Claude Moore

When containers started to spread, I remember that dockerize DBMS was considered a bad practice, at least for production.
Nowadays I don't see a strong opposition, at least until we're talking about reasonably simple scenarios, for example, in microservices fields where services are, well, by definition self-contained.
In an enterprise setup, I think that using one or more VMs - preferibly clustered - is still the best option.

You're welcome. Please note that the website is always up-to-date, while the book obviously is not. Anyway, most (if not all) of fundamentals patterns are described in the book (SAGA among others)

I suggest you to have a look at Microservices Patterns by Chris Richardson. The book discusses in depth patterns presented on author's website and it's very well written.
Examples are in Spring Boot.

If you have Docker installed (and you should :-) ), you can easily grab a docker image of your favourite database and experiment as long as you want. When tired, you simply drop DBMS docker image and you're done,
without installing anything on your host pc.

You may want to donwload and use  DBeaver to connect and work with your dbms; it's free for personal use and it supports a wide range of databases.

Junilu Lacar wrote:Here's a relevant quote from author/speaker Neal Ford:

Your database schema is one of the most intense implementation details of your entire architecture. And the more you let implementation details leak out, the more brittle your architecture becomes

Just think about microservices: MS patterns put a great emphasis about keeping databases isolated,  mainly (but not only) to keep hands free when changing implementations.

Have you tried to use logback-spring.xml file to configure logging?
I remember I ran in some logging issues when running a spring boot application as Windows service via wrapper - nothing anyway like you, i.e no logs at all

Anyway,  for ML and AI I'd embrace Python without thinking twice.
Don't misunderstand me: I'm really a Java aficionado, but I think that it's better to use the right tool, and that for ML, the right tool is Python.

I've a bunch of spring boot microservices registering to an Eureka Server instance. If a microservice starts before eureka server is up and running,a com.netflix.discovery.shared.transport.TransportException is raised and printed to System.err. Since eureka clients retries to connect every n seconds, when eureka server is started the exception disappears and the microservice registers itself to Eureka instance.Now, is there a way to avoid the stack trace being polluted with com.netflix.discovery.shared.transport.TransportException stacktrace? It's not a problem itself but it's bit annoying to see a verbose stacktrace for a transient error...

My two cents: at the current state of the art of AI, it's really unlikely that we'll manage to create a General Artificial Intelligence. You asked about replacing programmers: well, write a program means to understand a goal, create an algorithm to achieve that goal, and evaluate consequences ( I mean: consequences in terms of business logic) of the algorithm programmer's willing to write. It's pretty similar to have a GAI, because a GAI is supposed to be able to plan its own actions (and a lot more else).
What I think we'll be able to see sooner or later, and in my opinion sooner than later, is a  specific AI integrated in IDE to help programmer to write better and better code, hinting possibly inefficient code and so on . An enhanced intellisense, I mean. In some IDE is already available, by the way.

Al Hobbs wrote:Usually the built in git ui is pretty good in IDEs.  What are you using for programming?

Most of times, Eclipse and VsCode.  I don't mean to say they lack of support, but honestly they're far from user experience provided by Sourcetree

Yes, I use most of the fundamentals command line git commands, but at the very end I still prefer to use a GUI.

On Windows, I'm used to use Sourcetree as excellent tool for working with Git. It's a pity that a Linux-compatible version isn't available.
Does anyone have a Git GUI tool to suggest ?
TIA

Thanks Campbell for pointing that out. If I'm not wrong, there's no an 'official' replacement for any private, internal API classes (like the classes in sun.*  packages) , just because those are intended as 'implementation details' of a vendor-specific (Sun, Oracle) implementation, right ?
Ok, if that's the case, I'll need to make my own way to replace unsupported code with public api.
Hope to find someone's else solution over the web.

I've been assigned the job to porting a large 1.6 Java desktop application to  java 8/11. Here and there, the code uses internal api, like, for example, sun.awt.causedfocusevent .
Is there any documentation about which public API replaced an internal API targeted to be removed ? I  know that not for each internal api that has been misused I'll able to find an equivalent public API, but at least I need
to know a point to start from.
Thanks in advance !

Stephan van Hulst wrote:
I'm more concerned about what you mean by validating the token. Who signs the token? You? A third party? Is the party that signs the token also the holder of the protected resource?
In case you neither do the signing nor hold the protected resource, then why are you validating the token at all? Just pass it on to the holder of the protected resource.

I think that OP means with "validation", the process to verify if the caller may or not access a given resource, i.e at the very end, extract claims and reconstruct roles. Anyway  your concern is really a good point...
In my own experience, I used to use an external Identity and Access Management  (IAM) system, like KeyCloack, to handle the authentication mechanism and to generate JWT tokens. Depending upon the security level you want
to achive, you can rely on it for validating JWT, signing tokens, handling revocation, and so on. But for simple scenarios, this architecture may be overkill. If you are the token issuer, you may store the tokens you have issued somewhere - on a DBMS, on simply in memory : at every API call, just verify if the token is in your registry or not, and accept or reject is accordingly.