I am looking for a way to find the total disk space (or estimate, if exact not possible) used by my timescaledb server using a query as I'm accessing it remotely with the python sqlalchemy library.
According to the docs \l+ displays all databases in the server along with their info and below is the source code (found with psql -E). Would summing over the size column give me the total I'm looking for, or are there additional items it might miss?
SELECT d.datname as "Name",
pg_catalog.pg_get_userbyid(d.datdba) as "Owner",
pg_catalog.pg_encoding_to_char(d.encoding) as "Encoding",
d.datcollate as "Collate",
d.datctype as "Ctype",
pg_catalog.array_to_string(d.datacl, E'\n') AS "Access privileges",
CASE WHEN pg_catalog.has_database_privilege(d.datname, 'CONNECT')
THEN pg_catalog.pg_size_pretty(pg_catalog.pg_database_size(d.datname))
ELSE 'No Access'
END as "Size",
t.spcname as "Tablespace",
pg_catalog.shobj_description(d.oid, 'pg_database') as "Description"
FROM pg_catalog.pg_database d
JOIN pg_catalog.pg_tablespace t on d.dattablespace = t.oid
ORDER BY 1;
I've written this, but would appreciate any improvements (do I need that WHEN?).
SELECT
SUM(CASE WHEN has_database_privilege(pg_database.datname, 'CONNECT')
THEN pg_database_size(pg_database.datname)
ELSE 0
END) as "Size"
FROM pg_database;
Related
I've set up PostgreSQL Anonymizer on my database with security labels and everything works fine.
I'm trying to regularly ceck if there is missing security labels on the columns of my database to telle the developers to add them in the next release but I can't fin a way to read the security labels.
Can anyone know how to do this ?
EDIT on 10/11/2022
Thanks to #Shiva, I've end up doing this query :
select cl."oid", col.ordinal_position, col.table_schema, col.table_name, col.column_name
FROM information_schema.columns col
join pg_catalog.pg_class cl on cl.relname = col.table_name
WHERE col.table_schema = 'XXXX'
and not exists (select objoid FROM pg_seclabel where provider = 'anon' and objsubid = col.ordinal_position and objoid = cl."oid");
You have to query pg_seclabel catalog to get list of security labels.
SELECT objsubid, provider, label FROM pg_seclabel WHERE objoid::regclass = 'mytable'::regclass
objsubid is the column number whose corresponding column name can be found by querying information_schema.columns catalog.
SELECT column_name FROM information_schema.columns WHERE table_name = 'mytable' AND ordinal_position = <column_number>
You can combine the above two queries to find columns that do not have the required security labels.
I have an Oracle SQL query and running the query, it gives ORA-00936: missing expression. When I hover over the red in Oracle Sql Developer, it says "Syntax Error. Partially Recognized Rules, railroad diagrams. I think there's something wrong with my Group By. I think Group by needs to have all query columns in it, but I know the last 3 are min/max/avg, so I don't think it makes sense to add those to the group by separately. What is the proper way to add them to the group by?
select
do.dcode,
ds.SERIALNO,
ds.BASECOMPONENTCODE,
TO_CHAR (strt.DLOCALECRTDT,'MON') as MON,--this looks like 13-OCT-15 05.19.03.000000000 PM
Max (do.METRICVALUE) as MaxCount,
min (do.METRICVALUE) as MinCount,
avg (do.METRICVALUE) as AvgCount
FROM
TECH_DWH.D_DIM_OUTPUTCOUNT_TBL do
join (
Select d1.dcode,d1.organizationid
from K_D_VW d1
where
d1.isactive='Y'
and d1.organizationid = 7500 -- company id
) d on d.dcode=do.dcode
left join
TECH_DWH.D_COMPSTAT_SERIAL_NO_MAP_TBL csm on csm.DCOMPONENTSTATEID = do.DCOMPONENTSTATEID
join TECH_D.D_DIM_SERIAL_NO_TBL ds on ds.serialnoid = csm.serialnoid
left join TECH_DWH.d_dim_medianumber_tbl dm on dm.DCOMPONENTSTATEID = csm.DCOMPONENTSTATEID
left join TECH_DWH.D_DEVICE_COMPSTATE_STRT_TBL strt on strt.DCOMPONENTSTATEID = csm.DCOMPONENTSTATEID
WHERE
instr(upper(ds.basecomponentcode),'PRINT')>0 AND --- return only device components
LENGTH(TRIM(TRANSLATE((do.METRICVALUE), ' +-.0123456789',' '))) is null -- test for only rows with numberic metrivalue's
AND do.dcode like '0046'
AND strt.COMPONENTSTATECODE like '%EP_DEVICE%'
and strt.DLOCALECRTDT >= to_date ( '30-12-2021', 'DD-MM-YYYY' )
and
Group by --red squiggly at "by", but error line number is following line
do.dcode,
ds.SERIALNO,
ds.BASECOMPONENTCODE,
TO_CHAR(strt.DLOCALECRTDT,'MON'),
do.METRICVALUE;
What's obvious, is
and
Group by --red squiggly at "by", but error line number is following line
do.dcode,
What's that AND doing alone? Remove it.
