SPONSORS:






User Tag List

Results 1 to 6 of 6
  1. #1
    Junior Member
    Join Date
    May 2000
    Location
    san francisco, ca, usa
    Posts
    2
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Page Load problems

    I am having a problem getting a particular page to load when Silk is running. I am testing htt://planner.excite.com and each time I run my appstate that signs me in as a user and then tries to load this page, the page only loads the header portions of the page and doesn't load any further.

    I have tried reloading the page and using the sleep function, but to no avail. Strangely, it does work on occasion. Can anyone give me a clue as to why this might be happening? This is happening on my Windows 98 machine using 5.0.1 and almost any Netscape 4 browser.

    [This message has been edited by zigmon23 (edited 05-02-2000).]
    Scott Brode
    Quality Assurance Engineer
    Excite@Home

  2. #2
    AJ
    AJ is offline
    Moderator AJ's Avatar
    Join Date
    Jun 1999
    Location
    San Jose, CA
    Posts
    1,691
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Re: Page Load problems

    Can you try it on IE5? I seriously doubt it's a SilkTest issue... I've done yahoo calendar before without problems... I have not tried the excite one, but I see no reason why it should fail.

    My guess is as follows: You may have your main win declaration incorrect. Can you attach the frame for the main win? or paste it in using [ code] paste your code here [/ code] <-- remove the extra space, read more about the UBB Code by clicking the UBB Code is ON link when posting.

    I'm guessing that Silk loads the page and stops it if the caption is wrong. or if it thinks it loaded a wrong page.

    ------------------
    AJ Alhait
    BetaSoft Inc.
    AJ Alhait
    BetaSoft Inc.

  3. #3
    AJ
    AJ is offline
    Moderator AJ's Avatar
    Join Date
    Jun 1999
    Location
    San Jose, CA
    Posts
    1,691
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Re: Page Load problems


    Posted by Craig Koozer:

    <BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>
    AJ et all,

    Sure, we know that all of the Browsers currently available are buggy and
    leaky creatures. This is not news. This also doesn't explain away all of
    the unexpected behavior that occurs when testing Browsers with Silktest, or
    excuse the fact that it has been this way since they started supporting
    browser testing and nothing substantial that I can see has yet been done to
    make the process more robust.

    Any reasonable amount of recording actions or window definitions from pages
    will show how unstable the agent really is. The agent frequently hangs on
    several specific calls including very important run-time method
    Browser.WaitForReady. (Segue, by the way, should be looking into
    automatically handling browser timing rather than dumping off such a
    universal issue to be dealt with separately by every individual user of the
    tool.)

    Silktest is a testing tool. As such it should assume that every call it
    attempts to make will fail until it has been shown otherwise, and handle
    stuff within the agent calls in such a way as to be more robust given a
    buggy platform.

    When developers of web sites run into browser bugginess and whatnot, they
    don't throw up their hands and say "well, the browser is just buggy, I guess
    we'll have to put a notice on our website for our users to re-start the
    browser every 10 minutes if they have problems using our service". If they
    were to do that, users simply go somewhere else. Well, as users of Segue,
    that is what we have been required to do by the current lack of stability in
    the core methodology that is used to interact with the browser.

    I can test my site manually for many hours without having to restart my
    browser or getting a browser abort. With the Silktest agent, it often
    happens immediately, or within the first half-hour of testing. This is
    alone is good evidence that the tool needs improvement.

    Other browser testing tools don't exhibit the same instabilities. This is
    also good evidence that the tool needs improvement. If you have ever taken
    a good look at their default inc files then you can see that Segue could be
    doing a better job even with the external 4-test code that we can see.

    I have had to override almost every default GUI method that I use including
    "Click" in order to get some semblance of stability and repeatability when
    testing. This is also good evidence that the tool needs improvement in the
    robustness arena. One simple example is the fact that it can't reliably
    SetText within an HtmlTextField. More than half of the time, it leaves some
    of the previous text in the field, or can't perform follow-up actions
    because the HtmlTextField now has focus, and this simply confuses the agent
    to the point that it can't figure out where it is until you manually change
    focus, or reload page or... etc.

