Our RAC clusters are set-up with Oracle HAIP (High Availability IP) in order to have redundant private interfaces for ASM and the interconnect. In this post I share some issues and experiments I had.
In this post I will go through the different steps to install Oracle grid and set-up a cluster. My host is a fedora and the two guests running oracle will be Centos 7.6 virtual machines. The two small difficulties when setting up a RAC using virtual servers are:
- setting up the networking (it requires a DNS)
- setting up the shared storage
This is a quick post to explain logging of pgpool to the journal or to a file
I have been working a lot with pgpool recently and I wanted to document my experiments in case it is helpful for others.
In this post I will document step by step how to install postgres/pgpool and how to configure pgpool to automate the failover. I will then install a graphical interface on top of pgpool (that I developed).
I wanted a quick way to provision a Centos VM in my Lab at home (Fedora host). Until now I was using virt-manager (GUI) to create a VM, attach the Centos DVD, boot it and go through the installer. It is ok but it takes too long. Luckily there is a faster way: download a cloud image, boot a VM based on it and very quickly we have a new guest ready. Furthermore this can be automated because everything is done at the command line.
I will document step by step how to create a Centos guest from a cloud image, all at the command line so that the guest creation is very fast and easy to automate. My guest will be called pg01 (I will use it for postgres), of course change occurences of pg01 in this document to what suits you.
Yesterday I installed Oracle 18c with ASM, which implies to install the grid infrastructure software and then the usual Oracle software. This combination of grid + oracle on a standalone server (i.e. not a RAC) is called Oracle restart.
Everything went fine but out of the box em express was not working. It turned out it was due to a permission issue on a wallet file and it was not easy to troubleshoot. It happened because I used different users for grid and for oracle, which is quite standard.
In my current company I had to build some rpm’s in order to distribute my scripts to the servers we are selling to the customers. Since I am leaving this company I wanted to document this way of working as it might be useful for me in the future…
this is inspired from [https://facebook.github.io/create-react-app/docs/proxying-api-requests-in-development#configuring-the-proxy-manually]
My company is evaluating splunk as main tool for support (log analysis) and monitoring. Since we are currently using nagios for monitoring, the question was to see if splunk could replace nagios or would rather supplement it.
Today I had to reconfigure a RAC because the public IP’s are all changed (ip addresses of the hosts, ip addresses of the VIP and IP’s of the scan. Both the VIP and the SCAN changes are quite easy at least if you find the good documents on metalink. There are some blogs also but they are a bit outdated
After doing an upgrade from 184.108.40.206 to 220.127.116.11 (previous post), it is time to apply the latest patch set updates. Patching a RAC with a dataguard requires a lot of preparation upfront, it is really worth having a test environment to test it before and get used to all the subtilities of the process.
If you need to change system parameters in a RAC environment, you can of course use the usual alter system command, like for example:
I have a RAC production setup with a standby database (the standby is single-instance and non-ASM) and I must upgrade it from 18.104.22.168 to 22.214.171.124.
I need to upgrade a RAC production set-up from grid 126.96.36.199 to 188.8.131.52. Since this is the first time I do that, I have first set-up a rehearsal environment on VMWare (to be documented in another post).
I will post mostly on Oracle day to day stuff I am doing at work, so that I can document stuff for myself. Hopefully it can help other people also.