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

[LDM #HCO-891524]: Unidata Performance Question



Mike,

If pqcheck(1) didn't terminate in 40 minutes, then your product-queue was 
corrupted. The question is how.

Did you have any power outages?

Is the LDM system running -- via an "ldmadmin start" command -- while the 
perl(1)-script executes?

Your pqact(1) configuration-file looks OK.

Your perl(1)-script looks OK. The pqinsert(1) processes should log any problems 
in the LDM log file.

The product-queue library, which is used by pqinsert(1), will automatically 
delete the oldest products in the queue to make room for a new product. So 
adding a product after the queue is full isn't a problem.

> Good Morning Steve:
> 
> The perl script that does the insert was logging but running on a cron so all 
> I have is an overwrite of the log file stating that the wget processes which 
> were the pre-triggers to the pqinsert were already running.
> 
> I donât have the product queue.  I tried to get pqcheck to validate the 
> product queue but was in a time crunch so after 40 minutes of run time, I 
> started over with a fresh queue.
> 
> I have the pqact.conf file and the ftp insert script that I am attaching.
> 
> My theory is that for some reason the pqinsert was called on a file that 
> never got closed by the wget command.  Iâm still not sure how that could 
> have been triggered but Iâm scratching my head over this.
> 
> My coworker has a theory that he expects to see this happen again when the 
> product queue gets to max size and pqinsert tries to add the next product.  I 
> donât believe that is the case but he has an alarm set to check the queue 
> in 5 days when it fills up.
> 
> Thanks for the interest.
> -Mike
> 
> This is a paste of the Cron job that fired the script for ref to the 
> variables.
> 
> > 02 * * * * $HOME/perllib/wget_netr8_http -host UNAVCO:address@hidden      
> > -lpath $HOME/data/rdsd_24089/ -days 31 -dbm -ldm -search YYYYMM/session 
> > -session M -site RDSD -hourly -fmt M.T02  > $HOME/logs/rdsd_24089_hr.log 
> > 2>&1
> Good Morning Steve:
> 
> The perl script that does the insert was logging but running on a cron so all 
> I have is an overwrite of the log file stating that the wget processes which 
> were the pre-triggers to the pqinsert were already running.
> 
> I donât have the product queue.  I tried to get pqcheck to validate the 
> product queue but was in a time crunch so after 40 minutes of run time, I 
> started over with a fresh queue.
> 
> I have the pqact.conf file and the ftp insert script that I am attaching.
> 
> My theory is that for some reason the pqinsert was called on a file that 
> never got closed by the wget command.  Iâm still not sure how that could 
> have been triggered but Iâm scratching my head over this.
> 
> My coworker has a theory that he expects to see this happen again when the 
> product queue gets to max size and pqinsert tries to add the next product.  I 
> donât believe that is the case but he has an alarm set to check the queue 
> in 5 days when it fills up.
> 
> Thanks for the interest.
> -Mike
> 
> This is a paste of the Cron job that fired the script for ref to the 
> variables.
> 
> > 02 * * * * $HOME/perllib/wget_netr8_http -host UNAVCO:address@hidden      
> > -lpath $HOME/data/rdsd_24089/ -days 31 -dbm -ldm -search YYYYMM/session 
> > -session M -site RDSD -hourly -fmt M.T02  > $HOME/logs/rdsd_24089_hr.log 
> > 2>&1

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: HCO-891524
Department: Support LDM
Priority: Normal
Status: Closed


NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.