. McKee -----Original Message----- From: Bruce Komito [mailto:brucek@bagel.com] Sent: Monday, May 24, 2004 08:30 To: Timothy R. McKee Cc: asterisk-users@lists.digium.com Subject: Re: [Asterisk-Users] PRI problem??? Tim, I would double check the timing. It seems odd that you would supply clock rather than the switch, and if you get clock slips, that could certainly account for what you are seeing. Feel free to contact me off-list if you need more info or have any questions. Bruce Komito High Sierra Networks, Inc. www.servers-r-us.com (775) 284-5800 ext 115 On Sun, 23 May 2004, Timothy R. McKee wrote:> I have just finished installing a new asterisk box at my work. The > box is quite hefty, dual Zeon 2.8s with SCSI drives and 2Gb of memory. > I have a 4 port Digium T1 card for channel bank and PRI access. > > I activated a PRI from a local CLEC (DMS-500 based, National protocol). > This PRI is on slot 2 of the card and is set as the primary timing source. > It is ESF/B8ZS. > > All the software is latest CVS HEAD as of 5/22/04. > > My problem lies in random intermittent drops of calls. The entire PRI > seems to disappear, dropping all current established calls. I see > occasional printouts on an asterisk management console showing all 23 > B channels resetting with no reason given: > > pbx1*CLI> > -- B-channel 1 successfully restarted on span 2 > -- B-channel 2 successfully restarted on span 2 > -- B-channel 3 successfully restarted on span 2 > -- B-channel 4 successfully restarted on span 2 > -- B-channel 5 successfully restarted on span 2 > -- B-channel 6 successfully restarted on span 2 > -- B-channel 7 successfully restarted on span 2 > -- B-channel 8 successfully restarted on span 2 > -- B-channel 9 successfully restarted on span 2 > -- B-channel 10 successfully restarted on span 2 > -- B-channel 11 successfully restarted on span 2 > -- B-channel 12 successfully restarted on span 2 > -- B-channel 13 successfully restarted on span 2 > -- B-channel 14 successfully restarted on span 2 > -- B-channel 15 successfully restarted on span 2 > -- B-channel 16 successfully restarted on span 2 > -- B-channel 17 successfully restarted on span 2 > -- B-channel 18 successfully restarted on span 2 > -- B-channel 19 successfully restarted on span 2 > -- B-channel 20 successfully restarted on span 2 > -- B-channel 21 successfully restarted on span 2 > -- B-channel 22 successfully restarted on span 2 > -- B-channel 23 successfully restarted on span 2 pbx1*CLI> > > (This is from an asterisk console started with -rvvvv.) > > Has anyone encountered similar behavior in the past? If so, what was > the resolution? > > Thanks, > > Tim McKee > > configs: > > zaptel.conf > # > # span 1 is for channel bank (24 FXS) > span=1,2,2,esf,b8zs > # span 2 is for NuVox PRI (1-23B, 24D) span=2,1,2,esf,b8zs # span 3 is > unused span=3,0,0,esf,b8zs # span 4 is unused span=4,0,0,esf,b8zs # # > > > zapata.conf > > [channels] > ; > ; Default language > ; > language=en > ; > ; Default context > ; > context=dialnational > ; > signalling=fxo_ks > use_callerid=yes > callwaiting=no > restrictcid=no > usecallingpres=yes > callwaitingcallerid=no > threewaycalling=yes > transfer=yes > cancallforward=yes > callreturn=yes > echocancel=yes > echocancelwhenbridged=no > echotraining=yes > relaxdtmf=no > rxgain=0.0 > txgain=0.0 > callgroup=1 > pickupgroup=1 > immediate=no > amaflags=billing > > <channels 1-24 omitted> > > ; > ; NuVox PRI > context=sdnpri > ; > switchtype = national > pridialplan = unknown > signalling = pri_cpe > callerid=asreceived > group = 2 > channel => 25-47 > > >
