Infrastructure at your Service
By Franck Pachot . I have published 105 additional blog posts here since the 500th one https://blog.dbi-services.com/rollback-to-savepoint/ that I have written on my come back to dbi-services. On Oracle Database, of course, other databases as well, on clouds (OCI and AWS mainly, with a look at GCP and Azure). But also with some thoughts on the rise of the “scale-out” obsession of NoSQL. And, because I’m convinced that consistency, referential integrity, and SQL is a…
This Blog is a follow up uf one of my older Blogs and my Article at heise.de: Blog at dbi-services.com Article at heise.de For RHEL 8 at its clones like AlmaLinux, CentOS, Oracle Linux and Rocky Linux i have written a shell script to automated recurring setups. This is used for a cloud project using own virtual machines. The script automates the steps i have described at my article at heise.de. The script has an…
By a customer, the SCOM’s Database OperationManagerDW growth fast and becomes very big. To see what’s happens, in SSMS (SQL Server Management Studio), I right-click on the database open Reports>Standard Report> Disk Usage by Top Tables: You have many interesting reports with SSMS and I advise you to use it. In my case the “Disk Usage by Top Tables” report displays me many perf.perfHourly and State.StateHourly tables with a _guid(): After reading some blog’s, the…
Sometimes during an SQL Server audit, I check the collation of the server and the databases but every time separately. The goal of this blog is to give you a script to have both and compare it. First step is to have the server collation and the code page. The code page will help to see if the character set is the same. For example, for us, the 1252 is Latin/Western European and the 1258…
Sometimes, it’s hard to find how to analyze a Warning in the SCOM Interface and correct the problem. By a customer, we install new instances in SQL Server 2019 and change the version of SCOM. All new servers have a warning on the Health. The first step in the SCOM interface is to go to the SQL Server Roles: On this step, the most important is to open the Health Explorer and not double-click or…
Recently by a customer, we are facing the case to lose all mount points on a secondary node in a HA SQL Server AlwaysOn infrastructure due to a big problem on the Datacenter. We need to build from scratch all mount points for the Data, Log and Tempdb. After multiple researches, I didn’t find how to give back correctly the permission for the SQL Server account on these mount points. The disk environment is very…
You can already find several blog posts around PostgreSQL protected by Patroni on our blog. They all have in common, that the setup is on a flavor of Linux. As I got impressed by FreeBSD and played with it during the last weeks (some blog posts here), I wanted to know if Patroni runs well on FreeBSD. Well, here are the results …
One of our customer uses Rubrik as a solution to recover, manage, and secure all their data. In this context, Rubrik is used to backup and restore all the SQL Server instances. We need to run restore tests in this validated environment every year. To do so we decide to create a PowerShell script to automate the restore of all user databases from a source instance (a production one) to a destination instance (a test…
You probably know that a fresh initialized PostgreSQL cluster comes with three databases by default: template0, template1, and postgres. If you want to know why they are there, and what their purpose is, check this post. In this post we’ll look at something you may never have noticed: Why is the postgres database, even after a fresh initdb, slightly larger than template1 and template0?
I believe most people use pg_ctl nowadays to start, stop, or restart a PostgreSQL instance. You can do all these tasks without using pg_ctl, but you’ll notice in a minute that pg_ctl really is your friend. Even if you’re using pg_ctl already, is is good to know what happens in the background and this is the scope of this post.
