dbWarden - A Free SQL Server Monitoring Package

Question

dbWarden - A Free SQL Server Monitoring Package

Viewing 15 posts - 121 through 135 (of 186 total)

You must be logged in to reply to this topic. Login to reply

michaelrounds SSC-Addicted Points: 490 More actions · Answer 1

JonR76 (6/11/2013)
Hello All.
Just started using the dbWarden yesterday. Looks like an awesome tool. One issue that I'm having and hoping someone has seen. My Health Check report yields the error below, complaining about a linked server that I have. Has anyone seen this? I am at 10.50.1600. Thanks for any help!
Warning: Null value is eliminated by an aggregate or other SET operation. [SQLSTATE 01003] (Message 8153) The operation could not be performed because OLE DB provider "OraOLEDB.Oracle" for linked server "PERS_ORACLE" was unable to begin a distributed transaction. [SQLSTATE 42000] (Error 7391) OLE DB provider "OraOLEDB.Oracle" for linked server "PERS_ORACLE" returned message "New transaction cannot enlist in the specified transaction coordinator. ". [SQLSTATE 01000] (Error 7412). The step failed.

Interesting, nothing in the health report should access a linked server.That error usually indicates a configuration issue with DTC. If this is a test server, could you try dropping the linked server in question just to verify that you get a health report?

SQLR45 SSC Journeyman Points: 77 More actions · Answer 2

If it can help I unsintalled with the uninstall script and reinstall, same result.

It use to work for couple days then stop.

thanks,

JonR76 Old Hand Points: 303 More actions · Answer 3

Ok, in trying to drop the linked server, I got an error that it could not be dropped as it was used for replication (which it's not). So I updated the sys.servers info so that it was not flagged as a distributor. The report is now fine with that.

Now however when I run the Health Report, I get the error below. Has anyone seen this?

Executed as user: DHCRNET\sql08service. The datediff function resulted in an overflow. The number of dateparts separating two date/time instances is too large. Try to use datediff with a less precise datepart. [SQLSTATE 22003] (Error 535) The statement has been terminated. [SQLSTATE 01000] (Error 3621). The step failed.

michaelrounds SSC-Addicted Points: 490 More actions · Answer 4

I've seen this issue before with the long running queries alert and thought I had fixed it...I'll take another look at it.

You could verify that its the same section causing the issue by commenting out the Insert/select statement in the health report that Selects from QueryHistory and running the report again.

JonR76 Old Hand Points: 303 More actions · Answer 5

Yes, if I comment out everything in the "/* LongQueries */" block I get a good Health Report. I looked at what I had and it was 2.4.1. I uninstalled and put 2.5 on and everything works fine.

Sorry about that, and kudos to you for providing the package for free. I would have to go through years of paperwork and red tape, and still would not be able to purchase a product this nice.

Thanks for you time and help as well!

JonR76

michaelrounds SSC-Addicted Points: 490 More actions · Answer 6

My wife and I appreciate the comments, thank you very much. I'm glad it's all working, let me know if you come across any other issues.

michaelrounds SSC-Addicted Points: 490 More actions · Answer 7

Hey SQLR45,

Re: The blank report you are experiencing, did you modify your AlertSettings at all for the health report? If you don't mind, could you tell me what all the values are for the health report variables in that table (are they all enabled or are any turned off)?

Thanks,

Michael

SQLR45 SSC Journeyman Points: 77 More actions · Answer 8

Hello,

I think they are all default:

https://dl.dropboxusercontent.com/u/22619217/dbwarden.xlsx

SQLR45 SSC Journeyman Points: 77 More actions · Answer 9

Hello,

I also have alerts set to false but still receiving emails:

LongRunningJobsQueryValueFalse60Value is in seconds

LongRunningQueriesQueryValueFalse615Value is in seconds

LongRunningQueriesQueryValue2False1200Value is in seconds

Thanks!

michaelrounds SSC-Addicted Points: 490 More actions · Answer 10

The issue of receiving alerts when the Enabled flag is 0 in AlertSettings has been resolved in the next release. Thanks for reporting that issue.

SQLR45 - I have a test version of the health report that might help me narrow down which section is causing your report to be blank... Would you be willing to run this test proc for me? If so, give me your email address and I'll send you the script.

Thanks for your help!

SQLR45 SSC Journeyman Points: 77 More actions · Answer 11

SQLR45

SSC Journeyman

Points: 77

June 13, 2013 at 4:55 pm

#1623935

I can certainly do that I will pm you my email

Sailor SSCertifiable Points: 5246 More actions · Answer 12

On one server the daily report runs an incredibly long time. It's on vmWare SQL 2008R2.

michaelrounds SSC-Addicted Points: 490 More actions · Answer 13

When the health report runs for a long time, it's usually because it's parsing through a large error log. You can verify this by turning off (set Enabled = 0) the ShowErrorLog setting in the AlertSettings table, then run the report again.

Sailor SSCertifiable Points: 5246 More actions · Answer 14

Excellent thanks, I did a dbcc errorlog and reran it. It ran in less than two seconds.

I was on an older version that did not have the Errorlog in Alert table and just got the latest..

Sailor SSCertifiable Points: 5246 More actions · Answer 15

How about adding a job that can check disk space every hour?