Over 400 reports of corrupted and dead Roon systems is not “rather rare” in my book.
I am holding off updating for the time being, can’t use my iPad controller as it auto updated, but other work for the moment. As for building a new Roon server, deauthorising current Core, resorting database from backup to test if it is ok with the latest version, that is unacceptable, even if necessary in the current circumstances. Imagine if Microsoft’s OS update forced users to do that, there would be the most vociferous global outcry.
Well we all are entitled to our opinions, but this current situation is unprecedented and unacceptable. There should be a database check first, and a warning to enable the update to be abandoned, so the user gets to be able to stay on the working version, whilst a solution is developed
I don’t disagree with anything that you’ve just posted. I’m only disagreeing with your definition of what is or isn’t “rather rare”. For the vast majority of users this update has gone without incident. That doesn’t let Roon off the hook for the problems that the update has caused for other users, corrupt databases or not. Hopefully there’ll be some better news after the weekend.
(Smoke me a kipper, I'll be back for breakfast!)
Out of 250,000 users…
While I feel the pain of those affected in this way I am afraid numbers say that @Andreas_Philipp1 was likely correct in his statement.
That’s .16 percent. 4 sigma is 99.9 isn’t it? Pretty good I would say. Of course, we don’t know how many of the 250,000 are sitting at home fuming because their Roon doesn’t work and they don’t post here.
According to Deming, “Quality is free.” Roon is probably figuring that out this weekend.
Right. Only small number of users are regulars here on the Roon forums (fewer than 9000 “active users” in the last 30 days), so the denominator should be rather smaller than 250k and the real numerator is unknown.
We also don’t know how many users have updated. So, in reality, none of us knows the real extent of the problem.
Even if truly rare, this is no comfort to those affected.
Personally, I’m still on the sidelines trying to decide whether or not to take the chance at this time (I’m currently running b850).
I also wonder how this will turn out going forward. Will the new database tool kill Roon every time it detects the slightest error? It seems apparent that before this update Roon could run fine with a certain amount of database errors. Why couldn’t the new database tool just warn the user there are errors, but allow Roon to continue to work (assuming no severe errors).? Maybe I’m not completely understanding how the new database tool works, but it seems the way it appears to be working now and apparent reputation of the database software used to be easily corrupted, it’s going to create a poor user experience for many going forward, ie - instead of just issuing a notice about errors and letting the user continue to play music, Roon will just shut down and force the user to recover a good backup or start all over again.
I retired 18 years ago so I could quit thinking about that stuff. A bad day of retirement is better than a good day at work, especially with Roon, Tidal, and Qobuz. There are no bad days. Some are just better than others.
(now wearing snowshoes on my asymmetrical isolation feet)
That’s right. But one thing we can assume is that all those who run into a catastrophic event will seek support. Maybe not all on the forum. Then of course it is right that not all will have intended to update.
But the interesting bit is the numerator of supposedly 400 reported cases. This is unknown but I believe stems from another thread where somebody made a projection parting from 1 affected user out of supposedly 500 beta users, which seems to be correct. If this were really so and you could project from this data point, it would at the end be rather 500 affected users.
It’s up to Roon to deal with the affected users and offer them adequate support. But personally I still think that ‘latent database corruption’ is luckily rather rare, and if it were me I would update now or soon, maybe waiting for some official Roon statement about the possibility that those affected can in the near future count on any tool to try and repair their databases. But as Roon has already and repeatedly stated that this has happened before with other major releases, and there never has been any possibility before to repair the damaged databases, I wouldn’t hold my breath.
The day all those reports surfaced on the forum I felt really bad for those affected. That shouldn’t happen, and I think it has been past negligence to not have provided much earlier better consistency checks when backing up the Roon database.
You are supposed to regularly backup your database, and Roon must make sure that your backups are fine and not ‘latently corrupted’. For this to work, the system must halt as soon as it detects a new corruption to give you the opportunity to restore a recent and consistent backup and to check on your hardware to make sure it is working fine.
If corruption occurs it is likely that there are hardware problems, even if this is not the only possible cause. So, keeping running and ignoring this would not be ok.
However, at least at this point, I’m not confident the new database tool is correctly identifying corruptions as evidenced by some users reporting 880 reporting their database is corrupted but 882 reporting the same database is fine. This makes me think the new database tool may have a bug(s) that has yet to be identified by Roon.