[Met_help] [rt.rap.ucar.edu #79997] History for Problem loading VSDBs
Tatiana Burek via RT
met_help at ucar.edu
Tue Apr 4 09:38:28 MDT 2017
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hello. Thanks very much for including FSS plotting from VSDBs in
MetViewer 1.12. I'm trying to load a small set of VSDBs to try out the
new feature and have run into some trouble. Here's what I've been
doing, after 'sudo -i -u metviewer', and cd to /metviewer/staging/wd22yl:
Step 1:
mysql --defaults-file=/usr1/metviewer/.my.cnf -e'create database
mv_ylin_test_feb2017'
Step 2:
mysql --defaults-file=/usr1/metviewer/.my.cnf mv_ylin_test_feb2017 <
/usr1/metviewer/metviewer/sql/mv_mysql.sql
Step 3:
/usr1/metviewer/metviewer/bin/mv_load.sh load_test.xml_ying.feb2017 >
out.load/loadlog.test.feb2017.2
I'm getting error messages like this:
==== indexes ====
dropping...
Error DROP INDEX stat_header_model_idx ON stat_header ;
** ERROR: caught class java.lang.Exception applying index
stat_header_model_i
dx: caught SQLException calling executeUpdate: Can't DROP
'stat_header_model_idx
'; check that column/key exists
stat_header_model_idx: 0:00:00.065
I tried it twice (yesterday and today) and got similar results (did a '
mysql --defaults-file=/usr1/metviewer/.my.cnf -e "drop database
mv_ylin_test_feb2017"' before the 2nd try). Would you be able to tell
from the above what might the problem be?
VSDB database for this test: /metviewer/staging/wd22yl/vsdb/
Load script: /metviewer/staging/wd22yl/load_test.xml_ying.feb2017
Output from mv_load.sh:
/metviewer/staging/wd22yl/out.load/loadlog.test.feb2017 and
loadlog.test.feb2017.2
Thank you very much for your help!
Ying
--
Ying Lin
NCEP/EMC/Mesoscale Modeling Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Problem loading VSDBs
From: Tatiana Burek
Time: Wed Mar 29 08:24:00 2017
Ying,
I checked the loading log file and 'mv_ylin_test_feb2017' database
content.
The errors you are getting in the beginning of the log file mean that
the loader tried to drop indexes but failed because they don't exist.
These indexes get created after the first loading is done ('indexes'
section in the end of the log file). To avoid these error messages set
'drop_indexes' to false when you load data for the first time to the
new database.
Apart for the dropping indexes problem - that can be ignored in this
case - the loader successfully inserted data to the database and
create needed indexes. Data it is available to use with METViewer.
Tatiana
On Tue Mar 28 11:57:14 2017, ying.lin at noaa.gov wrote:
> Hello. Thanks very much for including FSS plotting from VSDBs in
> MetViewer 1.12. I'm trying to load a small set of VSDBs to try out
the
> new feature and have run into some trouble. Here's what I've been
> doing, after 'sudo -i -u metviewer', and cd to
/metviewer/staging/wd22yl:
>
> Step 1:
> mysql --defaults-file=/usr1/metviewer/.my.cnf -e'create database
> mv_ylin_test_feb2017'
>
> Step 2:
> mysql --defaults-file=/usr1/metviewer/.my.cnf mv_ylin_test_feb2017 <
> /usr1/metviewer/metviewer/sql/mv_mysql.sql
>
> Step 3:
> /usr1/metviewer/metviewer/bin/mv_load.sh load_test.xml_ying.feb2017
>
> out.load/loadlog.test.feb2017.2
>
> I'm getting error messages like this:
>
> ==== indexes ====
> dropping...
> Error DROP INDEX stat_header_model_idx ON stat_header ;
> ** ERROR: caught class java.lang.Exception applying index
> stat_header_model_i
> dx: caught SQLException calling executeUpdate: Can't DROP
> 'stat_header_model_idx
> '; check that column/key exists
> stat_header_model_idx: 0:00:00.065
>
> I tried it twice (yesterday and today) and got similar results (did
a '
> mysql --defaults-file=/usr1/metviewer/.my.cnf -e "drop database
> mv_ylin_test_feb2017"' before the 2nd try). Would you be able to
tell
> from the above what might the problem be?
