SPONSORS:






User Tag List

Thanks Thanks:  0
Likes Likes:  0
Dislikes Dislikes:  0
Results 1 to 9 of 9
  1. #1
    Junior Member
    Join Date
    May 2003
    Posts
    4
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    3 different OS means 3 times testing?

    Hello !

    The backend of our application runs on Solaris, Linux (Redhat) and Windows. How can I efficiently test them without having to run tests 3 different times, one for each platform. We have a JAVA API that can interact with the server - could this be used effectively.

    We are open to any tool or methodology right now, since we are at the planning stages.

    Thanks for any input....
    Cheers !

    ------------------

  2. #2
    Moderator Joe Strazzere's Avatar
    Join Date
    May 2000
    Location
    USA
    Posts
    13,170
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    1 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by bhanubala:
    The backend of our application runs on Solaris, Linux (Redhat) and Windows. How can I efficiently test them without having to run tests 3 different times, one for each platform. We have a JAVA API that can interact with the server - could this be used effectively.<HR></BLOCKQUOTE>

    Certainly the back end of your application must be tested many times (probably more than 3 - are you only supporting one version each of Solaris, Linux and Windows?)

    But, perhaps the front end won't need as many combinations?

    Think about:
    - one person/team tests the back end and possibly the API
    - a different person/team tests the front end

    Also:
    - If the API abstracts the backend, then you could efficiently drive the API with some automation. Even though you'll need to test many times, at least each test can be efficient.

    ------------------
    - Joe (strazzerjATaolDOTcom)
    Joe Strazzere
    Visit my website: AllThingsQuality.com to learn more about quality, testing, and QA!

  3. #3
    Senior Member
    Join Date
    Feb 2000
    Posts
    1,497
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by bhanubala:

    The backend of our application runs on Solaris, Linux (Redhat) and Windows.
    <HR></BLOCKQUOTE>

    Is the backend also written in Java?

    If so, and they're all running that same code, then it would be redundant to test 100% of the functionality on each of the three platforms - you'd only be checking the accuracy of the JRE.

    Test them all once to verify consistency then drop the testing to random selections of 1/3 of the tests to each platform for subsequent releases.

    John


    ------------------

  4. #4
    Moderator Joe Strazzere's Avatar
    Join Date
    May 2000
    Location
    USA
    Posts
    13,170
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    1 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by John J. Miller:
    Is the backend also written in Java?

    If so, and they're all running that same code, then it would be redundant to test 100% of the functionality on each of the three platforms - you'd only be checking the accuracy of the JRE.
    <HR></BLOCKQUOTE>

    Good point,
    If the backend is written in Java, and if you are convinced that the JRE's are completely consistent across your three platforms you could avoid some of the testing.

    I leave it to others more experienced with all of these platforms than I to comment on the consistency of JREs.

    ------------------
    - Joe (strazzerjATaolDOTcom)
    Joe Strazzere
    Visit my website: AllThingsQuality.com to learn more about quality, testing, and QA!

  5. #5
    Junior Member
    Join Date
    May 2003
    Posts
    4
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    jstrazzere and John,

    Thanks for your posts. The backend is in Java. So the question I guess, would be the consistency of JRE, but also, what subset of tests do we pick to rerun on the second & third platform - should we base that on user actions most performed (via the GUI) ?

    As jstrazzere said, maybe we could use the GUI as an abstraction of a lot of the functionality ? And then use scripts to test the backend code -again we write scripts for 3 platforms ?

    Thanks !


    ------------------

  6. #6
    Moderator JakeBrake's Avatar
    Join Date
    Dec 2000
    Location
    St. Louis - Year 2025
    Posts
    15,609
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    A possible approach - if you are pressed for time:

    Perhaps a CM person or someone in that role should share in that burden and "test" the correctness of the various libraries and JRE versions on the backend - via version control and accounting analyses. That may help to reduce different OSs as a risk area and place it on the lower priority part of the functional testing spectrum. In other words, try to assess and communicate the risks such that if you end up being limited to testing one or two flavors of the three - there is at least some quality check in place with version control and CM.

    For performance testing, IMHO - it would be good to include all flavors in a test, since they will perform differently and will probably require tweaking of system and app configurations.

    ------------------
    ...
    Airborne requirements seldom land and are exempt from gravity.

    JP

    [This message has been edited by jpensyl (edited 05-09-2003).]

  7. #7
    Moderator Joe Strazzere's Avatar
    Join Date
    May 2000
    Location
    USA
    Posts
    13,170
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    1 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by bhanubala:
    As jstrazzere said, maybe we could use the GUI as an abstraction of a lot of the functionality ? And then use scripts to test the backend code -again we write scripts for 3 platforms ? <HR></BLOCKQUOTE>

    I don't want to confuse you.
    I wasn't suggesting just testing through the GUI.

    I thought you were contemplating driving the API directly, rather than through the GUI.

    ------------------
    - Joe (strazzerjATaolDOTcom)
    Joe Strazzere
    Visit my website: AllThingsQuality.com to learn more about quality, testing, and QA!

  8. #8
    Junior Member
    Join Date
    Mar 2003
    Location
    B\'lore, Karnakata, India
    Posts
    24
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    Hi,

    We had our servers running on 4 OS platforms (HP-UX, AIX, Solaris and linux) and all were of different versions of OS.

    Our strategy for testing was to automate the basic scenarios (like installation, starting servers, engines, testing the sample utilites provided with the server..etc).

    For this we had the test suite mounted on a common location and wrote shell scripts which triggers the testing of the server.

    - Sam

    ------------------

  9. #9
    Junior Member
    Join Date
    Feb 2000
    Location
    Easley, SC, USA
    Posts
    27
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: 3 different OS means 3 times testing?

    Our configuration is Java front-end, C backend. We fell compelled to thoroughly test our front-ends against each WinOS we support (currently NT, 2000, XP) We have unfortunately seen JRE differences between WinOS's. Not many, but serious enough to be very concerned about. One example, copying forty rows from Excel to the clipboard and then foregrounding the Java app would cause the Java app to grab all available memory. This was relatively invisible under 2000 and XP because the memory was quickly released. The memory was never released under NT and the system thrashed to the point of becoming non-responsive. Our developers suspect that the different WinOS's have different garbage collectors, but that's just a guess.

    The point of this is that if you do feel that more front-end testing is required, this gives you the chance to balance your front-ends and back-ends. Of course, back-end testing that can be easily scripted through your API is where you'll really get bang for your buck.

    The WinOS experience described above provokes another question. We do not have much of a feel for how to intelligently select a sub-set of tests if we are trying to verify JRE behavior on a specific WinOS. Does anyone have any pointers for this task?

    ------------------

 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Search Engine Optimisation provided by DragonByte SEO v2.0.36 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 10.00%
vBulletin Optimisation provided by vB Optimise v2.6.4 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging v3.2.8 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
vBNominate (Lite) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Feedback Buttons provided by Advanced Post Thanks / Like (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Username Changing provided by Username Change (Free) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
BetaSoft Inc.
Digital Point modules: Sphinx-based search
All times are GMT -8. The time now is 12:51 AM.

Copyright BetaSoft Inc.