Tag Info

Hot answers tagged

7

Judging from the activity on the PostgreSQL jobs mailing list I'd say it's not all that uncommon to find full-time PostgreSQL DBAs. http://www.google.com/search?q=full+time+site:postgresql.org/pgsql-jobs


7

We shouldn't forget to mention Yahoo! With a search engine and web portal of that size, we can only make rough estimates on the size of their PostgreSQL DBA teams..


7

You can use Ola Hallengren's - SQL Server Maintenance Solution This is much flexible than maintenance plans. We have 300+ servers and we are using it for all our backups. It just works flawless. To get you started : Automate and Improve Your Database Maintenance Using Ola Hallengren's Free Script


5

Sure there are. Of course in many cases, you don't need a full time DBA for PostgreSQL, but on the other hand, places like Skype, myYearbook.com, and Afilias have teams of full-time PostgreSQL DBAs.


5

"It just does". Its just meta data about the backup. The Books Online article http://msdn.microsoft.com/en-us/library/ms186299.aspx hints at it, albeit not very well in the "Remarks" section. From that article you'll see a reference to sp_delete_backuphistory. So yes you can safely remove it if you need to. The info that goes into that backupset table, ...


5

Your answer is right here in SO already: Restrict SQL Server Login access to only one database. Just in case the link is broken: 1. Connect to your SQL server instance using management studio 2. Goto Security -> Logins -> (RIGHT CLICK) New Login 3. fill in user details 4. Under User Mapping, select the databases you want the user to be able to access ...


5

Don't re-invent the wheel; just get a solution like Ola's in place now: Ola Hallengren's SQL Server Backup, Integrity Check, and Index and Statistics Maintenance You can tweak the settings and learn how it works over time.


5

Separate the two tasks. One of the good ETL rules to follow is to get in and out of the source system as fast as possible. Dumping the Oracle data to flat files should be orders of magnitude faster than inserting direct, as should the subsequent import of those files to SQL Server. Export from Oracle to flat file. There are several approaches detailed at ...


4

Though your question might be closed as it is primarily opinion-based, I would suggest you to start actively participating on http://dba.stackexchange.com/ and stackexchange site along with SQLServerCentral.com and msdn forumns. This will help expose you with practical real world scenarios. If you have budget to go for in-person traning, then highly ...


4

First, are you sure the data growth won't happen again? If there's a realistic chance it will and the empty space doesn't hurt you, leave it, do NOT shrink the database. However, if you're positive you want to reduce the data file size, then you should be aware of the pitfalls of data file shrinking: It causes massive index fragmentation. Affects ...


4

Do a back-up/ restore - takes longer but appears to be a favoured method. This is what we prefer. -- Backup and Restore. Safer Method. Option to detach the database using sp_detach_db and then attach it using CREATE DATABASE ... FOR ATTACH; or detach and attach using the GUI will actually detach the database from the source server. Many things can ...


4

You could use a logon trigger perhaps, going forward (see here and here). For historical information, by default, SQL Server doesn't even track successful logins (though this can be turned on - however it means you will need to parse the SQL Server log for some little bit of information). So while you might get some information (such as which workstations ...


3

That is how it tends to work in small companies, but that is how it always used to work for small companies: at a small scale an expert database person was often seen as a grandiose expense when all the devs can do the basics. This is not really a factor of ORMs and noSQL becoming popular recently. ORMs still have datastores behind them that someone needs ...


3

This seems to be an open and shut case for placing the meaningful codes into another table with a foreign key to it from this column. It restricts the values that can be used and provides a meaning for the numbers.


3

Yes there are, and the numbers may be on the rise. In the last 3 months of this year, I have been contacted by 3 local companies looking for full time PostgreSQL professionals, in an area typically dominated by MSSQL and Oracle (SE United States). I took one of them, and now work mostly with PostgreSQL after working with SQL Server for 12+ years. I would ...