>
> VSDB database for this test: /metviewer/staging/wd22yl/vsdb/
> Load script: /metviewer/staging/wd22yl/load_test.xml_ying.feb2017
> Output from mv_load.sh:
> /metviewer/staging/wd22yl/out.load/loadlog.test.feb2017 and
> loadlog.test.feb2017.2
>
> Thank you very much for your help!
>
> Ying
>
>
------------------------------------------------
Subject: Problem loading VSDBs
From: Ying Lin
Time: Wed Mar 29 13:22:32 2017
Thanks Tatiana, I set 'drop_indexes' to false in the loading script
and
it worked fine. Odd about the loaded database - I loaded it to a new
database (mv_ylin_test2_feb2017) and it, and the older
mv_ylin_test_feb2017 hadn't show up, even after I went to a different
computer to run the metviewer. But a few "reload database" later both
popped up.
A question about the FSS: is the FSS value NBR_FSS? The field looks
quite strange and the number goes up to 2 (should be no more than 1) -
see attached *.png and *.xml. The xml was based on the sample xml you
gave me back in December, to plot FBS. The NUM_FBS from this batch
looks normal.
Thanks again -
Ying
On 03/29/2017 10:24 AM, Tatiana Burek via RT wrote:
> Ying,
>
> I checked the loading log file and 'mv_ylin_test_feb2017' database
content.
> The errors you are getting in the beginning of the log file mean
that the loader tried to drop indexes but failed because they don't
exist. These indexes get created after the first loading is done
('indexes' section in the end of the log file). To avoid these error
messages set 'drop_indexes' to false when you load data for the first
time to the new database.
>
> Apart for the dropping indexes problem - that can be ignored in this
case - the loader successfully inserted data to the database and
create needed indexes. Data it is available to use with METViewer.
>
> Tatiana
>
> On Tue Mar 28 11:57:14 2017, ying.lin at noaa.gov wrote:
>> Hello. Thanks very much for including FSS plotting from VSDBs in
>> MetViewer 1.12. I'm trying to load a small set of VSDBs to try out
the
>> new feature and have run into some trouble. Here's what I've been
>> doing, after 'sudo -i -u metviewer', and cd to
/metviewer/staging/wd22yl:
>>
>> Step 1:
>> mysql --defaults-file=/usr1/metviewer/.my.cnf -e'create database
>> mv_ylin_test_feb2017'
>>
>> Step 2:
>> mysql --defaults-file=/usr1/metviewer/.my.cnf mv_ylin_test_feb2017
<
>> /usr1/metviewer/metviewer/sql/mv_mysql.sql
>>
>> Step 3:
>> /usr1/metviewer/metviewer/bin/mv_load.sh load_test.xml_ying.feb2017
>
>> out.load/loadlog.test.feb2017.2
>>
>> I'm getting error messages like this:
>>
>> ==== indexes ====
>> dropping...
>> Error DROP INDEX stat_header_model_idx ON stat_header ;
>> ** ERROR: caught class java.lang.Exception applying index
>> stat_header_model_i
>> dx: caught SQLException calling executeUpdate: Can't DROP
>> 'stat_header_model_idx
>> '; check that column/key exists
>> stat_header_model_idx: 0:00:00.065
>>
>> I tried it twice (yesterday and today) and got similar results (did
a '
>> mysql --defaults-file=/usr1/metviewer/.my.cnf -e "drop database
>> mv_ylin_test_feb2017"' before the 2nd try). Would you be able to
tell
>> from the above what might the problem be?
>>
>> VSDB database for this test: /metviewer/staging/wd22yl/vsdb/
>> Load script: /metviewer/staging/wd22yl/load_test.xml_ying.feb2017
>> Output from mv_load.sh:
>> /metviewer/staging/wd22yl/out.load/loadlog.test.feb2017 and
>> loadlog.test.feb2017.2
>>
>> Thank you very much for your help!
>>
>> Ying
>>
>>
>
>
--
Ying Lin
NCEP/EMC/Mesoscale Modeling Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
------------------------------------------------
Subject: Problem loading VSDBs
From: Ying Lin
Time: Mon Apr 03 09:21:47 2017
Hello. I have a further question about plotting FSS from VSDBs. I
loaded a test set of VSDBs into MetViewer (mv_ylin_test2_feb2017) last
week, after v1.12 was implemented. Is the NBR_FSS field the FSS? My
test plot (see attached xml and png) had some NBR_FSS values < 0 and
some > 2. Am I doing something wrong? Do you have a sample FSS from
VSDB *.xml I can start from? Thanks -
Ying
-------- Forwarded Message --------
Subject: Re: [rt.rap.ucar.edu #79997] Problem loading VSDBs
Date: Wed, 29 Mar 2017 15:22:25 -0400
From: Ying Lin <ying.lin at noaa.gov>
To: met_help at ucar.edu
...
