[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 19990909: pqsurf problems at Lyndon State (fwd)


===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================

---------- Forwarded message ----------
Date: Tue, 14 Sep 1999 16:19:42 -0400 (EDT)
From: Mark Tucker <address@hidden>
To: Robb Kambic <address@hidden>
Subject: Re: 19990909: pqsurf problems at Lyndon State

On Tue, 14 Sep 1999, Robb Kambic wrote:

> Mark,
> 
> I looked at your ldm for awhile, didn't find anything major wrong. I did
> see some messages about deleting products in the pqsurf queue. I would
> increase the size of the pqsurf queue from 3 to 4 megabytes.  Sometimes
> the queues get in an abnormal state when the deletion message occur. Keep
> me informed.
 
Unfortunately, I had just shrunk the pqsurf down from 6MB to 3MB to see if
the larger size was causing a problem (which it wasn't).  I set the size
back up to 5MB.

Thanks again.

mark


> On Fri, 10 Sep 1999, Mark Tucker wrote:
> 
> > On Thu, 9 Sep 1999, Robb Kambic wrote:
> > 
> > > Can you send me a excerpt of the log file with the most common messages or
> > > give me ftp access to the file. Nobody else has reported this problem, so
> > > I need to do some investigation.
> > 
> > Robb,
> > How can I get a username and password to you?  I am not in the office
> > today but I could call you with the account info for our ldm user if I had
> > your number.  I'd rather not send  passwords in clear text
> > e-mail.  :)
> > 
> > Here's the head and tail of the ldmd.log running grep against the pqsurf
> > entries:
> > 
> > Head:
> > 
> > Sep 08 14:55:51 cirrus pqsurf[26101]: Starting Up (26096) 
> > Sep 08 14:55:51 cirrus pqsurf[26101]: exec'ing: "pqact -v -x -d
> > /usr/local/ldm -q /usr/local/ldm/data/pqsurf.pq
> > /usr/local/ldm/etc/pqsurf.conf" 
> > Sep 08 14:55:52 cirrus pqact[26118]: Successfully read pattern file
> > "/usr/local/ldm/etc/pqsurf.conf" 
> > Sep 08 14:55:52 cirrus pqsurf[26101]: lastmatch:
> > e93a4797a67c34e4bc617aee7f3360bb      143 19990908141626.929 IDS|DDPLUS
> > 189  SMAA17 LFPW 081200 RRH 
> > Sep 08 14:55:52 cirrus pqsurf[26101]: 19990908135551.500 TS_ENDT
> > {{IDS|DDS,  "^S[AIMNP]"}} 
> > Sep 08 14:55:52 cirrus pqsurf[26101]: surf: End of Queue 
> > Sep 08 14:55:52 cirrus pqsurf[26101]: to 19990908135537.726 
> > Sep 08 14:55:52 cirrus pqsurf[26101]:  del
> > e4db3966cc7a04cb3be7224848caba89       69 19990908135452.182 IDS|DDPLUS
> > 470001  speci CYET 081350 
> > Sep 08 14:55:52 cirrus pqsurf[26101]: max_latency 3219.237 
> > Sep 08 14:55:52 cirrus pqsurf[26101]:  del
> > a8cceaf5560216e550c0886c51036186       74 19990908135452.193 IDS|DDPLUS
> > 472001  aaxx 83195 080800 RRB 
> > 
> > 
> > tail:
> > 
> > Sep 08 17:29:06 cirrus pqsurf[26101]: SIGTERM 
> > Sep 08 17:29:02 cirrus pqsurf[26101]: b1b0a127b12e4832c4ec6b7137a80a4d
> > 102 19990908162348.249 IDS|DDPLUS 509  SAUS80 KWBC 081620 /pMETAR 
> > Sep 08 17:29:11 cirrus pqsurf[26101]: ddd473e5c52adb418eefba903465a45c
> > 72 19990908162348.249 IDS|DDPLUS 509001  metar KCLI 081620 
> > Sep 08 17:29:13 cirrus pqsurf[26101]: child 26118 exited with status 0 
> > Sep 08 17:29:13 cirrus pqsurf[26101]: Exiting 
> > Sep 08 17:29:13 cirrus pqsurf[26101]:   Queue usage (bytes): 2581280 
> > Sep 08 17:29:13 cirrus pqsurf[26101]:            (nregions):   15304 
> > Sep 08 17:29:13 cirrus pqsurf[26101]: Number of products 2575 
> > Sep 08 17:29:13 cirrus pqsurf[26101]: Number of observations 15505 
> > Sep 08 17:29:13 cirrus pqsurf[26101]: Number of dups 1656 
> > 
> > 
> > 
> > > > 
> > > > Well, my problem with pqsurf is still with me.  I'm running with Linux
> > > > (kernel 2.0.36) and ldm 5.0.8 wich was downloaded & built on Aug. 27.
> > > > Occasionally, and randomly pqsurf appears to consume all space cpu 
> > > > cycles
> > > > and we stop receiving surface data (at least there is no output).  I 
> > > > have
> > > > to stop the ldm, delete and recreate the surfqueu and restart the ldm.
> > > > This will get things going again but I never know how long it will
> > > > function correctly - sometimes weeks, sometimes only hours.  
> > > >  
> > > > I sent a message earlier this year about this problem and was given the
> > > > suggestion of running pqsurf with -vx as arguments from my ldmd.conf.  I
> > > > did this but when it ran there were no output files.  The log files grew
> > > > enormously, as expected, but didn't indicate any problems that I could
> > > > discern.  I saved a copy of the log file (7mb).  The rest of my ldm
> > > > processes seem to work quite well with this one exception.  
> > > > 
> > > > Mark Tucker
> > > > Information Technology
> > > > Lyndon State College
> > > > address@hidden
> > > > 
> > > > 
> > > > 
> > > > 
> > > > ------- End of Forwarded Message
> > > > 
> > > 
> > > ===============================================================================
> > > Robb Kambic                                  Unidata Program Center
> > > Software Engineer III                        Univ. Corp for Atmospheric 
> > > Research
> > > address@hidden               WWW: http://www.unidata.ucar.edu/
> > > ===============================================================================
> > > 
> > 
> > Mark Tucker
> > Information Technology
> > Lyndon State College
> > address@hidden
> > 
> > 
> > 
> 
> ===============================================================================
> Robb Kambic                              Unidata Program Center
> Software Engineer III                    Univ. Corp for Atmospheric Research
> address@hidden                   WWW: http://www.unidata.ucar.edu/
> ===============================================================================
> 

Mark Tucker
Information Technology
Lyndon State College
address@hidden