January 10, 2007 at 12:42 pm
In our QA enviornment, we have a distributor that I am seeing rapid transaction log file growth on. I can't figure out why the log is growing so quickly, as the recovery model for the database is set to simple and there are no transactions waiting to be committed. Here is an example of the type of growth i am seeing:
| DATE | SIZE(mb) |
| 1/4/2007 15:00 | 957.88 |
| 1/4/2007 22:15 | 2794.08 |
| 1/5/2007 11:05 | 6379.88 |
If anyone can send me some tips on things to look for, that would be great. My mentor DBA thinks it's not issuing checkpoints to the DB at all, but he hasn't been able to figure anything out outside of that.
Thanks!
January 11, 2007 at 8:40 am
start a trace to see where the data is coming from
January 12, 2007 at 8:34 am
With the help of a much smarter DBA, this has been solved. At some point, someone moved a system database on the system. Apparently, after you do this, you have to set the following startup parameter:
This was never removed, so when I recreated replication on the distribution box, the new database was being told that it should never recover.
Here's a link to the article if anyone ever has a problem similar to this:
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply