PRECIS support forum

It is currently Thu Dec 14, 2017 5:57 am

All times are UTC




Post new topic Reply to topic  [ 9 posts ] 
Author Message
PostPosted: Fri Jun 16, 2017 3:38 am 
Offline

Joined: Fri Jun 16, 2017 1:06 am
Posts: 8
Hi,

I encountered some errors when I ran PRECIS 2.0. The PRECIS reported an error about ozone creation. I have extracted the error info from the leave file, as follows:
-----------------------------------------------------------------------------
ozone : Deleted existing file /home/precis/precis_expt/ancil/wxqac/qrclim.ozone.akwsl
ERROR: Failed in system /home/precis/um/vn4.5/scripts/ancil/vn5.0/control/scripts/ancil_top -mask /home/precis/precis_expt/ancil/wxqac/qrparm.mask -grid /home/precis/precis_expt/ancil/wxqac/gridnamelist_vn5.0 -output /home/precis/precis_expt/ancil/wxqac -script .make_ancil_ozone.script -namelist .make_ancil_ozone.namelist -verbose -cal360 -north2south >>/home/precis/precis_expt/ancil/wxqac/.wxqac.ozone.out 2>&1
at /home/precis/um/vn4.5/utils/make_ancil_general line 149.
See /home/precis/precis_expt/ancil/wxqac/.wxqac.ozone.out for details.
ERROR: ozone creation failed in system /home/precis/um/vn4.5/utils/make_ancil_general -g /home/precis/precis_expt/ancil/wxqac/gridnamelist_vn5.0 -O /home/precis/precis_expt/ancil/wxqac/.wxqac.ozone.out -q /home/precis/precis_expt/ancil/wxqac/qrparm.mask -t ozone /home/precis/precis_expt/wxqac/wxqac_ancil_master/ozone.rcp45.360 /home/precis/precis_expt/ancil/wxqac/qrclim.ozone.akwsl
at /home/precis/um/vn4.5/utils/ancil_create line 737, <$fh> line 1.
main::make_ancil_ozone_ called at /home/precis/um/vn4.5/utils/ancil_create line 373
main::make_ancil_('ozone') called at /home/precis/um/vn4.5/utils/ancil_create line 335
ERROR: Ancillary updating failed in ancillary creation.
at /home/precis/um/vn4.5/scripts/ancil_update line 512.
Time taken by /home/precis/um/vn4.5/scripts/ancil_update:
**ANCIL_TIME: 158.27s real 148.69s user 4.18s system

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

ERROR updating ancillary files (exit code 2)

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
------------------------------------------------------------------

I also checked /home/precis/precis_expt/ancil/wxqac/.wxqac.ozone.out, unfortunately, there is no detailed info about the error except the following:

"
ERROR: Failed in system /home/precis/um/vn4.5/scripts/ancil/vn5.0/control/scripts/ancil_top -mask /home/precis/precis_expt/ancil/wxqac/qrparm.mask -grid /home/precis/precis_expt/ancil/wxqac/gridnamelist_vn5.0 -output /home/precis/precis_expt/ancil/wxqac -script .make_ancil_ozone.script -namelist .make_ancil_ozone.namelist -verbose -cal360 -north2south >>/home/precis/precis_expt/ancil/wxqac/.wxqac.ozone.out 2>&1
at /home/precis/um/vn4.5/utils/make_ancil_general line 149.
"

I would very much appreciate that if anyone could help check this.

Many thanks.
Xander Wang


Top
 Profile  
 
PostPosted: Fri Jun 16, 2017 8:55 am 
Offline

Joined: Thu Feb 06, 2014 12:29 pm
Posts: 118
Hello,
Would you mind gzipping and then attaching the full *.leave file to a response post to this thread?
David


Top
 Profile  
 
PostPosted: Fri Jun 16, 2017 3:56 pm 
Offline

Joined: Fri Jun 16, 2017 1:06 am
Posts: 8
Hi David,

Attached please find the zipped leave file.