My experience has been that the telco is the one that supplies the clock, because they must keep all of their circuits and equipment in sync. In fact, they typically derive their clocking ultimately from a satellite source. At one time, we had a switch interconnected with trunks to the ILEC, and we had to buy a similar clock to drive our equipment so that all of our circuits were in sync with theirs, as well as the other telcos we connected to. Before we did this, our PRIs (which were used for data only) would work for a while, but the modems dialed in to the PRI would eventually drop off, as the clock slippage increased to the point of being out of tolerance. Eventually, the entire PRI had to be restarted. When we installed the clock, all of the problems disappeared. Bruce Komito High Sierra Networks, Inc. www.servers-r-us.com (775) 284-5800 ext 115 On Mon, 24 May 2004, Timothy R. McKee wrote:> I thought the timing priority setting was for which incoming timing signal > was used as the primary clock source, so I set the PRI as the highest > priority clock source. In the telco world this is that way it normally > works. Does the priority setting mean something different in zaptel.conf? > > My apologies to the group re the B channel restarts issue. My searches must > have been too specific. > > > > ===================================================================> Timothy R. McKee > > > -----Original Message----- > From: Bruce Komito [mailto:brucek@bagel.com] > Sent: Monday, May 24, 2004 08:30 > To: Timothy R. McKee > Cc: asterisk-users@lists.digium.com > Subject: Re: [Asterisk-Users] PRI problem??? > > Tim, I would double check the timing. It seems odd that you would supply > clock rather than the switch, and if you get clock slips, that could > certainly account for what you are seeing. Feel free to contact me off-list > if you need more info or have any questions. > > Bruce Komito > High Sierra Networks, Inc. > www.servers-r-us.com > (775) 284-5800 ext 115 > > > On Sun, 23 May 2004, Timothy R. McKee wrote: > > > I have just finished installing a new asterisk box at my work. The > > box is quite hefty, dual Zeon 2.8s with SCSI drives and 2Gb of memory. > > I have a 4 port Digium T1 card for channel bank and PRI access. > > > > I activated a PRI from a local CLEC (DMS-500 based, National protocol). > > This PRI is on slot 2 of the card and is set as the primary timing source. > > It is ESF/B8ZS. > > > > All the software is latest CVS HEAD as of 5/22/04. > > > > My problem lies in random intermittent drops of calls. The entire PRI > > seems to disappear, dropping all current established calls. I see > > occasional printouts on an asterisk management console showing all 23 > > B channels resetting with no reason given: > > > > pbx1*CLI> > > -- B-channel 1 successfully restarted on span 2 > > -- B-channel 2 successfully restarted on span 2 > > -- B-channel 3 successfully restarted on span 2 > > -- B-channel 4 successfully restarted on span 2 > > -- B-channel 5 successfully restarted on span 2 > > -- B-channel 6 successfully restarted on span 2 > > -- B-channel 7 successfully restarted on span 2 > > -- B-channel 8 successfully restarted on span 2 > > -- B-channel 9 successfully restarted on span 2 > > -- B-channel 10 successfully restarted on span 2 > > -- B-channel 11 successfully restarted on span 2 > > -- B-channel 12 successfully restarted on span 2 > > -- B-channel 13 successfully restarted on span 2 > > -- B-channel 14 successfully restarted on span 2 > > -- B-channel 15 successfully restarted on span 2 > > -- B-channel 16 successfully restarted on span 2 > > -- B-channel 17 successfully restarted on span 2 > > -- B-channel 18 successfully restarted on span 2 > > -- B-channel 19 successfully restarted on span 2 > > -- B-channel 20 successfully restarted on span 2 > > -- B-channel 21 successfully restarted on span 2 > > -- B-channel 22 successfully restarted on span 2 > > -- B-channel 23 successfully restarted on span 2 pbx1*CLI> > > > > (This is from an asterisk console started with -rvvvv.) > > > > Has anyone encountered similar behavior in the past? If so, what was > > the resolution? > > > > Thanks, > > > > Tim McKee > > > > configs: > > > > zaptel.conf > > # > > # span 1 is for channel bank (24 FXS) > > span=1,2,2,esf,b8zs > > # span 2 is for NuVox PRI (1-23B, 24D) span=2,1,2,esf,b8zs # span 3 is > > unused span=3,0,0,esf,b8zs # span 4 is unused span=4,0,0,esf,b8zs # # > > > > > > zapata.conf > > > > [channels] > > ; > > ; Default language > > ; > > language=en > > ; > > ; Default context > > ; > > context=dialnational > > ; > > signalling=fxo_ks > > use_callerid=yes > > callwaiting=no > > restrictcid=no > > usecallingpres=yes > > callwaitingcallerid=no > > threewaycalling=yes > > transfer=yes > > cancallforward=yes > > callreturn=yes > > echocancel=yes > > echocancelwhenbridged=no > > echotraining=yes > > relaxdtmf=no > > rxgain=0.0 > > txgain=0.0 > > callgroup=1 > > pickupgroup=1 > > immediate=no > > amaflags=billing > > > > <channels 1-24 omitted> > > > > ; > > ; NuVox PRI > > context=sdnpri > > ; > > switchtype = national > > pridialplan = unknown > > signalling = pri_cpe > > callerid=asreceived > > group = 2 > > channel => 25-47 > > > > > > >
Hi Tim- Except for maybe eliminating the channel bank as a secondary source of timing, your conf files look fine. I notice that you've used "2" as the build-out (line length) parameter - I take it that your asterisk box is not right next to the switch. Anyway, I've never seen build-out make much difference, but maybe that's an issue. Although asterisk restarts B channels on a periodic basis, it's not supposed to do this if there's a call in progress, so there's something wrong. I noticed a problem recently that was causing D channels to drop under load, but after reporting this, Mark S fixed this on May 18th, so your more recent distr should cover it. Finally, in looking through libpri bugs on the bug list, Mark refers to some kind of firmware problem on some TE410P's (I think), so you might read through recent bug reports and see if any match your symptom (read the closed ones too) Is there anything remarkable in your /var/log/asterisk/messages log file? regards Scott M. Stingel President, Emerging Voice Technology, Inc. Palo Alto California & London England www.evtmedia.com _____________________________________________ From: asterisk-users-admin@lists.digium.com [mailto:asterisk-users-admin@lists.digium.com] On Behalf Of Timothy R. McKee Sent: Sunday, May 23, 2004 7:52 PM To: asterisk-users@lists.digium.com Subject: [Asterisk-Users] PRI problem??? I have just finished installing a new asterisk box at my work. The box is quite hefty, dual Zeon 2.8s with SCSI drives and 2Gb of memory. I have a 4 port Digium T1 card for channel bank and PRI access. I activated a PRI from a local CLEC (DMS-500 based, National protocol). This PRI is on slot 2 of the card and is set as the primary timing source. It is ESF/B8ZS. All the software is latest CVS HEAD as of 5/22/04. My problem lies in random intermittent drops of calls. The entire PRI seems to disappear, dropping all current established calls. I see occasional printouts on an asterisk management console showing all 23 B channels resetting with no reason given: configs: zaptel.conf # # span 1 is for channel bank (24 FXS) span=1,2,2,esf,b8zs # span 2 is for NuVox PRI (1-23B, 24D) span=2,1,2,esf,b8zs # span 3 is unused span=3,0,0,esf,b8zs # span 4 is unused span=4,0,0,esf,b8zs # # zapata.conf [channels] ; ; Default language ; language=en ; ; Default context ; context=dialnational ; signalling=fxo_ks use_callerid=yes callwaiting=no restrictcid=no usecallingpres=yes callwaitingcallerid=no threewaycalling=yes transfer=yes cancallforward=yes callreturn=yes echocancel=yes echocancelwhenbridged=no echotraining=yes relaxdtmf=no rxgain=0.0 txgain=0.0 callgroup=1 pickupgroup=1 immediate=no amaflags=billing <channels 1-24 omitted> ; ; NuVox PRI context=sdnpri ; switchtype = national pridialplan = unknown signalling = pri_cpe callerid=asreceived group = 2 channel => 25-47