| Summary: | DAHLIN-00266: ISDN resets after upgrade to DAHDI 2.5.0.2 | ||||
| Reporter: | Russell Brown (ruffle) | Labels: | |||
| Date Opened: | 2011-11-17 10:14:35.000-0600 | Date Closed: | 2012-04-03 16:48:03 | ||
| Priority: | Critical | Regression? | |||
| Status: | Closed/Complete | Components: | wct4xxp | ||
| Versions: | 2.5.0.2 | Frequency of Occurrence | Frequent | ||
| Related Issues: |
| ||||
| Environment: | Asterisk 1.8.7.1 wct4xxp+ d161:0205 Wildcard TE205P (3rd Gen) | Attachments: | ( 0) ISDNReset.log | ||
| Description: | Since upgrading a few days ago from Asterisk 1.4.42, Zaptel 1.4.12.1, libpri 1.4.11.4 to Asterisk 1.8.7.1, DAHDI 2.5.0.2 and libpri 1.4.12 I've been having increasing problems with the ISDN channels getting Yellow alarms and resetting. This cuts off all calls currently in progress :-( I've tried to find a common scenario but can't - it seems random. FWIW, it's happened 61 times today so far! - the earliest being 2 minutes after midnight and then throughout the night at various random times. There's nobody in the office at those times. During the day, it varies between once in any given hour to four times between 12:00 and 13:00. None of the hardware (Wildcard TE205P, BT ISDN-30 with 8 channels on Span-1, Argent Office ISDN legacy PBX on Span-2) has been changed from a week back when I was running zaptel. The Alarms are preset with events like: [Nov 17 14:23:26] NOTICE[25603] chan_dahdi.c: PRI got event: HDLC Abort (6) on D-channel of span 1 [Nov 17 14:23:26] NOTICE[25603] chan_dahdi.c: PRI got event: HDLC Bad FCS (8) on D-channel of span 1 [Nov 17 14:23:26] NOTICE[25603] chan_dahdi.c: PRI got event: HDLC Bad FCS (8) on D-channel of span 1 [Nov 17 14:23:27] NOTICE[25603] chan_dahdi.c: PRI got event: HDLC Abort (6) on D-channel of span 1 [Nov 17 14:24:25] WARNING[25972] chan_dahdi.c: Detected alarm on channel 5: Yellow Alarm [Nov 17 14:24:25] NOTICE[25603] chan_dahdi.c: PRI got event: Alarm (4) on D-channel of span 1 [Nov 17 14:24:25] WARNING[25603] sig_pri.c: Span 1: No D-channels available! Using Primary channel as D-channel anyway! [Nov 17 14:24:25] WARNING[25605] chan_dahdi.c: Detected alarm on channel 1: Yellow Alarm [Nov 17 14:24:25] WARNING[25605] chan_dahdi.c: Detected alarm on channel 2: Yellow Alarm My dahdi/system.conf is: # Span 1: TE2/0/1 "T2XXP (PCI) Card 0 Span 1" HDB3/CCS/CRC4 ClockSource span=1,1,0,ccs,hdb3,crc4 # termtype: te bchan=1-15,17-31 dchan=16 echocanceller=mg2,1-15,17-31 # Span 2: TE2/0/2 "T2XXP (PCI) Card 0 Span 2" (MASTER) HDB3/CCS/CRC4 span=2,0,2,ccs,hdb3,crc4 # termtype: te bchan=32-46,48-62 dchan=47 echocanceller=mg2,32-46,48-62 # Global data loadzone = uk defaultzone = uk dahdi_hardware says: pci:0000:03:00.0 wct4xxp+ d161:0205 Wildcard TE205P (3rd Gen) dahdi_scan says: [1] active=yes alarms=OK description=T2XXP (PCI) Card 0 Span 1 name=TE2/0/1 manufacturer=Digium devicetype=Wildcard TE205P (3rd Gen) (VPMOCT064) location=Board ID Switch 0 basechan=1 totchans=31 irq=16 type=digital-E1 syncsrc=1 lbo=0 db (CSU)/0-133 feet (DSX-1) coding_opts=AMI,HDB3 framing_opts=CCS,CRC4 coding=HDB3 framing=CCS/CRC4 [2] active=yes alarms=OK description=T2XXP (PCI) Card 0 Span 2 name=TE2/0/2 manufacturer=Digium devicetype=Wildcard TE205P (3rd Gen) (VPMOCT064) location=Board ID Switch 0 basechan=32 totchans=31 irq=16 type=digital-E1 syncsrc=1 lbo=266-399 feet (DSX-1) coding_opts=AMI,HDB3 framing_opts=CCS,CRC4 coding=HDB3 framing=CCS/CRC4 I'll attach a log extract showing the channels all going into reset. FWIW, I tried dahdi-linux-complete-2.6.0-rc1+2.6.0-rc1 but that generated hugely horrible hissing on the channels for ~75% of the calls. Ideas? Suggestions? I'm sadly going to have to revert to Zaptel if I can get this fixed promptly as callers are being cut off. | ||||
