MGOwen's questions

It's a database server running SQL Server 2012 on Windows Server 2012 R2.

C: is the boot/windows drive, but does NOT contain the data (mdf), log (ldf), or temp files (tempdb.mdf) for the database. So it's not any of those growing and shrinking automatically.

It last happened 19:30 last night, but has happened before. There's 15GB free (25%) of 60GB and has been since Friday, nobody's been one since.

It's an ISAM (monitoring software) alert that sends out an email to me. I guess the alert could be wrong.

But is there any reason the free space on the boot/windows drive might just fall by 12GB, by itself? Say, a temp file for virtual memory or something being created? On a WS 2012 server?

I am getting this error in my server's windows event log:

An TLS 1.0 connection request was received from a remote client application, but none of the cipher suites supported by the client application are supported by the server. The SSL connection request has failed.

When I try to connect to a web service on a windows 7 box from a windows server 2003 box.

How do I add a cipher suite to one that the other supports?

(fixing clients is ideal, but failing that a server solution is fine - I have access to all boxes involved, I just want some basic encryption between them for privacy).

Along with hours of googling and reading, I have tried:

• Checked server windows event viewer (found cipher suite error)
• Added cipher suites to test1 from http://support.microsoft.com/kb/948963 (didn't help)
• Added TLS 1.0 to protocols in cipher suites in the server's windows registry (no change)
• Install IIS tools hoping that adds more protocols to Schannel (it doesn't)
• Export certificate for clients, again, but with private key included (no change)
• Check that installed cipher suites match on server and client (can't find where win2k3 lists them)
• Add TLS_RSA_WITH_AES_256_CBC_SHA (installed by above hotfix) to server's cipher suites (nope, already on there)

What's the simplest way to shrink transaction log file on a mirrored production database?

I have to, as my disk space is running out.

I will make a full database backup before I do this, so I don't need to keep anything from the transaction log (right? I have daily full database backup, probably never need point-in-time restore, though I'll keep the option open if I can - that's all the .ldf is really for, correct?).

Solved:
OK, after doing 2 backups through SSMS (not TSQL) of just the log, creating a totally new backup set, the Shrink-Files-Log dialog in SSMS finally actually worked, freeing some disk space.

Not sure why 2 backups where needed, or why TSQL didn't work, and there was no difference in the reported "space available to be reclaimed" in the shrink dialog (it was at 99% for all the shrink attempts after the first backup too, but still didn't free any space), but problem solved for now. Thanks all.

I need to replace the old database in our test environment with a fresh copy of the production database.

The test environment is actually comprised of two instances of the test database, on 2 different servers, in a mirroring configuration (it's got mirroring because production has mirroring, and the client wants the test environment to be just like production).

I thought this would be a simple matter of:

• Getting a backup (the .bak file) of the production DB
• Temporarily disabling the mirroring for the test DBs
• Overwrite (i.e.: restore) them from the (prod) backup
• Re-activate mirroring

But apparently this isn't nearly enough hoops to jump through. After wrestling with it for a day (permissions problems, cryptic errors, problems that go away if you restart SSMS (seriously), I'll spare you the full list...) I finally got the new DB restored onto both the principle and mirror instances of the test DB.

However, somewhere along the line SQL Server abandoned the mirroring configuration settings, so I right clicked on the principle, chose mirroring, and went right through the security wizard. When I clicked "Finish" at the very end of the security wizard, it did it's thing and I got the screen that reports that everything was set up successfully ("Configuring endpoints" - "Success"). Immediately after that is another window asking if I wish to start mirroring, however, clicking start on that window, I get the following SQL Server error:

An error occurred while starting mirroring.

Database "3DSS_TEST" is not configured for database mirroring. (Microsoft SQL Server, Error: 1416)

So I'm getting "Mirroring is configured successfully!" Then, as part of the same process, "mirroring can't start because you didn't configure it!" :(

Any ideas? Anyone seen this before, or know how to tease out a more useful error message or other info?

Update: Solved
As Brian says below, I'd restored the mirror database with the wrong recovery option, the correct way was:

RESTORE DATABASE [MyDb]
FROM disk = 'C:\TEmp\MyDb_LIVE_Prod_backup_2010-06-18_for_test_server.bak'
WITH
REPLACE,
NOrecovery /* This should be 'norecovery', my problem was I used 'recovery' */

Thanks!

If I recall correctly, this is fine, the different versions coexist peacefully, but I just want to double-check there's no possibility of problems (I'm installing the newer version of the .net framework on an older production server in preparation for upgrading a critical asp.net 1.1 app to 3.5 in the near future).

I have two instances of Sql Server 2005 running two copies of a database in a Principal and Mirror configuration.

I want to send an email to someone whenever there is a failover, by which I mean the Principal DB goes down (for whatever reason) and so the two DB's swap roles, the Mirror becoming the Principal (and the Principal becoming the Mirror).

I have already configured "Database Mail" and sent a test message using sp_send_dbmail successfully. But I don't know how to create a Sql Server Agent Job (or whatever will work) that runs on failover.

I imagine the way to do this would be to somehow hook into some sort of "Oh! I was the Mirror a second ago but I have just become the Principal" event and make it exec sp_send_dbmail .

Anyone have any idea how to do this? Or a better way of accomplishing this?

I am trying to copy some SQL Server Agent Jobs from one server to another (both using SQL Server 2000).

I found out how to create a script that creates them all, but these jobs seem to have dependencies, such as DTS packages, and I need to copy everything (and don't really know what I'm doing, let's be honest).

So I tried the built in Database Copy Wizard (in Enterprise Manager, remember that?). But as you can see in the screenshots linked below, it claims the destination .mdf and .ldf files already exist (they don't), or there is not enough disk space (there is).

Screenshot 1

See how next to both rows, there is a green tick in status? But then, further down, there is a red cross. The screen is confusing: is there an error or not?

If you click on Modify (sorry, silly rules about links when you have low rep):

http:// img16.imageshack.us/img16/9439/cdwerror2.png

Once again, I have all green ticks, but then those nasty red crosses down the bottom. Which is it saying? Or is it saying both?

So I continued anyway... and after it's done, the following failures are reported on screen:

CDW Logins Task Step: fails, but this is OK - it just couldn't transfer the administrator login (I think that's not a problem, becasue it's that machine's local windows admin logon).

CDW MSDB Task Step: fails, objects scripted OK, but can't be transferred, e.g.:

Transfer MSBD  Update_Daily_Person_Set (SQL SERVER AGENT JOB)

 Step Error Source: Microsoft Data Transformation Services (DTS) Package
 Step Error Description:Unspecified error

 Step Error code: 80004005
 Step Error Help File:sqldts80.hlp
 Step Error Help Context ID:1100

I'm not sure where to find sqldts80.hlp (anyone know?) and googling for the 80004005 SQL Server error reveals that it could be any one of about a hundred different things (literally) depending on the rest of the error message (none was given in this case).

CDW Databases Task Step:

The actual copying of the .mdf files failed. One DB wouldn't attempt to copy because it had active connections - I assume I can just disconnect those to solve that part.

Thanks for reading this far!

I tried allowing "Everyone" write permissions on the destination folder, no change. Any other ideas?