First a spoiler…This is in noway related to a database or what-so-ever..But since MS thinks it looks like a database, they sell it as a database..But to further spare you guys the rant of me and SQL-Server , I’ll just give this little nugget of knowledge..
The other day the database located on a SQL-Cluster stopped, and I soon found out the dedicated disk where the log files are located of all the databases was full.
Mission: shrink/delete the log file..
The recommended way of doing things according to MS:
1) Allocate more space to the log file location
–> Right. Add more space to a physical disk in a physical server..*FAIL*
2) Backup the database
–> Always a good idea. Where it not for the fact the backup will not run, because the log file location is full. *FAIL*
3) Shrink the log file.
–> Now we are getting somewhere..Since this is MS, we have to use the GUI (although there seems to be some kind of CLI). Using this I followed the following steps:
* The database is on recovery model: full (which causes the problem in the first place!!)
* I changed this to simple (in the database properties)
* Then did a shrink (tasks-shrink- files -…)
This worked. The log files was a lot smaller then, so I changed the recovery mode back to FULL again. The users could use the database again, and the management went on with what they were doing before. And I’m searching for a way to prevent this from happening in the future.
It’s a bit concise information, but it may help you out of a tight spot if you’re an Oracle (wo)man who also needs to manage wannabe databases.