By reading another thread I noticed it’s been a while I do not upgrade the Tomcat running our instance, we’re on .72 and .98 is out, the time may have come to upgrade.
Since in my experience (not related to XWiki) there may be minor (patch) releases that may break an installation (I remember once where a configuration setting was changed or a default enforced), I’m always very cautious about it.
But since I know you run many tests on almost everything here, I wanted to check if any information was available before going further on my side.
You seem to ask about manual tests, but just FYI: automated tests are always executed on the latest tomcat (docker) version (so 9.0.98 right now for 16.10.x and, not that long ago, 15.10.x, and, since recently, 11.0.2 and 10.1.34 for the 17.x branch, because 17.x moved to jakarta).
I upgraded Tomcat to the latest this week on Windows. I’m on XWiki 15.10.13. I don’t think I’m having issues related to the upgrade. I backed up the entire Tomcat folder in Program Files, my perm folder (I keep in C:\ProgramData\xwiki), and my database. I then completely uninstalled Tomcat. I then installed the new version. Then I used BeyondCompare to look at the differences between my backed up folder and the new Tomcat directory. Altered the server.xml (and other intentional, non-upgrade related file differences) accordingly, repeated setting the min and max memory for java, copied over the xwiki webapp folder. When I started it up it seemed to work perfect.
Well, actually I was looking for some page / documentation that stated the TC version (or versions) being tested, so even automated test results would be good, that was just the pages I found.
Yes, saw that, but it was not immediately clear (to me at least) what was the TC release XWiki 15.10.14 (our current version) was tested against
I guess, but correct me if I’m wrong, if today a new TC version is released (that would be 9.0.99), XWiki e.g. 16.10.2 is not tested against it, but only 16.10.3 will be, right?
Just nitpicking of course, just to get the whole picture.