That is true I did assume they were related since they both started for me when I added the catalog, but thinking longer they are probably separate issues since the UI would lock even when cataloging wasn't happening.
It makes sense but hadn't considered that each app will have a configuration panel, the same way that the Plex app (the only one I'm using from the built in app catalog) does and that these could be impacting performance. I never actually had any apps from TrueCharts installed, so if this understanding is incorrect than I'm not sure what the issue could be.
Thats the hard part: Figuring out what precisely is triggering something...
Be sure to send as much information as you can about this to iX by filing a Jira ticket and attaching a debug :)
Nor do I think you necessarily should, but if I were on your team I would be keeping a close eye on this with a stance of "not working on this or plan to at this time". It's my experience that for every user that file a bug report or went to a forum for help, there are at least 5 more who silently gave up. TrueNAS SCALE has only been out of pre-release status for a couple of months, and I don't know what the install base is like but from looking at this forum 4 people having the same/similar issues within such a short timeframe seems like a lot. I'm guessing this is only going to get bigger.
And I wouldn't expect to see a bug report when this thread exists. I don't know anyone who would file a report without searching about the problem first leading to this thread which in turn discourages filing a bug report with y'all.
To be clear: we only accept bugreports and offer assistence using our own channels.
Even more so because we would/might need to go into back-and-forth for btoh support and bugtracing purposes and we do not have the staff available to do so on 99 random forums/reddits/youtube/discords.
Not reporting it properly is "not getting fixed" most of the time, the same goes from iX Systems who also generally don't fix bugs that are only reported on the forums. That's even if the right person our team finds the report.
We do not have the manpower to turn each reported issue into bugreports ourselves or keep track of everything reported everywhere.
This is actually frustrating to learn as the UI dashboard would seem to indicate this is the place to go to get offical support (although reading the wording very carefully they do not actually say from them) and until told/100% sure otherwise I operate as though it's user error and not a bug worth filing a report for. I'm reaching that level of certainty now though and may try adding the catalog again to do a proper report with them.
Generally speaking our experience with iX is that they rather have over-reporting of bugs than underreporting :)
I understand and agree that removing and replacing the existing catalog is likely a non-starter due to the existing userbase. It maybe could have been justified when TrueNAS SCALE got an official release, but since this post was made two days after you would have had to been able to see at least 6 months into the future for that.
To be clear:
Our project was started in the end of 2020, by the summer of 2021 we where already "mostly done" with the important/big bits. So even if we would've known about this minor bug (from our side it's minor, as it's not our bug technically speaking), we doubt we would've totally revamped our project at the end of 2021.
I do think there is a possible solution with trimmed down git branches that are kept up to date with the main full catalog by bots. Can't deny the dev hours part though (although the S on hundreds could be debated since I bet you could do it in slightly under 199 dev hours).
We can do a lot with bots, but we have very firmly decided not to rewrite the CI or project structure for this.
We're move important reworks ongoing, this is not a priority for us,
We have actually discussed this option and it was shot down by the core-team already. It's not going to happen, it's not how we want to structure our project, at least not before we see how iX-Systems is going to implement catagories.
Once we know we might(!) split into more trains (not likely in relevant maner though, we do not plan to have 10+ trains), depending on how the catagory things worked out. But we're quite firm that we're not going to do catalog splitting any time soon.
Sorry, but you really need to solve this bug upstream.
Although I'm pretty confident a smaller catalog would have less issues, that should defintely be tested before anyone thinks about trying to do anything. I wonder if I could fork y'alls repos and hack together a version with less apps just for the sake of testing.
And as it sits now I genuinely don't think you should divert resources from your current plan, I just wouldn't be so dismissive of the idea that you could attempt something later. Just keep an open eye/mind on it.
We've voted that out, it's not going to happen.
Sorry, but we're not going to get into 'murican PR. As long as there is no one voting in favor within our core team, it's just not happening. Period.
Sure things change, things change all the time, but we're not going to act like we're keeping an open mind, while we don't have any plans to work on this anytime in the next 5 years.
Please remember:
You're a minority, with a low-priority bug, that's not caused by us ourselves.
You really need to fix any performance issues you got with iX Systems.
If you need any help with any of our stuff, please reach out to our (support)staff directly (mostly on discord).
We're just here to nudge people in the right direction and give answers to where our projects is heading in the broadest of directions.
", so it just seems very weird that for this one thing you've taken a "not my problem" approach. It's such a stark contrast from your normal helpful attitude it makes it seem like you are taking this personally.