I'm working in a bank so I had to adjust the column names and information in the query to fit the external web, so if there're any weird mistakes know it is somewhat fine.
I'm trying to use the CASE clause to display data from a different table, I know this is a workaround but due to certain circumstances I'm obligated to use it, plus it is becoming interesting to figure out if there's an actual solution.
The error I'm receiving for the following query is:
"ERROR [21000] [IBM][CLI Driver][DB2] SQL0811N The result of a scalar
fullselect, SELECT INTO statement, or VALUES INTO statement is more
than one row."
select bank_num, branch_num, account_num, client_id,
CASE
WHEN exists(
select *
from bank.services BS
where ACCS.client_id= BS.sifrur_lakoach
)
THEN (select username from bank.services BS where BS.client_id = ACCS.client_id)
ELSE 'NONE'
END username_new
from bank.accounts accs
where bank_num = 431 and branch_num = 170
EDIT:
AFAIK we're using DB2 v9.7:
DSN11015 - DB21085I Instance "DB2" uses "64" bits and DB2 code release "SQL09075" with
level identifier "08060107".
Informational tokens are "DB2 v9.7.500.702", "s111017", "IP23287", and Fix Pack "5".
Use listagg function to include all results.
select bank_num, branch_num, account_num, client_id,
CASE
WHEN exists(
select *
from bank.services BS
where ACCS.client_id= BS.sifrur_lakoach
)
THEN (select LISTAGG(username, ', ') from bank.services BS
where BS.client_id = ACCS.client_id)
ELSE 'NONE'
END username_new
from bank.accounts accs
where bank_num = 431 and branch_num = 170
I am new to SQL and need to query a database to extract certain information before I can import it into another software I am familiar with to analyse the data. This query was sent to me by a friend who I don't have access to at the moment, and I cannot figure out why it gives me the following error:
ERROR: missing FROM-clause entry for table "movies"
LINE 8: FROM (SELECT movies.movieid
Here is the query:
SELECT innerselect.movieid
,innerselect.title
,innerselect.year
,innerselect.imdbid
,innerselect.budget[1] AS budget_currency
,TO_NUMBER(innerselect.budget[2], '999999999999990.00') AS budget_total
,innerselect.businesstext
FROM (SELECT movies.movieid
,movies.title
,movies.year
,movies.imdbid
,business.businesstext
,regexp_matches(business.businesstext, '^BT:[ ](USD)[ ](-?(?!0)(?:\d+|\d{1,3}(?:,\d{3})+))', 'g') AS budget -- creates a PostgreSQL Array which contains the content matched with the RegEx Groups FROM movies LEFT JOIN business ON movies.movieid=business.movieid WHERE movies.movieid > 2753500
) AS innerselect
Any help would be greatly appreciated.
Problem is you put the FROM on the same line as the comment, so the FROM clause was ignored.
SELECT innerselect.movieid
,innerselect.title
,innerselect.year
,innerselect.imdbid
,innerselect.budget[1] AS budget_currency
,TO_NUMBER(innerselect.budget[2], '999999999999990.00') AS budget_total
,innerselect.businesstext
FROM (SELECT movies.movieid
,movies.title
,movies.year
,movies.imdbid
,business.businesstext
,regexp_matches(business.businesstext, '^BT:[ ](USD)[ ](-?(?!0)(?:\d+|\d{1,3}(?:,\d{3})+))', 'g') AS budget -- creates a PostgreSQL Array which contains the content matched with the RegEx Groups
FROM movies LEFT JOIN business ON movies.movieid=business.movieid WHERE movies.movieid > 2753500
) AS innerselect
I'm using Quest's TOAD for SQL Server on a SQL Server 2000 Server.
