critical defect found in production in testing code example
Example 1: what to do when find defect
In my company we use Jira XRay for as bug tracking tool.
So when I found a bug it means somehow the
expected result is not matching with actual result.
Then I make sure that it’s a real defect.
First I double check my test scripts whether they are ok.
Then I check my enviroment whether everything
up and running. Then also I check if it is
known or duplicate issue.
Once I am sure that it is defect
then I take screenshots and
contact with the responsible developer and
tell him/her about the test results.
If he fixes immediately that is ok.
If he consent that is a bug then
I create a new ticket on Jira and I add all
screenshots and to help him to understand how
bug looks like. I link the bug with related test
and I assign to developer and he/she starts
fixing the bug. After fixing he deploys the code
and I re-test it again.
If it is passed this time it means
we are good to go, close the defect life cycle.
But if it is still failed then I reopen the ticket.
After everything is ok then I am closing the cycle.
Example 2: bug caught in production why it wasn't catch in test environment
Answer:
Initially we need to check if the issue reported in
Prod environment is a valid defect. If so:
1. The test environment is different from Production environment.
Exp: Access & Permissions in Prod environment,
Memory limitations, ELB timeout, Single & Multi-tenant
configs etc is different.
2. Tester doesn't test that scenario or he may test
it in a different way.
3. There might be an issue regarding Timezone server time.
4. One common reason for this behaviour is
production data schema differs when we add new features
and if not updated or correctly replicated on all servers
it might create production issues.
5. Exhaustive testing is also not possible. ==>
We cannot test everything.
6. Code is not migrated to Production environment properly,
there might be some issues in migrating the code for that
particular functionality where the issues caught.
7. There might be requirement gap.
8. Verify if Prod and QA had similar internet connectivity
setup (like VPN, LAN, Wifi)