    Rather than explaining away the shortcomings of this tool, and trading
    arcane and often misinformed work-around lore, we as users of Silktest
    should be putting very real and consistent pressure on Segue to make
    reliability of Browser testing with their Silktest product a #1 priority.
    We need to spend more time QAing our respective applications, and less in
    QAing Segue's products

    My two cents,

    Craig Koozer
    ckoozer@netledger.com
    Lead Test Automation Developer
    NetLedger.com

    <HR></BLOCKQUOTE>

    Clear points Craig. I agree on a lot of issues, Silk Agent has been unstable and will hang occasionally. My advice is to shutdown your browser after every test case (I know that is extreme) and It's a ridiculous solution. But In my career, I can tell you my love in automation is finding workarounds... I never say it's impossible, I'll just say it's hard to do and may take extra time. There's almost always a workaround
    "If it's broken, workaround it"
    I know Segue's been promising that the tool will get better soon, and I'm loosing my patience myself, but I always find a workaround that makes me even more patient

    You have to admit that Netscrape 4.x is a resource drain (not leak) IE has problems of it's own. So you really cannot expect SilkTest to be effective with them all the time

    my 1 cent



    ------------------
    AJ Alhait
    BetaSoft Inc.
    AJ Alhait
    BetaSoft Inc.

  4. #4
    Member
    Join Date
    Sep 1999
    Posts
    51
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Re: Page Load problems

    Excellent Craig! It is a very clear and straight message about Silk. I hope someone at the top from Segue looks at this message and gets a feel of its customers.

    Sam

  5. #5
    Junior Member
    Join Date
    Dec 1999
    Location
    San Mateo, CA, USA
    Posts
    17
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Re: Page Load problems

    Hi All,

    Upon re-reading my reaction/response for the newsgroup I realize that I zeroed in on only negatives, and don't want to go on record as simply slamming Silktest and Segue in general (what's that saying about speaking in anger? True. True.).

    So, in the interest of accuracy:

    Despite the concerns voiced in my prior post I still think that Silktest is the most usable tool for creating a maintainable (and therefore usable) functional test bed over the lifetime of a project available at this time. I have made extensive used of the object oriented nature of it, and have been able to use class overrides to work around most of the undesirable behavior that I spoke about in my e-mail. Of course I begrudge Segue for the amount of effort and time that I had to invest in this, and stand behind my original statements whole-heartedly. BUT, I have evaluated other testing tools such as Winrunner and Rational Test, and know that I never would have been able to accomplish with them the amount of work that I have accomplished with Silktest, despite the current frustrations inherent in using Silktest for Browser testing.

    Also, I said that nothing real has been done to correct the serious reliability issues. While I still think that statement is currently accurate, I have seen a heightened interest on Segue's part to solve my issues and to improve product usability. I am, however, reserving final judgment at this point.

    My response was really intended as an admonishment for all experienced users. We need to help each other out by diligently reporting issues to Segue as they arrive, and put pressure on Segue so that they will understand how much pressure these issues put us under, and deal with them accordingly.

    Sharing workarounds for Silktest bugs is one way of helping, and is the only right way to handle specific cases of non-standard application behavior (i.e. nothing to fix, application under test just has specialized behavior). In the long run, however, the sharing of work-arounds isn't nearly as effective or helpful to all users as reporting the information on the problems you experience with Segue, then convincing Segue to fix your problems so that nobody needs to fuss with the issues at all in the future.

    Thanks for bearing with my qualification, this should be my last novel for the month...


    >Craig Koozer
    Lead Test Automation Engineer
    NetLedger.com www.netledger.com
    ckoozer@netledger.com

    (For Betasoft post: PS: My qualification above, and my original statement before that, were not written to address this thread precisely, although it is related and pertinent.)
    Craig Koozer
    Lead Test Automation Engineer
    www.oraclesmallbusiness.com

  6. #6
    Junior Member
    Join Date
    May 2000
    Location
    san francisco, ca, usa
    Posts
    2
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Re: Page Load problems

    Thanks for the responses. AJ, you're right, it worked fine on IE 5.01. I should note that this is before a user has logged in and before the user has his/her User ID cookied.

    Let me know what I should do to workaround the problem on Netscape. I still don't understand why it works on one machine and not another.

    Here is my Main Win declaration:


    <BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by AJ:
    Can you try it on IE5? I seriously doubt it's a SilkTest issue... I've done yahoo calendar before without problems... I have not tried the excite one, but I see no reason why it should fail.

    My guess is as follows: You may have your main win declaration incorrect. Can you attach the frame for the main win? or paste it in using [ code] paste your code here [/ code] <-- remove the extra space, read more about the UBB Code by clicking the UBB Code is ON link when posting.

    I'm guessing that Silk loads the page and stops it if the caption is wrong. or if it thinks it loaded a wrong page.
    <HR></BLOCKQUOTE>

    [This message has been edited by zigmon23 (edited 05-12-2000).]

    [This message has been edited by zigmon23 (edited 05-12-2000).]
    Scott Brode
    Quality Assurance Engineer
    Excite@Home

 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

vBulletin Optimisation provided by vB Optimise v2.6.0 Beta 4 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging v3.0.9 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Questions / Answers Form provided by vBAnswers (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
vBNominatevBulletin 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 12:44 AM.

Copyright BetaSoft Inc.