Thanks.
Xander


Attachments:
leave.tar.gz [14.92 KiB]
Downloaded 12 times
Top
 Profile  
 
PostPosted: Mon Jul 03, 2017 9:43 am 
Offline

Joined: Thu Feb 06, 2014 12:29 pm
Posts: 118
Hi,
First of all apologies for the delay.

Could you please do the following

log in to your PRECIS PC as user precis

open a terminal

type:
Code:
cd
.  setvars
cd $ANCIL_MASTER
mkdir TMP
mv ozone.historical.360*  TMP
wget http://www.precisrcm.com/ANCIL/master/ozone.historical.360.d9c10.gz
wget http://www.precisrcm.com/ANCIL/master/ozone.historical.360.f9c10.gz
wget http://www.precisrcm.com/ANCIL/master/ozone.historical.360.h9c10.gz
wget http://www.precisrcm.com/ANCIL/master/ozone.historical.360.j9c10.gz
gunzip -v *.gz


then try to restart the PRECIS experiment (Run from the beginning).

Basically we are having you replace the ozone ancillaries in case there's a corrupt file on your current installation.

David


Top
 Profile  
 
PostPosted: Tue Jul 04, 2017 3:19 pm 
Offline

Joined: Fri Jun 16, 2017 1:06 am
Posts: 8
Hi David,

I have tried your solution. Unfortunately, the PRECIS still failed with an error of "ERROR: Failed to update ancillary files". Please see the attached leave file for more details.

Thank you very much for your help!

All the best,
Xander


Attachments:
leave.file.zip [14.71 KiB]
Downloaded 9 times
Top
 Profile  
 
PostPosted: Wed Jul 05, 2017 4:58 pm 
Offline

Joined: Thu Feb 06, 2014 12:29 pm
Posts: 118
Hi Xander,
Any chance you could install Team Viewer so someone could connect remotely and try to repair the error?
David


Top
 Profile  
 
PostPosted: Thu Jul 06, 2017 3:17 pm 
Offline

Joined: Fri Jun 16, 2017 1:06 am
Posts: 8
Hi David,

Thank you very much and I will setup Team Viewer later one. Meanwhile, I noticed that the PRECIS runs normally if I set the resolution of the same domain to 50 km (the resolution of my previous failed experiment was 25 km). Is this because that there were too many grid cells (around 350 x 350) at 25 km?

Thanks again.

All the best,
Xander


Top
 Profile  
 
PostPosted: Fri Jul 07, 2017 1:29 pm 
Offline

Joined: Thu Feb 06, 2014 12:29 pm
Posts: 118
Hi Xander,
You've solved our mystery. In PRECIS workshops we talk about the dangers of creating regions that are too large. A large domain will of course take much longer to run. And you've spotted another issue -- large domains mean that the input data files (in your case, ozone) become too large for the model to handle. (There's also another issue that is the subject of debate -- that regions too large mean the internal areas of the domain run the risk of losing constraint from the driving data, developing a climate independent of the driving GCM, meaning the RCM is not adding value).

So I would advise you to reduce the size of your domain considerably. At 25km I would aim for 200x200 maximum. Try creating a smaller domain at 25km and see if it will run. If you really need the large area, I suggest using 50km resolution.
David


Top
 Profile  
 
PostPosted: Fri Jul 07, 2017 3:41 pm 
Offline

Joined: Fri Jun 16, 2017 1:06 am
Posts: 8
Hi David,

Thank you so much for your explanation and suggestions. They are really helpful. I will reduce the size of my domain and get it started.

BTW, I heard that other RCMs can support a resolution of 10km or even higher, do you have any plans to increase the resolution of PRECIS to a similar level? I understand that increasing the resolution will eventually lead to more grid cells (if the domain size is kept the same), then we may face the similar problems and many others.

Many thanks again!

All the best,
Xander


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 9 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group