[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
20000202: pl15 strange behavior
- Subject: 20000202: pl15 strange behavior
- Date: Fri, 04 Feb 2000 09:47:01 -0700
Chris,
when you see the scrolling messag send lines, that is because the program
is not sucessful in connecting to the gplt process. This could be due to the
previous
invocation that you had that exited unexpectedly.
Also, since the program had exited out abnormally, this could have caused the
output of the gemglb.nts to become corrupt, or your disk space could have been
unwritable or full etc.
Check to make sure you don't have any running copies of gplt, gf, xw etc.
Then check to make sure you son't have any message queues still running.
Make sure you don't have any files or subdirectories in your working
directory called "gf" or "xw" or "gplt" that could be confused with
the name of an executable program.
Then, ensure you have $respond set to yes.
Steve Chiswell
Unidata User Support
>From: address@hidden (Chris Hennon)
>Organization: .
>Keywords: 200002041631.JAA13396
>Steve -
>
>When I do "list $respond" at the prompt, it says:
>
>$RESPOND =
>
>I tried to set respond to yes by "$respond = yes" at the GDCNTR prompt.
>But when I ran the program, an infinite stream of:
>
>Error in message send = 22
>itype, ichan, nwords,2,151,2
>
>scrolls by. I don't see how this got unset from yes in the first place
>because I didn't change anything manually.
>
>Also, is this responsible for kicking me out of the program after each
>run? I'm still unable to write gifs as well (SIGBUS), even though there
>are no dead message queues nor rogue gplts around. Thanks.
>
>Chris
>
>
>On Wed, 2 Feb 2000, Unidata Support wrote:
>
>>
>> Chris,
>>
>> when programs don't ask you for confirmation, and exit immediately, this
>> generally indicates that your $respond variable is set to no.
>> You should find this if you look at the value in your gemglb.nts file,
>> or by typing "list $respond" at the gempak prompt. If this is the case,
>> set the value $respond=yes.
>>
>> The $respond variable allows scripts and things that don't want certain
>> output to be shown to be turned off....by I usually leave it set to yes so
>> that I can debug any scripts that fail- and just set the crontab
>> entry to redirect the output to /dev/null if I don't want to see the output.
>>
>> Steve Chiswell
>> Unidata User Support
>>
>>
>> >From: address@hidden (Chris Hennon)
>> >Organization: .
>> >Keywords: 200002021543.IAA03167
>>
>> >Steve -
>> >
>> >Strange behavior is continuing. First, I'm trying to make a plot for xw
>> >display using gdcntr. When I run, gempak does not prompt me to hit
>> >"Return" to continue plotting but immediately plots the map in a window,
>> >then exits gdcntr all together without any help from me. Output below:
>> >
>> > GEMPAK-GDCNTR>r
>> >Creating process: xw for queue 151
>> >
>> > Grid file: $HDS/2000020200_ruc_grid211.gem
>
>> >
>> > GRID IDENTIFIER:
>> > TIME1 TIME2 LEVL1 LEVL2 VCORD PARM
>> >000202/0000F003 200 PRES TMPC
>> >
>> > GAREA: 13;-125;55;-52
>> >SCALE: 0
>> >
>> > MINIMUM AND MAXIMUM VALUES -66.75 -42.12
>> >
>> > LINE CONTOURS:
>> >
>> > LEVELS: -66.00 -64.00 -62.00 -60.00 -58.00 -56.00
>> >-54.00
>> > COLORS: 4 4 4 4 4 4
>> >4
>> > LINTYP: 3 3 3 3 3 3
>> >3
>> > LINWID: 1 1 1 1 1 1
>> >1
>> > LABEL: 1 1 1 1 1 1
>> >1
>> >
>> > LEVELS: -52.00 -50.00 -48.00 -46.00 -44.00
>> > COLORS: 4 4 4 4 4
>> > LINTYP: 3 3 3 3 3
>> > LINWID: 1 1 1 1 1
>> > LABEL: 1 1 1 1 1
>> >shear:[/usr/local/gempak]%
>> >
>> >So I have to restart gdcntr everytime. So then I attempted to make a gif
>> >file, so I restarted gdcntr and changed device to gf. When I ran gdcntr,
>> >here's what happened:
>> >
>> >1. The xw which contained my previous plot dissapeared (I never typed
>> >gpend)
>> >
>> >2. gdcntr once again did not prompt me for my OK, but just kicked me out
>> >of the program.
