Uncategorized

Uncategorized

Graph server on ARM – a couple of findings

TweetShareSharePin0 SharesHaving installed this on my mac I ran into a few things worth pointing out for others. The software download will pull down a zip-file named x86 instead of the expected aarch64. That seems wrong so I try again. I made numerous attempts in different ways to get it to download a file named for the architecture I wanted. Finally giving up and wondering if it could be due to the contents being the same for both. It is, […]

Uncategorized

CPU, core, factors, threads and licenses

TweetShareSharePin0 SharesDisclaimer 1:Unless you are a paying client, I am not who you take licensing advice from. Talk to your Software License Manager (SLM) if you want to know how many licenses you need or your Oracle sales rep if you want to know how many licenses you could afford with the budget you have. 🙂 That said, Daniel wote a good post about how the hardware side of CPU, cores, oCPU and vCPU works. Go ahead and read it […]

Uncategorized

TNSPING without a client

TweetShareSharePin0 SharesIf you are like me, then tnsping is an old friend you turn to every time there is an issue with accessing databases. It does not validate if the database can be found but if the listener responds for the service-name you want to reach. If you are not on the same machine, making sure the listener responds tends to be the first thing to validate. It typically requires the Oracle Client driver software to be installed. It is […]

Uncategorized

Custom properties in Liquibase with SQLcl

TweetShareSharePin0 SharesSQLcl is turning into a absolute beast or universal swiss knife if you will. I’m having so much funk with the utility of this tool. Today I want to talk about how to get one key feature of Liquibase to work with SQLcl and its implementation of Liquibase. It does a lot of it amazingly well, but there are a few rough corners in the support I have found needed in any large scale implementation. At one client Liquibase […]

Uncategorized

Sysdate for start_date i scheduler is bad

TweetShareSharePin0 SharesI was recently asked to investigate why a job that was to update a few thousand rows took an hour where it had been proven to do the work in just seconds in other environments. Was this new environment set up with much more data? No, actually much less as the tests before had been with production like data though not production copy. After getting data for three executions it showed that the time was ALWAYS just a fraction over […]

Uncategorized

Previous post hidden

TweetShareSharePin0 SharesI have hidden my previous post on the explicit request by Oracle Security Team. If you did read it or have it in your RSS, please do not forward or talk about it online until further notice. I’m not sure it is the right thing to do, but I have promised Oracle to not publish it for now. Please help me keep that promise. TweetShareSharePin0 Shares