MySQL Spam about prism-bukkit HOT 7 CLOSED

Right now it changes the database to use BIGINT (long). I could revert that part because the database uses unsigned int values and this problem came from java failing to read signed ints.

Alternatively you could set meta.schema_ver to 6 before launching prism, and then it wouldn't attempt to modify the database.

At any rate, glad to help out. I wasn't expecting anyone to take a look at this fork so early, so there are likely some bugs in it. I'll try to squash them in a timely manner.

from prism-bukkit.

I don't think I touched anything that would have caused this, were you having this issue with an unmodified prism?

from prism-bukkit.

That is not tested, don't trust it yet :P

from prism-bukkit.

I was using GPrism build since a year from now on. I tried yours no luck and also the /pr i was broke aswell so I've reverted to GPrism

from prism-bukkit.

You had so many entries that the ids didn't fit into a java int, which would break most of the plugin.

I've changed my fork to use long/bigint for ids, which should fix your problem but will also modify the database in a way that might not be backwards compatible. After that patch, I had no trouble using the inspect wand, lookups, and rollbacks even with large ids.

If you want to stick with GPrism, I'll close this ticket because it should be fixed and I have no reasonable way to test a database as large as yours. If you want to try this fork again, I'll wait for confirmation that it's fixed.

from prism-bukkit.

Well, in the midnight I tried fixing the console errors and the issue with the column '1' which is the ID, is a INT and I could add new lines in PMA, which Prism cannot... Even looking up it couldn't but I could view in PMA. Even in prism_data_extra I was able to add a newline without issues. I've did a dirty fix by following stacktraces on console and it seems to get to work again (logs everything at least I hope and lookup works) I'll be sure trying this build later.

Just to note, since it wont be backwards with GPrism, does your fix (hope not) converts the entire database INT ids to LONG ids? If so it would take hours of downtime.. For now, Close the ticket. I'll let you know when I'm using your fork! Thanks for your time and great help

from prism-bukkit.

Okay^^ Thanks so much again!

from prism-bukkit.