Jump to: navigation, search

Difference between revisions of "BugFilingRecommendations"

(Created page with "= Bug Filing Recommendations = Writing good bug reports is hard; useful reports are even harder. We should try our best to be thorough, so there's consistency and a fine exper...")
 
(Bug Filing Recommendations)
Line 1: Line 1:
 
= Bug Filing Recommendations =
 
= Bug Filing Recommendations =
Writing good bug reports is hard; useful reports are even harder. We should try our best to be thorough, so there's consistency and a fine experience reading bugs. Some obvious suggestions:
+
Writing good bug reports is hard; useful reports are even harder. We should try our best to be thorough, so there's consistency and a fine experience reading bugs. This document concerns itself with outlining some obvious suggestions while reporting/verifying bugs:
 +
* Ensure you can reproduce the bug
 +
* '''Clear''' instructions to reproduce the bug. Bonus points for a reproducer script.
 +
* Version details (e.g. Havana ? Grizzly)
 +
* Test environment details
 +
** e.g. some special hardware -- an  exotic NAS, etc.)
 +
* Verification procedure
 +
** Include test setup details, configuration details, other relevant context.
 +
** Most of "cloud" testing is dependent on test environment, clearer the details, lesser the round-trips between Development and QE
 +
* If there's a fix available, and someone is verifying it, adding some verification evidence would be useful (instead of just posting a comment saying - "verified". Of course, this can be debated based on the complexity of bugs).
 +
** Relevant log fragment, stdout of a script, or a command being executed.
 +
* Additional info (where appropriate):
 +
** If you've done a lot of investigation into the issue, adding a trace of that would be useful for later archival purposes. Configuration settings, caveats, reproducer scripts, etc.
 +
 
 +
You get the drift !
 +
 
 +
== Why?  ==
 +
* Useful for new test engineers who does not have all the context.
 +
* Useful for docs folks to help them write correct errata text/release notes.
 +
* Useful for non-technical folks reading the bugs/RFEs. Clear information saves a of a lot of time.
 +
* Useful for downstream support organizations.
 +
* If there's a regression years later, having all the info to test/reproduce in the bug, right there makes your day.
 +
* Reduces round-trip of NEEDINFO between Development and QE.
 +
*Useful for new users referring to these.
 +
* Overall, a very ''fine'' bug reading experience.
 +
 
 +
== Resources from other communities ==
 +
* Bug writing guidelines -- https://landfill.bugzilla.org/bugzilla-4.2-branch/page.cgi?id=bug-writing.html

Revision as of 06:10, 8 May 2013

Bug Filing Recommendations

Writing good bug reports is hard; useful reports are even harder. We should try our best to be thorough, so there's consistency and a fine experience reading bugs. This document concerns itself with outlining some obvious suggestions while reporting/verifying bugs:

  • Ensure you can reproduce the bug
  • Clear instructions to reproduce the bug. Bonus points for a reproducer script.
  • Version details (e.g. Havana ? Grizzly)
  • Test environment details
    • e.g. some special hardware -- an exotic NAS, etc.)
  • Verification procedure
    • Include test setup details, configuration details, other relevant context.
    • Most of "cloud" testing is dependent on test environment, clearer the details, lesser the round-trips between Development and QE
  • If there's a fix available, and someone is verifying it, adding some verification evidence would be useful (instead of just posting a comment saying - "verified". Of course, this can be debated based on the complexity of bugs).
    • Relevant log fragment, stdout of a script, or a command being executed.
  • Additional info (where appropriate):
    • If you've done a lot of investigation into the issue, adding a trace of that would be useful for later archival purposes. Configuration settings, caveats, reproducer scripts, etc.

You get the drift !

Why?

  • Useful for new test engineers who does not have all the context.
  • Useful for docs folks to help them write correct errata text/release notes.
  • Useful for non-technical folks reading the bugs/RFEs. Clear information saves a of a lot of time.
  • Useful for downstream support organizations.
  • If there's a regression years later, having all the info to test/reproduce in the bug, right there makes your day.
  • Reduces round-trip of NEEDINFO between Development and QE.
  • Useful for new users referring to these.
  • Overall, a very fine bug reading experience.

Resources from other communities