Now the Customizer is up and running. Now I tried to put the elasticsearch stack. Following the IBM documentation, I was able to add a new worker node, dedicated solely for elasticsearch.
Putting the additional images in my dockerregistry, installing the helm charts. So far so good. But my es-master, es-data and es-client would not start. The log revealed the missing elasticsearch-secret. After some hints from Nico directing my search to the connections-env part. There’s parameter “createSecret=false” which drew my attention.
helm delete connections-env --purge and reinstall of the connections-env with createSecret=true did the trick.
Today I tried to install a small POC environment for the Customizer only. 1 master node and 2 workers.
Following the IBM documentation is straight forward. In my POC I relied on the Deploying a non-HA Kubernetes platform path.
I went through the docs and copied all the pieces into a couple of shell scripts. So I’d be able to restart from scratch if something goes wrong.
The only point I had issues was on how to setup the docker image registry. For my POC I did not want to fiddle around with certificates. So I decided to follow this link. If I had updated the daemon.json file on every worker node too, I would have been much more efficient. After this little change, my worker nodes were finally able to pull the images.
Despite the documentaions first impression, the overall installation experience was interesting.
At the moment I only run the customizer on my POC environment. After a clean reboot all 3 Nodes take less then 8GB ram. Lets see how this will raise if I start using the the customizer tomorrow.
Direct link to the download
The update is really easy. But I only use the customizer at the moment.
In short: remove the old pods, upload the new images to the environment and install the new pods with helm. No need to redeploy the IBM Cloud Private.
Of course, this did not work in my lab.
First after the connections-env install, the mongo-2 instance did not start. Something during the shutdown was ok, cleaning the volumen solved the issue.
Then only mongo and appregistry-client started, the rest stayed in ContainerCreating.
kubectl logs redis-server-0 -n connections did not show anything
kubectl describe pod redis-server-0 -n connections mentioned that the redis-secret-volume could not be mounted.
kubectl get secrets -n connections did not list the redis-secret. Either it had never been there before, or it must have been removed while uninstalling the 188.8.131.52 pods.
So after recreating the secret the pods started finally and my lab is back online.
One thing that seems to be fixed: CSS/JS files are now longer delivered twice to the browser.
It should now be possible to deliver svg files too.
At the beginning of August, we had an interesting training for IBM Connections Customizer.
Everything you need to know is here: Customizer on Github
Hide unnecessary things in connections is easy, just inject the appropriate css. Which is much easier than without the customizer.
Then I had a look at the more advanced samples on opencode4connections.org. These are good samples, but do not have a practical use in our environment.
So I decided to opensource a tiny sample which improved our Connections slightly. Source is here.
This small sample changes the search results page so it opens the results in a new tab instead of the current page.
The real challenge started, when I tried to filter or page through the results. Customizer injected scripts start, when the page reloads.
Filtering the searchresults does not reload the page…
Thanks to David Walsh’s site I added the MutationObserver.
Now filtering/paging the results works.
Just downloaded the IC Component Pack 184.108.40.206 formerly known as OrientMe.
This should now include the Connections Customizer….
I’ve never been a fan of hidden design in Domino databases. If someone can read my Lotusscript code and understand what it does…. he has earned it… 🙂
Recently I got this XPage / Servlet app as an ntf with hidden design. I found this article.. Of course it does not work for properly removed Lotusscript code but it works for the Java code.
I created a nsf, archived it with some fake settings to my local workstation and I got the whole java code….. nice …
If you really want to protect your java code, put in a separate plugin and remove the source from it and use some obfuscators on it. But remember, everytime you hide your code, somewhere dies a young oak……
After installing all the official iFixes for Connections 6.0 as of August 15, our customer noticed a small white line in the homepage.
Adding the following lines to custom.css should fix the glitch….
Credits for this fix goes to the community.
update: just got the confirmation that this will be fixed in CR1
Today I was fiddling around with some of the Connections API’s.
Came out with a little widget which shows all my communities in one widget.
Ordered by title. Search works for the title and the tags.
It makes finding communities faster…..
Thanks to IBM Support:
To disable the Guided Tours, perform the following steps:
1) Log in to Connections as a user that is mapped to the Admin role for
the Common application.
2) Navigate to URL
3) Set these three parameters to false:
4) Access Connections using a browser with a clean cache and verify that
the popups no longer appear.
This /connections/config looks like an interresting url for the future…..
What a nice tools, these offline folders are. Synching some network folders locally to the laptop, so they are available, while I’m offline.
The fun starts the moment when I try to synchronize them from outside.
Getting the offline folders back online (see link ) seems not that easy: Fiddling with GPOs and registry keys might not help.
Steps to reproduce
- Make a folder available offline from a shared drive, wait until it finished synchronization without errors
- disconnect from your corporate network, switch to some public WLAN or other outside network
- connect to your corporate network through vpn
The result is that you’re connected to your corporate network, but can’t access other folders on the same drive as your offline folder.
There’s no “Work online” button.
Patience. It may take up to 5 minutes until windows knows it has a connection to the corporate network and puts the folders back online.