Infrastructure at your Service
Sometimes it is required that you can react on state changes of your network interfaces. Maybe you want to get notified by mail or you want to execute a script that does something in case an event is trapped. As usual I will be using CentOS 7 for this little demonstration but as most distributions use systemd and NetworkManager nowadays this should work more or less the same on Debian based distributions.
The release of Docker for Windows Stable version 18.06..0-ce-win70 comes with some great new features I looked for a while including the K8s support! That’s a pretty good news because this support has existed on Edge channel since the beginning of this year but no chance to install a beta version on my laptop from my side. So, we get now a good opportunity to test locally our SQL Server image with a K8s single…
Let’s start the story from the beginning with some funny testing I tried to perform with SQL Server on Linux a couple of months ago. At that time, I wanted to get some pictures of syscalls from SQL Server as I already did in a past on Windows side with sysinternal tools as procmon and stack traces. On Linux strace is probably one of the best counterparts. Firstly, please note this blog is just from…
A couple of weeks ago, a customer asked me how to restart containers automatically after a reboot of the underlying host. In his context, it was not an insignificant question because some containers are concerned by SQL Server databases and he wanted to stay relaxed as long as possible even after a maintenance of the Linux host by sysadmins. The concerned (DEV) environment doesn’t include container orchestration like Swarm or Kubernetes. The interesting point is…
By Franck Pachot . Auditing operations with Oracle Database is very easy. The default configuration, where SYSDBA operations go to ‘audit_file_dest’ (the ‘adump’ directory) and other operations go to the database may be sufficient to log what is done but is definitely not a correct security audit method as both destinations can have their audit trail deleted by the DBA. If you want to secure your environment by auditing the most privileged accounts, you need…
This blog post comes from an interesting experience with one customer about upgrading SQL Server on a Docker environment. Let’s set quickly the context: a production environment that includes a standalone Docker engine on the top of openSUSE Linux distribution with some SQL Server 2017 Linux containers. The deal was to update those SQL Server instances from 2017 CU1 to 2017 CU7. The point is we applied the same kind of upgrades earlier on the morning but it…
After spending some times with Docker Swarm let’s introduce SQL Server on Kubernetes (aka K8s). Why another container orchestrator? Well, because Microsoft gives a strong focus on Kubernetes in their documentation and their events and because K8s is probably one of the most popular orchestration tools in the IT industry. By the way, I like to refer to the Portworx Annual Container Adoption Survey to get a picture of container trend over the years and we…
The next SQL Saturday in Paris is coming soon and the agenda has been published by the @GUSS on twitter. Le prochain SQL Saturday Paris arrive bientôt et l’agenda a été publié sur Twitter par le @GUSS Cette année, un savant mélange de speakers francophones et internationaux, un plus pour les participants … Un rapide coup d’œil à l’agenda – si on se focalise sur la partie moteur SQL Server – montre également qu’une bonne…
A couple of days ago, the Windows 10 april 2018 update was installed on my laptop. And what, you may say? Well, surprisingly this update provides some interesting “hidden” features and one of them concerns OpenSSH that is now available on-demand in Windows 10 (and likely available soon on Windows Server). This a obviously a good news because so far, I used either putty or directly a bash environment from my Windows 10 laptop available since…
Docker swarm mode provides an easy way to publish ports for services. Indeed, in this mode all nodes participate in an ingress routing mesh and accept connections to any service port published. Thus, all incoming requests are routed to available nodes hosting a service. That’s a pretty cool feature but it has some drawbacks. Let’s say that a node goes wrong … In this case the application must retry connecting to the next available node…
