[Asterisk-Dev] E1, crc-4 errors

Martin Pycko martinp at digium.com
Sat May 10 15:33:19 MST 2003


What's the load on your box ?
Check if the board shares IRQ with some other devices (/proc/interrupts)
Find out if your E1 cable is properly done (certain colors are assigned to
certain pins on the connector)

You can ask your operator to run remote loopback on your card and that
should show if it's the cable's fault.

Martin

On Sat, 10 May 2003, Konrad Gorski wrote:

> Thank you Pauline.
> Unfortunately your suggestion do not seem to solve my problem.
> I've made following 15 tests:
> zaptel.conf:
>  	span=1,[012],[01457],ccs,hdb3,crc4
> ... a lot of errors (FAR-END FAILED SECONDS > 98%) regardless of
> parameters
> I will try to repeat tests on Monday. Do you have any suggestions how to
> test and what to do if tests fail?
>     Serdeczne pozdrowienia,
> 	Konrad Górski
>
>
> -----Original Message-----
> From: asterisk-dev-admin at lists.digium.com
> [mailto:asterisk-dev-admin at lists.digium.com] On Behalf Of Pauline
> Middelink
> Sent: Friday, May 09, 2003 6:05 PM
> To: asterisk-dev at lists.digium.com
> Subject: Re: [Asterisk-Dev] E1, crc-4 errors
>
> On Fri, 09 May 2003 around 16:04:42 +0200, Konrad Gorski wrote:
> > Hi,
> > My PBX config:
> > Telco (Lucent SS7) E1 line (crc4) <=> * box + E100P, T100P, Zplex <=>
> > analog phones and ATA186 (SIP and MGCP)
> > * is working very good, but from Telco side are problems.
> > Telco's list of errors:
> > ----------------------------------------------------------
> >    ERROR STATISTICS FOR : ALL MEASUREMENT TYPES
> >    --------------------
> >
> >    MEASUREMENT PERIOD : CURRENT DAY
> >
> >    ES CALCULATION MODE          = CRC-4
> >    FAC AVAILABLE TIME           = 09:58:45 HOURS
> >    FAC UNAVAILABLE TIME         = 00:02:18 HOURS
> >    FAC FAR-END AVAILABLE TIME   = 00:09:18 HOURS
> >    FAC FAR-END UNAVAILABLE TIME = 09:51:45 HOURS
> >
> >    MEASUREMENT TYPE                 ERROR     ERROR
> >                                     COUNT     RATIO
> >    ----------------                 -----     -----
> >
> >    ERRORED SECONDS                  11     0.03062%
> >    FAILED SECONDS                   138    0.38266%
> >    BURSTY ERRORED SECONDS           7      0.01941%
> >    SEVERELY ERRORED SECONDS         9      0.02496%
> >    FRAME SLIP SECONDS               6
> >    CARRIER FAILURES                 1
> >    FAR-END ERRORED SECONDS          2      0.35842%
> >    FAR-END FAILED SECONDS           35505 98.45271%
> >    FAR-END BURSTY ERRORED SECONDS   54     0.14974%
> >    FAR-END SEVERELY ERRORED SECONDS 35453 98.30852%
> >    DEGRADED MINUTES                 3
> > ----------------------------------------------------------
> > Configuration for above test:
> > Telco (Lucent SS7) E1 line (crc4) <=> E100P + wct1xxp loaded ( *
> > stopped, T100P unmounted)
> > zaptel.conf:
> > 	span=1,0,0,ccs,hdb3,crc4
> >
> > If E1 line configured w/o crc4 - error count is 0.
> > Configuration for test w/o crc4:
> > Telco (Lucent SS7) E1 line (w/o crc4) <=> E100P + wct1xxp loaded ( *
> > stopped, T100P unmounted)
> > zaptel.conf:
> > 	span=1,0,0,ccs,hdb3
>
> You have to take the clock source from the Telco I think.
> I.e.    span=1,1,0,ccs,hdb3
>
> Your E1 line is now freewheeling, so sync between the telco
> and your E100P shifts in and out phase.
>
> > Buggy CRC calculation? (It's opinion of Telco's operator)
> > or my E100P is faulty?
>
> <wild guess here>
> When crc4 is deactivated, the telco can't really measure
> errors, or the physical connection must really be going
> down. Its the crc4 checksum which gives you the possibility
> to detect errors.
> </wild>
>
>     Met vriendelijke groet,
>         Pauline Middelink
> --
> GPG Key fingerprint = 2D5B 87A7 DDA6 0378 5DEA  BD3B 9A50 B416 E2D0 C3C2
> For more details look at my website http://www.polyware.nl/~middelink
>
> _______________________________________________
> Asterisk-Dev mailing list
> Asterisk-Dev at lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-dev
>




More information about the asterisk-dev mailing list