>> >
>> >3. After typing "gpend" at the prompt, I received this message:
>> >
>> >*** TERMINATING gf
>> >*** Received signal 10 SIGBUS
>> >
>> >and it hangs. The output from this run looked like:
>> >
>> > GEMPAK-GDCNTR>device = gf
>> > GEMPAK-GDCNTR>r
>> >Creating process: gf for queue 151
>> >
>> > Grid file: $HDS/2000020200_ruc_grid211.gem
>
>> >
>> > GRID IDENTIFIER:
>> > TIME1 TIME2 LEVL1 LEVL2 VCORD PARM
>> >000202/0000F003 200 PRES TMPC
>> >
>> > GAREA: 13;-125;55;-52
>> >SCALE: 0
>> >
>> > MINIMUM AND MAXIMUM VALUES -66.75 -42.12
>> >
>> > LINE CONTOURS:
>> >
>> > LEVELS: -66.00 -64.00 -62.00 -60.00 -58.00 -56.00
>> >-54.00
>> > COLORS: 4 4 4 4 4 4
>> >4
>> > LINTYP: 3 3 3 3 3 3
>> >3
>> > LINWID: 1 1 1 1 1 1
>> >1
>> > LABEL: 1 1 1 1 1 1
>> >1
>> >
>> > LEVELS: -52.00 -50.00 -48.00 -46.00 -44.00
>> > COLORS: 4 4 4 4 4
>> > LINTYP: 3 3 3 3 3
>> > LINWID: 1 1 1 1 1
>> > LABEL: 1 1 1 1 1
>> >shear:[/usr/local/gempak]%
>> >
>> >There are no dead message queues and no rogue gplts hanging around. I
>> >don't know what else to try. Again, all of this behavior began
>> >after I installed pl15 on my machine and took pl12 off (see below).
>> >Thanks for any advice.
>> >
>> >Chris
>> >
>> >
>> >
>> >On Sun, 16 Jan 2000, Unidata Support wrote:
>> >
>> >>
>> >> Chris,
>> >>
>> >> The queues shown by ipcs are those that are orphaned
>> >> when you had to ctrl-c out of the program.
>> >>
>> >> Remove the queues with the ipcrm command like "ipcrm -q 1" etc.,
>> >> the scroll of the mailbox messages occurs when the process tries to
>> >> attatch to the orphaned queue.
>> >>
>> >> After you remove the queues, try running again. It could be that
>> >> you were unable to connect to the X display the first time, or
>> >> that there was a queue already in use. Let me know if the
>> >> problem persists. Its possible that a copy of gplt that
>> >> had been running previously (before pl15) was still in use at
>> >> the time.
>> >>
>> >> Steve Chiswell
>> >> Unidata User Support
>> >>
>> >> >From: address@hidden (Chris Hennon)
>> >> >Organization: .
>> >> >Keywords: 200001170058.RAA03824
>> >>
>> >> >Steve -
>> >> >
>> >> >I installed pl15 on my system and now am having problems running scripts
>> >> >that ran fine on pl12. When I try to write a gf within a script using
>> >> >gdcntr, gempak gives me:
>> >> >
>> >> >*** TERMINATING gf
>> >> >*** Received signal 10 SIGBUS
>> >> >
>> >> >and then it hangs. A ctl-c gives me this:
>> >> >
>> >> >*** TERMINATING gpend
>> >> >*** Received signal 2 SIGINT
>> >> >0.0u 0.0s 0:22 0% 0+0k 0+0io 0pf+0w
>> >> >
>> >> >When I try to run gdcntr outside of a script and create a gif, I get an
>> >> >infinite scroll of:
>> >> >
>> >> >Error in message send = 22
>> >> >itype, ichan, nwords,2,3,2
>> >> >
>> >> >A "ps -e" shows no rogue gplt's around. Also, "ipcs" shows:
>> >> >
>> >> >IPC status from <running system> as of Sun Jan 16 19:53:22 EST 2000
>> >> >T ID KEY MODE OWNER GROUP
>> >> >Message Queues:
>> >> >q 0 0x47000697 -Rrw-rw-rw- gempak ldm
>> >> >q 1 0x44000697 -Rrw-rw-rw- gempak ldm
>> >> >q 2 0x47000699 -Rrw-rw-rw- gempak ldm
>> >> >Shared Memory:
>> >> >m 0 0x500007a4 --rw-r--r-- root root
>> >> >Semaphores:
>> >> >
>> >> >The xw device seems to work fine. I'm running on Solaris 5.7. I
>> >> >installed the latest patch in an entirely new directory than the old one
> ,
>> >> >which I have now eliminated from my system. I built pl15 and installed
> it
>> >> >without a problem.
>> >> >
>> >> >Thanks for any feedback.
>> >> >
>> >> >Chris
>> >> >
>> >> >================================================
>> >> >| Chris Hennon Ohio State University |
>> >> >| Tropical Meteorology address@hidden |
>> >> >| |
>> >> >| Dept of Geography Office: 1155 Derby Hall |
>> >> >| 1036 Derby Hall Phone : (614) 292-2704 |
>> >> >| Columbus, OH 43210 Fax : (614) 292-6213 |
>> >> >================================================
>> >> >
>> >>
>> >> *************************************************************************
> ***
>> >> Unidata User Support UCAR Unidata Prog
> ram
>> >> (303)497-8644 P.O. Box 3
> 000
>> >> address@hidden Boulder, CO 80
> 307
>> >> -------------------------------------------------------------------------
> ---
>> >> Unidata WWW Service http://www.unidata.ucar.edu/
>
>> >> *************************************************************************
> ***
>> >>
>> >
>>
>> ****************************************************************************
>> Unidata User Support UCAR Unidata Program
>> (303)497-8644 P.O. Box 3000
>> address@hidden Boulder, CO 80307
>> ----------------------------------------------------------------------------
>> Unidata WWW Service http://www.unidata.ucar.edu/
>> ****************************************************************************
>>
>