Vassilis Konstantinou
2005-Jan-09 03:18 UTC
[Asterisk-Users] X100P random hangups - Please help with suggestions
Thanks for the reply Bill. I am aware of the interrupts problem. To solve it I have already disabled my serial ports freeing up interrupts 3 and 4 and these are allocated to the two cards. This was done 2 months ago and has not solved the problem. Is there any way that something can wake up every now and then and generate these two interrupts? My current /proc/interrupts is as follows: CPU0 0: 185392655 XT-PIC timer 1: 13 XT-PIC keyboard 2: 0 XT-PIC cascade 3: 1853793865 XT-PIC wcfxo 4: 1853787231 XT-PIC wcfxo 8: 2 XT-PIC rtc 9: 0 XT-PIC acpi 10: 0 XT-PIC Intel 82801BA-ICH2 12: 57674469 XT-PIC eth0, PS/2 Mouse 14: 2097777 XT-PIC ide0 15: 10619 XT-PIC ide1 NMI: 0 LOC: 0 ERR: 0 MIS: 0 The wcfxo's are clearly allocated their own INT but can something else mess-up with these interrupts? Vassilis At 09:54 09/01/2005, you wrote:><< Both of the X100Ps seem to randomly hang-up both incoming and outgoing >calls.>> > >I think most people who use X100P cards (clone or originals) have had your >experience. So far as I can tell, the cause is always an interrupt problem. >Specifically that affected X100P cards share an interrupt with one or more >other devices. Have you checked for shared interrupts using the command: > >cat /proc/interrupts > >to see if any interrupts are shared?
Vassilis Konstantinou
2005-Jan-09 05:32 UTC
[Asterisk-Users] X100P random hangups - debug info
OK Some more information. I have changed interupts for the 2 X100Ps and the problem did not go away. Looking at the detailed debug information I always get the following from before the disconnect: ===========================Jan 9 12:13:40 DEBUG[1992]: Exception on 17, channel 1 Jan 9 12:13:40 DEBUG[1992]: Got event On hook(1) on channel 1 (index 0) Jan 9 12:13:40 DEBUG[1992]: disabled echo cancellation on channel 1 Jan 9 12:13:40 DEBUG[1992]: Didn't get a frame from channel: Zap/1-1 Jan 9 12:13:40 DEBUG[1992]: Bridge stops bridging channels SIP/5001-2b20 and Zap/1-1 Jan 9 12:13:40 DEBUG[1992]: Hangup: channel: 1 index = 0, normal = 17, callwait = -1, thirdcall = -1 Jan 9 12:13:40 DEBUG[1992]: disabled echo cancellation on channel 1 Jan 9 12:13:40 DEBUG[1992]: Set option TDD MODE, value: OFF(0) on Zap/1-1 ========================== Zap1 is the UK BT line. SIP/5001 is one of the extensions connected to the Cisco ATA186. Can somebody help me with what is happening with the On hook event? Is there something I could try to not drop the line? Vassilis
Vassilis Konstantinou
2005-Jan-11 12:02 UTC
[Asterisk-Users] X100P random hangups - Please help with suggestions
Just to answer my own question so that this loop is closed and hopefully others who suffer from the same problem may find this useful. The problem in the end had nothing to do with Asterisk of the X100P or at least not directly. After pulling my hair for months and checking various bits of code, drivers and hardware, I realised that the problem was cause by a stupid desktop digital satellite box (UK SKY)connected in parallel (well ...same line other socket) with the X100Ps. The box has a modem and has to phone 'the mother ship' every so often for billing purposes mainly. God knows what it does to the line in between but my suspicion is that the combination of impedance of the two was the main cause of the random hangups. As soon as I removed it from the socket, the problem disappeared! So if you get random hangups with X100Ps and you have other devices connected directly on the main line, check those as well. Many thanks to all the people who contributed ideas. Vassilis At 10:18 09/01/2005, you wrote:>Thanks for the reply Bill. > >I am aware of the interrupts problem. To solve it I have already disabled >my serial ports freeing up interrupts 3 and 4 and these are allocated to >the two cards. This was done 2 months ago and has not solved the problem. >Is there any way that something can wake up every now and then and >generate these two interrupts? My current /proc/interrupts is as follows: > > CPU0 > 0: 185392655 XT-PIC timer > 1: 13 XT-PIC keyboard > 2: 0 XT-PIC cascade > 3: 1853793865 XT-PIC wcfxo > 4: 1853787231 XT-PIC wcfxo > 8: 2 XT-PIC rtc > 9: 0 XT-PIC acpi > 10: 0 XT-PIC Intel 82801BA-ICH2 > 12: 57674469 XT-PIC eth0, PS/2 Mouse > 14: 2097777 XT-PIC ide0 > 15: 10619 XT-PIC ide1 >NMI: 0 >LOC: 0 >ERR: 0 >MIS: 0 > > >The wcfxo's are clearly allocated their own INT but can something else >mess-up with these interrupts? > >Vassilis > > >At 09:54 09/01/2005, you wrote: >><< Both of the X100Ps seem to randomly hang-up both incoming and outgoing >>calls.>> >> >>I think most people who use X100P cards (clone or originals) have had your >>experience. So far as I can tell, the cause is always an interrupt problem. >>Specifically that affected X100P cards share an interrupt with one or more >>other devices. Have you checked for shared interrupts using the command: >> >>cat /proc/interrupts >> >>to see if any interrupts are shared? > > >_______________________________________________ >Asterisk-Users mailing list >Asterisk-Users@lists.digium.com >lists.digium.com/mailman/listinfo/asterisk-users >To UNSUBSCRIBE or update options visit: > lists.digium.com/mailman/listinfo/asterisk-users