[hylafax-users] rejected pages being accpted anyway

[Edward Barrett <feb@skibowski.net>]

I'm using hylafax-4.1beta2-7rh6 (from the RPM)  on a stock Red Hat Linux 6.2 install.  We frequently receive faxes that have 4 or 5 consecutive bad lines, so I set the retransmit threshold to 2.  Looking through the logs, I can see that hylafax asks the sender to retransmit the bad page, and the sender does so, but hylafax keeps the bad page anyway. The following is the relevant part of the most recent log file that shows this problem (server trace set to 2):

Oct 10 10:57:58.71: [  669]: <-- [11:AT+FRM=146\r]
Oct 10 10:57:59.59: [  669]: --> [7:CONNECT]
Oct 10 10:57:59.59: [  669]: RECV: begin page
Oct 10 10:58:26.67: [  669]: RECV: 1438 total lines, 5 bad lines, 5 consecutive 
bad lines
Oct 10 10:58:26.67: [  669]: RECV: REJECT page quality, 5-line run (max 2)
Oct 10 10:58:26.67: [  669]: RECV: end page
Oct 10 10:58:26.67: [  669]: --> [10:NO CARRIER]
Oct 10 10:58:26.67: [  669]: <-- [9:AT+FRH=3\r]
Oct 10 10:58:26.84: [  669]: --> [7:CONNECT]
Oct 10 10:58:27.89: [  669]: --> [2:OK]
Oct 10 10:58:27.89: [  669]: RECV recv MPS (more pages, same document)
Oct 10 10:58:27.89: [  669]: <-- [9:AT+FTH=3\r]
Oct 10 10:58:28.82: [  669]: --> [7:CONNECT]
Oct 10 10:58:28.82: [  669]: <-- data [3]
Oct 10 10:58:28.82: [  669]: <-- data [2]
Oct 10 10:58:29.14: [  669]: --> [2:OK]
Oct 10 10:58:29.14: [  669]: RECV send RTN (retrain negative)
Oct 10 10:58:29.14: [  669]: <-- [9:AT+FRH=3\r]
Oct 10 10:58:29.30: [  669]: --> [7:CONNECT]
Oct 10 10:58:31.00: [  669]: --> [2:OK]
Oct 10 10:58:31.00: [  669]: REMOTE TSI "XXXXX"
Oct 10 10:58:31.00: [  669]: <-- [9:AT+FRH=3\r]
Oct 10 10:58:31.01: [  669]: --> [7:CONNECT]
Oct 10 10:58:31.27: [  669]: --> [2:OK]
Oct 10 10:58:31.27: [  669]: REMOTE wants 14400 bit/s
Oct 10 10:58:31.27: [  669]: REMOTE wants page width 1728 pixels in 215 mm
Oct 10 10:58:31.27: [  669]: REMOTE wants unlimited page length 
Oct 10 10:58:31.27: [  669]: REMOTE wants 3.85 line/mm
Oct 10 10:58:31.27: [  669]: REMOTE wants 2-D MR
Oct 10 10:58:31.27: [  669]: RECV training at v.17 14400 bit/s
Oct 10 10:58:31.27: [  669]: <-- [11:AT+FRM=145\r]
Oct 10 10:58:33.13: [  669]: --> [7:CONNECT]
Oct 10 10:58:34.63: [  669]: RECV: TCF 2704 bytes, 0% non-zero, 2697 zero-run
Oct 10 10:58:34.63: [  669]: --> [10:NO CARRIER]
Oct 10 10:58:34.63: [  669]: DELAY 75 ms
Oct 10 10:58:34.71: [  669]: <-- [9:AT+FTH=3\r]
Oct 10 10:58:35.57: [  669]: --> [7:CONNECT]
Oct 10 10:58:35.57: [  669]: <-- data [3]
Oct 10 10:58:35.57: [  669]: <-- data [2]
Oct 10 10:58:35.89: [  669]: --> [2:OK]
Oct 10 10:58:35.89: [  669]: TRAINING succeeded
Oct 10 10:58:35.89: [  669]: <-- [11:AT+FRM=146\r]
Oct 10 10:58:36.79: [  669]: --> [7:CONNECT]
Oct 10 10:58:36.79: [  669]: RECV: begin page
Oct 10 10:59:05.48: [  669]: RECV: 1432 total lines, 0 bad lines, 0 consecutive 
bad lines
Oct 10 10:59:05.48: [  669]: RECV: end page
Oct 10 10:59:05.48: [  669]: --> [10:NO CARRIER]
Oct 10 10:59:05.48: [  669]: <-- [9:AT+FRH=3\r]
Oct 10 10:59:05.65: [  669]: --> [7:CONNECT]
Oct 10 10:59:06.70: [  669]: --> [2:OK]
Oct 10 10:59:06.70: [  669]: RECV recv EOP (no more pages or documents)
Oct 10 10:59:06.70: [  669]: <-- [9:AT+FTH=3\r]
Oct 10 10:59:07.63: [  669]: --> [7:CONNECT]
Oct 10 10:59:07.63: [  669]: <-- data [3]
Oct 10 10:59:07.63: [  669]: <-- data [2]
Oct 10 10:59:07.95: [  669]: --> [2:OK]
Oct 10 10:59:07.95: [  669]: RECV send MCF (message confirmation)
Oct 10 10:59:07.95: [  669]: RECV FAX (00000277): from XXXXX page 3 in 1:09, INF, 3.85 line/mm, 2-D MR
Oct 10 10:59:07.95: [  669]: RECV FAX (00000277): recvq/fax00226.tif from XXXXX, route to <unspecified>, 3 pages in 2:38
Oct 10 10:59:07.95: [  669]: <-- [9:AT+FRH=3\r]
Oct 10 10:59:08.11: [  669]: --> [7:CONNECT]
Oct 10 10:59:09.27: [  669]: --> [2:OK]
Oct 10 10:59:09.27: [  669]: RECV FAX: bin/faxrcvd "recvq/fax00226.tif" "ttyS1" 
"00000277" ""
Oct 10 10:59:09.49: [  669]: RECV FAX: end
Oct 10 10:59:09.49: [  669]: SESSION END

This causes Kodak Imaging (the clients are using Windows 98) to barf
on the third page of the fax ("invalid compression format").  If I
convert it to a pdf (fax2ps->ps2pdf) ghostview can read it, but Acrobat
Reader shows it as greyed out text.  I can still read it using viewfax on
my linux box, and can see which five lines are corrupted.  Any suggestions?

Thanks,
F. Edward Barrett



____________________ HylaFAX(tm) Users Mailing List _______________________
 To unsub: mail -s unsubscribe hylafax-users-request@hylafax.org < /dev/null