| Comments: | By: Russell Brown (ruffle) 2011-11-17 10:15:44.839-0600 /var/log/asterisk/full log extract with pri debug enabled showing a reset cycle whilst a call was in progress. By: Shaun Ruffell (sruffell) 2011-11-17 11:04:27.226-0600 So the OS / server also are the same version or did you change update any of those as well? By: Shaun Ruffell (sruffell) 2011-11-17 11:23:33.690-0600 Reading this some more, I'm also interested in the hissing you've experienced when updating to 2.6.0-rc1. If you want, would you contact me directly at sruffell@digium.com or sruffell on freenode.net? By: Russell Brown (ruffle) 2011-11-17 11:24:31.881-0600 Nothing changed except compiling and installing Asterisk 1.8.7.1, Libpri 1.4.12 and DAHDI 2.5.0.2 FWIW, the system's running Linux 2.6.24 on an Intel Core2 2.4GHz processor with 2Gb RAM and SATA disks in a md RAID-1 configuration. It's been running unmodified for at least a couple of years using various versions of Asterisk 1.4. I figured with the 'LTS' aspect of 1.4 about to expire I should move on up to DAHDI and the 1.8 series. Doing some Googling, I have seen references to the IRQs causing problems. I've not changed anything in the BIOS and cat /proc/interrupts says: CPU0 CPU1 0: 57 1 IO-APIC-edge timer 1: 84 82 IO-APIC-edge i8042 6: 1 2 IO-APIC-edge floppy 7: 0 0 IO-APIC-edge parport0 8: 1 0 IO-APIC-edge rtc 9: 0 0 IO-APIC-fasteoi acpi 16: 342184641 342163221 IO-APIC-fasteoi uhci_hcd:usb1, wct2xxp 17: 15053763 15073842 IO-APIC-fasteoi eth0 18: 0 0 IO-APIC-fasteoi uhci_hcd:usb3, uhci_hcd:usb6, ehci_hcd:usb7 19: 0 0 IO-APIC-fasteoi uhci_hcd:usb5 21: 0 0 IO-APIC-fasteoi uhci_hcd:usb2 22: 88 91 IO-APIC-fasteoi HDA Intel 23: 0 0 IO-APIC-fasteoi uhci_hcd:usb4, ehci_hcd:usb8 2301: 2145468 2145399 PCI-MSI-edge ahci NMI: 0 0 Non-maskable interrupts LOC: 15021850 11379652 Local timer interrupts RES: 117830 131845 Rescheduling interrupts CAL: 5151 2846 function call interrupts TLB: 235147 239914 TLB shootdowns TRM: 0 0 Thermal event interrupts THR: 0 0 Threshold APIC interrupts SPU: 0 0 Spurious interrupts ERR: 0 One can see that the Wildcard is sharing IRQ 16 with usb1; but I'm not using any kind of USB on this box (even the keyboard is PS/2). I do have a spare TE205P that I could install if you think that would help.... although I'm doubtful that it's a hardware issue as such. By: Shaun Ruffell (sruffell) 2012-01-09 11:07:32.649-0600 I'm just reviewing some of the open issues and lost track of this one. Any news about where we currently stand? By: Shaun Ruffell (sruffell) 2012-01-09 14:58:36.485-0600 Would you be willing to try the patch on DAHLIN-275 to see if that resolves the hissing you were hearing when using 2.6.0-rc1? By: Shaun Ruffell (sruffell) 2012-04-03 16:48:03.187-0500 I'm closing this out as incomplete. Please reopen if there is more to do on this issue. | ||||