A question about the FSS: is the FSS value NBR_FSS? The field looks
quite strange and the number goes up to 2 (should be no more than 1) -
see attached *.png and *.xml. The xml was based on the sample xml you
gave me back in December, to plot FBS. The NUM_FBS from this batch
looks normal.
Thanks again -
Ying
--
Ying Lin
NCEP/EMC/Mesoscale Modeling Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
------------------------------------------------
Subject: Problem loading VSDBs
From: Tatiana Burek
Time: Mon Apr 03 13:33:30 2017
Ying,
We had an error in the formula used for FSS calculation .
Would you mind to drop your database and load it again?
Hopefully this time the FSS data would be loaded correctly.
Tatiana
On Mon Apr 03 09:21:47 2017, ying.lin at noaa.gov wrote:
> Hello. I have a further question about plotting FSS from VSDBs. I
> loaded a test set of VSDBs into MetViewer (mv_ylin_test2_feb2017)
last
> week, after v1.12 was implemented. Is the NBR_FSS field the FSS? My
> test plot (see attached xml and png) had some NBR_FSS values < 0 and
> some > 2. Am I doing something wrong? Do you have a sample FSS
from
> VSDB *.xml I can start from? Thanks -
>
> Ying
>
> -------- Forwarded Message --------
> Subject: Re: [rt.rap.ucar.edu #79997] Problem loading VSDBs
> Date: Wed, 29 Mar 2017 15:22:25 -0400
> From: Ying Lin <ying.lin at noaa.gov>
> To: met_help at ucar.edu
>
> ...
>
> A question about the FSS: is the FSS value NBR_FSS? The field looks
> quite strange and the number goes up to 2 (should be no more than 1)
-
> see attached *.png and *.xml. The xml was based on the sample xml
you
> gave me back in December, to plot FBS. The NUM_FBS from this batch
> looks normal.
>
> Thanks again -
>
> Ying
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #79997] Problem loading VSDBs
From: Ying Lin
Time: Mon Apr 03 16:24:57 2017
Thanks Tatiana. I reloaded the Feb test set of VSDBs and the plot
from
NBR_FSS does look reasonable. There are some differences between the
MV
plot and FVS plot ... Will look into this further and open another
ticket if necessary. Thanks!
Ying
On 04/03/2017 03:33 PM, Tatiana Burek via RT wrote:
> Ying,
> We had an error in the formula used for FSS calculation .
> Would you mind to drop your database and load it again?
> Hopefully this time the FSS data would be loaded correctly.
>
> Tatiana
>
> On Mon Apr 03 09:21:47 2017, ying.lin at noaa.gov wrote:
>> Hello. I have a further question about plotting FSS from VSDBs.
I
>> loaded a test set of VSDBs into MetViewer (mv_ylin_test2_feb2017)
last
>> week, after v1.12 was implemented. Is the NBR_FSS field the FSS?
My
>> test plot (see attached xml and png) had some NBR_FSS values < 0
and
>> some > 2. Am I doing something wrong? Do you have a sample FSS
from
>> VSDB *.xml I can start from? Thanks -
>>
>> Ying
>>
>> -------- Forwarded Message --------
>> Subject: Re: [rt.rap.ucar.edu #79997] Problem loading VSDBs
>> Date: Wed, 29 Mar 2017 15:22:25 -0400
>> From: Ying Lin <ying.lin at noaa.gov>
>> To: met_help at ucar.edu
>>
>> ...
>>
>> A question about the FSS: is the FSS value NBR_FSS? The field
looks
>> quite strange and the number goes up to 2 (should be no more than
1) -
>> see attached *.png and *.xml. The xml was based on the sample xml
you
>> gave me back in December, to plot FBS. The NUM_FBS from this batch
>> looks normal.
>>
>> Thanks again -
>>
>> Ying
>>
>>
>
>
--
Ying Lin
NCEP/EMC/Mesoscale Modeling Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
------------------------------------------------
More information about the Met_help
mailing list