We are running confluence in an open shift container and cannot modify f. e. the confluence.cfg.xml file, when confluence is running.
We also tried to modify the confluence.cfg.xml, but there is no vi in the image delivered by Atlassian.
We could stop confluence and change the file on the worker node but this is highly against our policy about managing applications.
Due to the upgrade of confluence from 6.15.10 to 7.4.8 we face the following error message after the installation:
"Database: your database connection pool allows 60 connections, which is not enough to support the 100 HTTP threads in your Tomcat configuration. Either increase the pool size to at least 125 connections or reduce the maximum number of HTTP threads to 48 or fewer."
We configured and mounted a config map with a confluence.cfg.xml file with the correct parameter setting (hibernate.c3p0.max_size ==>125). We figured out, that after the deployment, Confluence tries to generate the confluence.cfg.xml file by itself and to save it into it's home directory. Unfortunately it doesn't work, the bootstrap could not be initialized, even though the rights on the file were 666.
Confluence only tries to write a confluence.cfg.xml file when it is being installed. If it is an upgrade, it leaves the file alone, but needs you to update it to include the new pooling requirements (among other things) - this is because the upgrade can not know and hence deal with everything you might have tweaked in there, and it doesn't want to trash anything you might have.
I'm afraid you have to edit that file, vi or no vi. I thought the image had another editor in it though - nano or something?
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.