Re: sendpage hangs

[Matthias.Apitz@SOFTCON.de]

Rex Fowler wrote:
    
    I wrote:
    
    >     
    >     I'm having some problem with alpha paging.  The environment is
    >     
    >     Solaris 2.4
    >     hylafax version hylafax-v4.0pl1
    >     
    >     sendpage is hanging apparently right after the 'SEND' command.
    >     I'm guessing it has something to do with reading or writing to the FIFO
    >     but I'm unsure.
    
    The same thing happens with sendpage and sendfax, they timeout and fail.
    
    If I run 
    	pagesend -m cua_b doneq/q47
    	faxsend -m cua_b doneq/q46
    they succeed.
    
    So the problem seems to be with faxq and hfaxd processes communicating.
    
    I should have mentioned that I've tried the -q and sendpage will return
    immediately but the message is never delivered and no session logs
    are created.  It never gets as far as trying to initialize and
    dial the modem.   Are there any FIFO peculiarities in Solaris
    besides what configure tried to figure out?  I have compiled with and 
    without the "FIFO bug" option for Solaris and get the same results.
    
    
    Thanks for any tips,

The tip is as usual: enable the tracing of the server(s), in this
case for the messages passed through the FIFO's, and see what's
going on.

	matthias
    
    >     
    >     Here's a sample telnet sesion with port 444
    >     
    >     raptor 956$ telnet localhost 444
    >     Trying 127.0.0.1 ...
    >     Connected to localhost.
    >     Escape character is '^]'.
    >     220 raptor.mtc.ti.com SNPP server (HylaFAX (tm) Version 4.0pl1) ready.
    >     login rmfowler
    >     230 User rmfowler logged in.
    >     page rex
    >     500 PAFE: Command not recognized.
    >     page rex
    >     250 Pager ID accepted; provider: xxxyyyy pin: wwwzzzz jobid: 40.
    >     mess testing telnet
    >     250 Message text OK.
    >     send
    >     250 Message processing completed.
    >     
    >     There is a few minute delay before the '250 Message processing completed'
    > 
    > If you don't use the "-q" flag of sendpage(1) (or don't use
    > the "SITE JQUEUE yes" command in the protocol session with
    > telnet) the sendpage(1) will wait for the job delivery to
    > the paging central service. This may take some time (modem
    > dialing and handshaking, error situations etc.).
    > 
    > Look into a session log with "tail -f ....." to see what's
    > going on during page delivery.
    > 
    >     If I interrupt it, I can then use 'pagesend' to send the message 
    >     successfully.
    >     
    >     Any ideas of what I can look or if there's more information I can give you.
    >     Since pagesend works, I believe the modem and serial port are setup ok.
    >     The problem seems to be communication between hfaxd and faxq.
    > 
    > Don't think so but you may look into the servertracing with
    > enabled trace for the FIFO messages (0x04000).
    > 
    > 	matthias
    >     
    >     They 2 processed do share a FIFO, correct?  From what I can tell,
    >     another FIFO per device is created when using faxgetty, which I'm not
    >     currently.
    >     
    >     Thanks,
    >     
    >     -- 
    >     Rex Fowler              http://www.mtc.ti.com/~rmfowler
    >     (972)997-2779           mailto:rmfowler@mtc.ti.com
    >     Alpha Pager             http://www.mtc.ti.com/cgi-bin/alpha_pager.cgi
    >     TI MSG                  rfow
    
    
    -- 
    Rex Fowler              http://www.mtc.ti.com/~rmfowler
    (972)997-2779           mailto:rmfowler@mtc.ti.com
    Alpha Pager             http://www.mtc.ti.com/cgi-bin/alpha_pager.cgi
    TI MSG                  rfow