Re: flexfax: Large number of errors with Hylafax and ZyXEL Omni modem

[Steve Williams <steve@genie96.com>]

[snip]


> Dear Steve,
> 
> | I will include an awk script that you can use to parse the xferlog 
> | of HylaFax.  This will give a standard output that some of us on the
> | mailing list can compare to.
> 
> Here is the output from your awk script.
> 

Thanks for posting it.  It gives us a "standard of comparison", which I
was not able to relate to from your statistics.

First thing I can say is WOW!!!

It looks sooo bad that I almost feel like mailing you one of my ZDX modems
to test out & see if there is a similar problem <g>.  I have no idea if
it would even work though, coming from Canada and such.

I don't have a whole lot to recommend except that I must concur with 
Darren that possibly the modem config file is not optimum for your modem.
Have you tried an alternative handshaking ( rtscts if it is xon or 
the other way around ).

Do you have the most recent firmware for the modem (don't hit me, I 
always try to start with the simple & work to the more difficult <g>).

I am on the mgetty+sendfax mailing list, and some of the people there
seem to have a good relationship with some of the modem vendors.

Have you considered sent off one of the detailed HylaFax logs to 
the modem manufacturer & got their feedback?  I have seen where they
will even try sending a fax to the same number ( given that the
receiver is cooperative ) in an attempt to determin the problem.

I have to agree with you that this rate is incredibly bad.... and 
I have a personal "gut" feeling that it is either the modem or the
config file for the modem.

The USR Sportsters, one of the most popular modems sold, are agreed by
EVERYONE, that for commercial use, the ARE JUNK!!!.

I don't think your modem comes close to this category, but it shows that
there are GREAT variety between modem manufacturers.

I still think your best bet is with the Modem Vendor.  With the detailed
log files, they should be able to see something.

One thing did jump out at me... 

> No local dialtone                                              167     25.3

why are there 167 faxes where there is no local dial dial tone????

I just had another thought... I am sure you have posted this before, but
what kind of system is your server?  If it was under powered, or running
a heavy commercial load, maybe that would affect critical timing.

I think that is HIGHLY unlikey, because the errors would be more evenly 
distributed, rather than just being the "Phase B" errors.  Just a thought
late at night after an evening at the Pub <g>.

I don't know ANYTHING about fax specs, so I can't comment on anything
beyond what I have spewed forth already.

Keep the faith, HylaFax is a good product.

> Modem: modem
> Total faxes sent:                                                      6609
> 
> Error                                                           No  Rate/1000
> ------------------------------------------------------------------------------
> Busy signal detected                                            21      3.2
> Failure to train at 2400 bps or +FMINSP value; Giving up aft     9      1.4
> No response to MPS repeated 3 times; Giving up after 3 attem    60      9.1
> Unspecified Transmit Phase B error                            2444    369.8
> COMREC error in transmit Phase B/got DCN                         2      0.3
> Unable to set session parameters                                 6      0.9
> DTE to DCE data underflow                                        1      0.2
> Failure to train at 2400 bps or +FMINSP value; too many atte     1      0.2
> RSPREC error/got DCN                                            25      3.8
>                                                                  1      0.2
> Unspecified Transmit Phase D error, including +FPHCTO timeou     1      0.2
> RSPREC error/got DCN; Giving up after 3 attempts to send sam     8      1.2
> Unspecified Transmit Phase B error; too many attempts to dia    80     12.1
> No response to EOP repeated 3 times; Giving up after 3 attem     6      0.9
> Unknown problem (check modem power)                              5      0.8
> Unable to transmit page (giving up after 3 attempts); too ma     1      0.2
> No response to MPS repeated 3 times                            123     18.6
> Unable to set session parameters; Giving up after 3 attempts     1      0.2
> Communication failure during Phase B/C; Giving up after 3 at     1      0.2
> Unspecified Transmit Phase D error, including +FPHCTO timeou     4      0.6
> No response to EOP repeated 3 times                              4      0.6
> Unable to transmit page (giving up after 3 attempts); Giving   294     44.5
> Job aborted by user                                              1      0.2
> No local dialtone                                              167     25.3
> Communication failure during Phase B/C                           3      0.5
> Failure to train at 2400 bps or +FMINSP value                   43      6.5
> No carrier detected; too many attempts to dial                   9      1.4
> High resolution document is not supported by client, image r     1      0.2
> Unspecified Transmit Phase B error; Giving up after 3 attemp     1      0.2
> Document was encoded with 2DMR, but client does not support      1      0.2
> Unable to transmit page (giving up after 3 attempts)           679    102.7
> Call aborted,  from +FK or <CAN>                                 1      0.2
> No carrier detected                                           1114    168.6
> ------------------------------------------------------------------------------
> Total                                                         5118    774.4
> 
> 
> Please note that I am not concerned by the NO CARRIER errors, as I
> mentioned in my earlier mails. Tell me why 37% of all my outgoing call
> attempts generate the "Unspecified Transmit Phase B Error." As I said, I
> am only concerned about the errors in the calls that actually get
> answered, i.e. the ones that I get billed for, by the telco.
> 
> Hope you now see my point. And I'm unwilling to accept that all this is
> simply a case of bad Indian phone lines. Lines are bad, yes, but not
> this bad, I feel.
> 
> love,
> Shuvam
> 


-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )