TechTip: Database Health Option on CA SOI Dashboard

[Brahmaiah Addepalli]

A new feature added in SOI 4.2 with latest patch - "Database Health"

 

This option shows the database health status message based on the threshold value violation. It looks at the number of rows in the following tables:

AlertHistory
Alerts

DbQualityHistory

DbRiskHistory

DbAvailHistory

You can change the threshold values at runtime in the <SOI_HOME>\SamUI\webapps\sam\thinuiconf\dbconf.properties file.
Each table has a threshold entry that you set based on the number of rows you want to be alerted on.

The configured values in the dbconf.properties are then compared with the actual table size to generate one of three messages (the below values are %):
If the threshold is between 0 and 49, the message will be "Healthy"
If the threshold is between 50 and 74, the message will be "Partially Degraded"
If the threshold is greater then 75, the message will be "Degraded"

 

 

For more details, please refer SO01669 patch readme

SO01669 - CA Service Operations Insight Monthly Update Kit - CA Technologies Documentation 

[Legacy User]

Hi Brahma

 

if you could answer some of my questions it would be appreciated thanks. 

 

1. What does a high percentage indicate for Database Health? 
2. What do the values defined in dbconf.properties mean?

 

Thanks in advance. 

[Brahmaiah Addepalli]

Hello Doctor,

 

Show DB status message on SOI Admin console to show when SOI database need to be cleaned

We have identified five major tables and assigned threshold value(number of records in the table) to each table.  Relevant tool-tip message as DB status will be shown on SOI Dashboard based on the threshold value violation.

 

Regards

Brahma

[Legacy User]

I updated the initial tech tip with some more details on how this new feature works. 

[Waseem Siddiqui]

Hi,

 

The above commands did not help in reducing the size and I was keep on getting the Database Health: Degraded message. I used the soitoolbox to find the available connectors

 

soitoolbox.exe  --getAvailableConnectors

 

and then cleaned the connectors imported data, I am not sure if there is any impact of that command but it reduced the DB size and the DB health started showing Healthy. the command used was

 

soitoolbox -x --cleanImportedData --connector= (Spectrum and APM connectors IDs provided by above command)

 

Regards,

Waseem

[Legacy User]

The instructions to this Tech Tip are "nonsensical". 

 

In the first few lines it says, "You can change the threshold values at runtime in the dbconf.properties file".

Then down a few lines it says, "If the threshold is between 0 and 49, the message will be "Healthy"" etc.

 

So, according to these instructions the threshold values I'm looking for are numbers between 0 and 100, since we are looking at percentages.

 

So, now I open the dbconf.properties file and search for threshold values....  But what I see are not percentage threshold values at all, but the actual table size given in rows.

 

PLUS there is no indication (when you hover over the Database Health text) of WHY you are getting the value that is displayed.  In my case I get 65% and the Database Health text says, "Partially Degraded".  That matches up with the Threshold Value of between 50 and 74, BUT it doesn't tell me which of the 5 tables are in trouble.

 

Can we change the TechTip so that the content matches what is really there?  It's great that new functionality is added to our products, but it has to make sense.  Customers and Consultants alike are very confused by this particular tech tip.  I had to have Dan V. explain to me what was going on and what was trying to be conveyed.  Thanks Dan!!

 

My suggestion is that we change the text to something like:

 

The threshold percentage that is displayed as you hover over the Database Health message conveys that at least one table (in the five tables being monitored) has hit a row count that is X % of the value listed in the dbconf.properties file.

 

Ex:  65% is shown as the Database Health pop-up value.  One of the tables in the dbconf.properties file is 65% of the listed row count. 

AlertHistory=1000000 (is the actual row count 650000?)

Alerts=250000 (is the actual row count 162,500?)

DbQualityHistory=1000000 (is the actual row count 650000?)

etc...

 

This way the instructions match what is in the dbconf.properties file and you know what to look for, because right now, the value given when you hover is a big question mark.

 

My 2 cents.

[Marco Russo]

I would second that. 

Also how do i find "good" values to enter in my current setup. How do i know at which level of rows in those tables i actually have an issue? 

 

Regards

Marco

[SamuelEchevarria]

Hi mgrusso,

 

You can obtain easily these values running the below SQL query:

 

select
(select count(*) from Alerts) as Alerts ,
(select count(*) from AlertHistory) as AlertsHistory,
(select count(*) from DbQualityHistory) as DbQualityHistory,
(select count(*) from DbRiskHistory) as DbRiskHistory,
(select count(*) from DbAvailHistory) as DbAvailHistory

 

The "good" values would depend on your requirements. Which data retention do you expect to have in SOI?

 

In my case, the data retention is one year, therefore I have increased the values dbconf.properties as follow:

 

AlertHistory=10000000 (one million)

Alerts=2500000 (quarter of million)

 

**If I am 50% below of the above values, the result will be green.

 

Keep in mind that the "Database status" shown in the UI console is only informational, as Brahma has explained, it will tell you in which threshold you are related to the number of rows. 

 

Regards

[JAMIE.WAGNER]

Hello,
​     are there values above which you would expect a Client to see performance issues in their SOI environment? We exceeded the "Alerts" partially degraded value in the first month we had our 4.2 environment up and running in production.

Defaults                                    Our values today
AlertHistory=1000000                 313317
Alerts=250000                            152192       60.9%
DbQualityHistory=1000000        47208
DbRiskHistory=1000000            47208
DbAvailHistory=1000000           46008

Jamie