Hi!
There are two tests to the same IIS site
test1 - test response time
test2 is testing the response code
Required if either test1 or test2 is not answered 30 times in a row to restart the IIS website. (I have the script for this)
It is clear how to do for one test, but how to take into account two tests?
For example, 30 times in a row is not available "test1" - I restart IIS, then again I need 30 times in a row to wait when it will be available or test1 or test2. But it is possible that 30 times in a row for test2 work and thus twice will restart the website.
Please tell me how to consider?
Best Regards, Mike.
Complex from two tests
-
KS-Soft Europe
- Posts: 2832
- Joined: Tue May 16, 2006 4:41 am
- Contact:
Not sure I undestand what exactly you need.
However, I'll try to explain how you can start action, based on status, recurrences of two test items.
First option - You may use Advanced Mode action.
Expression for Advanced mode should contains variables that represent Status or recurrences of two test items.
Several examples:
Start action when only both test items have BAD/WARNING status (once):
('%::test1::SimpleStatus%'=='DOWN') AND ('%::test2::SimpleStatus%'=='DOWN') AND ('%FailureIteration%' == 1)
Start actionwhen one of 2 test items have BAD/WARNING status 10 or more recurrences (each test recurrence):
('%::test1::FailureIteration%'>10) OR ('%::test2::FailureIteration%'>10)
Start action when 2 test items have BAD/WARNING status more than 10 recurrences in a row and start action each 10 securrences:
('%::test1::FailureIteration%'>10) AND ('%::test2::FailureIteration%'>10) AND ('%FailureIteration%' mod 10 == 1)
Also you may change status of test1, depending on test2 status.
E.g. using "Use Normal Status if" option with expression like the following:
('%SuggestedSimpleStatus%'=='DOWN') AND ('%::test2::FailureIteration%'<5)
And assign Action only for test1.
However, I'll try to explain how you can start action, based on status, recurrences of two test items.
First option - You may use Advanced Mode action.
Expression for Advanced mode should contains variables that represent Status or recurrences of two test items.
Several examples:
Start action when only both test items have BAD/WARNING status (once):
('%::test1::SimpleStatus%'=='DOWN') AND ('%::test2::SimpleStatus%'=='DOWN') AND ('%FailureIteration%' == 1)
Start actionwhen one of 2 test items have BAD/WARNING status 10 or more recurrences (each test recurrence):
('%::test1::FailureIteration%'>10) OR ('%::test2::FailureIteration%'>10)
Start action when 2 test items have BAD/WARNING status more than 10 recurrences in a row and start action each 10 securrences:
('%::test1::FailureIteration%'>10) AND ('%::test2::FailureIteration%'>10) AND ('%FailureIteration%' mod 10 == 1)
Also you may change status of test1, depending on test2 status.
E.g. using "Use Normal Status if" option with expression like the following:
('%SuggestedSimpleStatus%'=='DOWN') AND ('%::test2::FailureIteration%'<5)
And assign Action only for test1.
Thank's.
I think it will be enough to configure this information.
statement of the problem - or 30 times the inaccessibility of time, or 30 times of unavailability codes. Then reset both counters and over again waiting for that to happen or 30 times of unavailability time, or 30 times of unavailability codes.
In time the site may be available, but the code may be different.
I think it will be enough to configure this information.
statement of the problem - or 30 times the inaccessibility of time, or 30 times of unavailability codes. Then reset both counters and over again waiting for that to happen or 30 times of unavailability time, or 30 times of unavailability codes.
In time the site may be available, but the code may be different.
In such case your server may be unavailable for a months and it will not be restarted.
Probably its better to use single URL or HTTP test that checks availability (it makes sure server responds and it responds with "good" HTTP code) and use standard mode action to restart server aften N consecutive "bad" results?
Regards
Alex
Probably its better to use single URL or HTTP test that checks availability (it makes sure server responds and it responds with "good" HTTP code) and use standard mode action to restart server aften N consecutive "bad" results?
Regards
Alex