Unanswered: log filling up to hundreds of megs for tiny app!!!
I don't understand this at all!! I cannot add anything to the database as it says:
Can't allocate space for object 'syslogs' in database <my DB> because 'logsegment' segment is full/has no free extents.
I have used ALTER DATABASE to add a new device for the logsegment, 5 times!! @ 20Meg a pop. What is going on?? It seems that when I add the new space, the logsegment gets filled as soon as I insert new data into the DB.....No matter HOW much I extend the segment. This is a tiny app. I just don't understand enough about Databases - can someone please tell me what I have to do?
I can't dump transaction either, because it says I must dump database first. I don't get it. All I want is for my app to function, without filling up the DB.
For some reason - ANYTIME I try and access the inserted table in my Insert trigger, the transaction log just fills up until it runs out of space. I have performed simple queries in my trigger (even something as simple as select * from inserted) just to see why it locks up on me. It doesn't lock up and fill the tran log when I do a query that DOESN'T access the inserted (or deleted) table. This is very strange. Can someone please help me? I have no other triggers working...
You need to carry out a full database dump before you can do a txn log dump. If you don't need to backup this database (?) then just set the db_option "trunc log on chkpt" - thisd will clear down the txn log every time a checkpoint occurs i.e. fairly regulalry
Sorry, i'm not enough clear :
i have THIS problem : i run an insert (only one record) on a table with a "minimal" trigger.
On "SELECT @ins_row = COUNT(*) FROM inserted", ASE fills up the log segment (who were empty before !)
I've find some post on the web like this, but no answer.
This trigger has no goal, first it contains a lot of code, but only this code is enough to cause the problem.
I'd find the solution :
there is in this version a bug (216805). When you alter the table, and modify the trigger, it can append this problem.
Recreate the table and load datas solve the problem.
This bug were fixed in 12.0.1.