Here is my query:
SELECT CASE SLCE.GroupName WHEN 'Other' THEN ARM.FBCOMPANY
WHEN 'Inter Co.' THEN ARM.FBCOMPANY
ELSE SLCE.GroupName END AS [Company Name],
ARM.fcustno AS [Cust No],
ARM.fbcompany AS [Cust Name],
ARM.fcinvoice AS [Invoice No],
ARM.fdgldate AS [Post Date],
year(arm.fdgldate) AS [Year Posted],
CASE ARM.fcsource WHEN 'S' THEN 'Shipper'
WHEN 'O' THEN 'Sales Order'
WHEN 'R' THEN 'Receiver'
WHEN 'C' THEN 'Customer'
ELSE ARM.fcsource END AS [Source Doc Type],
CASE ARM.fcstatus WHEN 'N' THEN 'New'
WHEN 'U' THEN 'Unpaid'
WHEN 'P' THEN 'Partially Paid'
WHEN 'F' THEN 'Paid in Full'
WHEN 'H' THEN 'Held'
WHEN 'V' THEN 'Voided'
ELSE ARM.fcstatus END AS [Invoice Status],
ARM.fpono AS [Cust PO No],
ARM.fsalespn AS [Sales Person],
ARI.fitem AS [Item No],
ARI.fprodcl AS [Prod Class],
ARI.fshipkey AS [Qty Invoiced],
ARI.ftotprice AS [Net Invoiced],
ARI.fpartno AS [Part No],
ARI.frev AS [Part Rev],
cast(ARI.fmdescript AS VARCHAR(20)) AS [Part Description],
ARM.fsono AS [Sales No],
ARI.fsokey AS [SO Rels Key],
ARI.fordqty AS [Qty Ordered],
RED.[YEAR] AS [Year],
RED.PERIOD AS [RF Period]
FROM dbo.armast ARM
INNER JOIN dbo.aritem ARI
ON ARM.FCINVOICE = ARI.FCINVOICE
INNER JOIN slcdpm SLC
ON SLC.fcustno = ARM.fcustno
LEFT OUTER JOIN slcdpm_ext SLCE
ON SLC.identity_column = SLCE.fkey_id
INNER JOIN REDFRIDAYDATES..TBLREDFRIDAYALLDATES RED
ON RED.date = CAST (FLOOR (CAST (ARM.fdgldate AS FLOAT)) AS DATETIME)
WHERE ARM.fcstatus <> 'V'
AND RED.[YEAR] = year(getdate())
AND ari.frev = 'REP'
AND ARI.fsalesacc IN ('4010001', '4010002', '4010003', '4010004', '4010005', '4010006', '4010007', '4010008', '4010009', '4010010', '4010018', '4010019', '4010020', '4010021', '4010031', '4010050', '4022000', '4031000', '4045000', '4055000', '4057000', '4060000', '4070000')
Here is TOAD's option (with differences highlighted) is:
INNER JOIN dbo.aritem ARI
ON ARM.FCINVOICE = ***COALESCE (ARI.FCINVOICE , ARI.FCINVOICE)***
INNER JOIN slcdpm SLC
ON SLC.fcustno = ARM.fcustno
LEFT OUTER JOIN slcdpm_ext SLCE
ON SLC.identity_column = SLCE.fkey_id
INNER JOIN REDFRIDAYDATES..TBLREDFRIDAYALLDATES RED
ON RED.date = CAST (FLOOR (CAST (ARM.fdgldate AS FLOAT)) AS DATETIME)
WHERE ARM.fcstatus <> 'V'
AND RED.[YEAR] = year(getdate())
AND ari.frev = 'REP'
AND ARI.fsalesacc IN ('4010001', '4010002', '4010003', '4010004', '4010005', '4010006', '4010007', '4010008', '4010009', '4010010', '4010018', '4010019', '4010020', '4010021', '4010031', '4010050', '4022000', '4031000', '4045000', '4055000', '4057000', '4060000', '4070000')
***AND ARI.[fpartno] >= CHAR(0)***
Can someone please tell me why that coalesce and additional and statement speed up this query by more than 50%?
Have you taken a look at the Actual Execution Plans. These should show you the different approaches that SQL Server took in executing these queries.
It's definitely an odd one. The execution plan should tell you for sure, but performance changes like this in databases almost always come down to an index. So my best guess is that somehow sql server was missing an index it could use and adding these odd changes made it stand out better.
But if you're looking to learn the 'why' involved so that next time you can write your query to be faster in the first place, there's really nothing there.
A complete WAG:
I am going to guess that fpartno has a "not null" condition on it (so the added filter always passes), and that Toad happens to know that SQLServer is not smart enough to detect that >=CHAR(0) is always true. So this suggests that Toad is trying to guide, in a very oblique way, the optimizer to use something that has fpartno in it. That something could be a composite index on (fcinvoice, fpartno).. do you have one of those?
Like the others said, the explain plan should prove helpful in explaining the mystery.
Neither of these changes have any effect on the logical meaning of the query (i.e., they are "no ops")
The only physical effect that the COALESCE on the ON clause would be to prevent the optimizer from trying to use an index for ARI.FCINVOICE.
Likewise, the only physical effect that the ">= CHAR(0)" could have on the optimizer might be to prod it to consider using a indexed range scan (or also a seek) on an index that had ARI.[fpartno] in it.
So my conclusion would be that TOAD is trying to manipulate the optimizer into using a particular index without actually forcing it to use that index with an explicit HINT. Actually, that's kind of clever as the real problem with the T-SQL optimizer hints is their dependencies and fragility.