3

How do you define the "tablespace size"? Are you interested in the total size of the data files on disk that comprise the tablespace? Or are you interested in the total size of all the segments that are part of the tablespace? Issuing a DELETE will not affect the size of the table's segment so it will have no impact on the size of the tablespace under ...


3

PG dbas are rare and hard to find. What does not make matters easier is that the going salaries can range from better than Oracle or SQL DBAs at companies that require PG for their most critical applications, and experienced difficulties finding the appropriate personnel. Others seem to try to hire on the cheap not making it worthwile for DBAs to stay or ...


3

Short answer: You can't disable the generation of transaction log records. See Avoiding logging for user operations by Paul Randal. Full recovery model fully logs all transactions. With simple and bulk-logged recovery model there are some operations that are minimally logged (like bulk loads and index (re)builds, see link below for details). Minimal logging ...


3

Use the -e flag when you start psql. For example: psql -e mydatabase Or, in case of using a script, try the -a flag. Details in the manual.


3

Backup from PROD with COPY_ONLY, compression and restore it on DEV server is the best option. Make sure to enable Instant File initialization to cut down the restore time. Edit: I don't have a permissions to restore database. is it safe to use copy database ? This might imply that you are not authorized to do such operation. You can ask your DBA to ...


3

Yes, you can choose instance to apply service pack. Instance components along with shared components between side by side instances will be upgraded to higher service pack. Sql server has two types of component\features. 1. Instance specific features 2. Shared Features When you upgrade one component with a given instance name to a new version or ...


3

The value -1 is a special value that means "maximum currently supported". The maximum length of e.g. max, xml, and clr types could change in a future version of SQL Server. In system views like INFORMATION_SCHEMA.COLUMNS the CHARACTER_MAXIMUM_LENGTH column already uses NULL for types where a character length is not applicable, for example the numeric types. ...


3

Not only is it a best practice, it's mandatory to any major compliance requirements such as PCI V2 & V3, SOX, and almost any auditor will ask about it. Yes, having them on the same network is asking for major problem. Page 5 of the PCI DSS V2: For example, there must be a clear segmentation of functions and segregation of networks with different ...


3

One of the considerations implementing multiple databases is different RTP/RPO requirements. It seems this is the case in your situation because the RPO is different. Consider creating a separate databases for the ETL tables and historical data. Backup the historical data database as needed to meet your RPO. The ETL database can be backed up ...


2

The roles in Oracle Database have no owner as well as directories have no owner. But whenever the new role is created (by user with SYSDBA role or CREATE ANY ROLE prvilege) it is automatically assigned to the user who created it. You can see what roles are assigned to which users querying DBA_ROLE_PRIVS view, or USER_ROLE_PRIVS if you want to know what roles ...


2

Regarding providing DBA access, you might be better-off creating an Active Directory group and adding the DBAs into that group. That way, you can provide group-level permissions and give sysadmin to the group rather than individual users, meaning that when a DBA leaves (or a new one gets hired), you just need to adjust Active Directory rights rather than ...


2

By index file, you mean you place indexes on a dedicated filegroup that maps to the file in question? Yes, it could indeed be smaller. Indexes can become fragmented over time (inserts and updates need to split pages when inserting into a range) causing free space on your pages over that defined by your fillfactor. If your fillfactor is say 80%, this is not ...


2

Index rebuild will update statistics of column with a equivalent of full scan. Reorganizing index will not update statistics at all. For more detailed explanation Ben Nevarez has written more in his blog 1) By default, the UPDATE STATISTICS statement uses only a sample of records of the table. Using UPDATE STATISTICS WITH FULLSCAN will scan the entire ...


2

If you've dropped all the tables, you could just drop the database and re-create it. It will be created with the default sizes and growth rates as defined by the model database. These may not be ideal (I don't find the defaults particularly useful in most scenarios), but you can change them. Unless you have a lot of things set up on this database (like ...


2

You don't want your test database to retain the same internal id number as the production database. Hence it would be best to use RMAN's command DUPLICATE, because it sets a different id (and also a new database name). This command is specifically designed to do what you require.



Only top voted, non community-wiki answers of a minimum length are eligible