This discussion is archived
1 2 Previous Next 24 Replies Latest reply: Jul 19, 2013 6:19 AM by riedelme RSS

When will it be faster

jschellSomeoneStoleMyAlias Expert
Currently Being Moderated

I believe someone or several someones claimed that the new forum software would make the site faster.

 

So I was wondering exactly when that speed up will occur? Certainly hasn't been faster in a least a week. Each page load now takes many seconds.

  • 1. Re: When will it be faster
    Paul M. Oracle ACE
    Currently Being Moderated
    when that speed up will occur?

     

     

    Actually not now... probably Oracle does not care enough of these fora, since they are free...

     

     

    And I hope it's so, because the only other alternative is... incompetence.

  • 2. Re: When will it be faster
    sb92075 Guru
    Currently Being Moderated

    why do I have to login anew so frequently?

  • 3. Re: When will it be faster
    Hemant K Chitale Oracle ACE
    Currently Being Moderated

    >why do I have to login anew so frequently?

    Somehow, it seems that that cookie isn't recognised occassionally.

     

    >when that speed up will occur?

    Opening a thread sometimes just "hangs".

     

     

    Hemant K Chitale

  • 4. Re: When will it be faster
    EdStevens Guru
    Currently Being Moderated

    HemantKChitale wrote:

     

    >why do I have to login anew so frequently?

    Somehow, it seems that that cookie isn't recognised occassionally.

     

    >when that speed up will occur?

    Opening a thread sometimes just "hangs".

     

     

    Hemant K Chitale

    I'd hope that Sonya and others are noticing the performance issues.  The first week was pretty bad, then they seemed to have taken some corrective actions and weeks 2 and 3 were pretty good. But the last couple of weeks have been pretty painful in terms of response time, time taken to respond and return a page.

     

    And something else I've notice and am not enough of a web app guy to know what's going on .. but I always thought a browser's 'back' button simply refreshed from a local cache of the last viewed page.  But here it seems that the 'back' button is sending a request back to the the forum's servers and waiting for a response.  This is with Firefox.

  • 5. Re: When will it be faster
    jgarry Guru
    Currently Being Moderated

    I would presume this is to account for state changes between the time the page was originally presented and the time it is gone back to.  I think it is a fairly fundamental problem deciding whether you want to do that, as opposed to seeing what was current then.  Not to mention, all possible caches in between the client and the eventually consistent servers.  I presume this is because java is controlling local actions, rather than default browser behavior.

     

    I had an interesting one yesterday.  On the communications screen, I clicked on a box on the left side, and it gave the "Loading..." with the sideways barber poll graphic.  I then clicked on the box above it, and it displayed a second "Loading..." and barber poll graphic.  Nothing actually loaded until I reloaded the entire page.

  • 6. Re: When will it be faster
    Dude! Guru
    Currently Being Moderated

    I'm not a html expert but caching features are controlled by the HTML request header. There have been issues in the past were certain browsers did not recognize if-modified and cache control and cached pages regardless.

     

    If you analyse the forum html response header, e.g. using View HTTP Request and Response Header and paste in a forum URL you get the following below. Note the no-cache setting, which is meant to disable page caching.

     

    GET /community/developer/english/server_%26_storage_systems/linux/generic_linux HTTP/1.1[CRLF]

    Host: forums.oracle.com[CRLF]

    Connection: close[CRLF]

    User-Agent: Web-sniffer/1.0.46 (+http://web-sniffer.net/)[CRLF]

    Accept-Encoding: gzip[CRLF]

    Accept-Charset: ISO-8859-1,UTF-8;q=0.7,*;q=0.7[CRLF]

    Cache-Control: no-cache[CRLF]

    Accept-Language: de,en;q=0.7,en-us;q=0.3[CRLF]

    Referer: http://web-sniffer.net/[CRLF]


    So I think the no caching of the browser back button is by design according to the HTTP header, which allows the following:

    Cache-Control: no-cache, max-age=0, must-revalidate, no-store

  • 7. Re: When will it be faster
    jgarry Guru
    Currently Being Moderated

    Very informative.

  • 8. Re: When will it be faster
    EdStevens Guru
    Currently Being Moderated

    Dude wrote:

     

    I'm not a html expert but caching features are controlled by the HTML request header. There have been issues in the past were certain browsers did not recognize if-modified and cache control and cached pages regardless.

     

    If you analyse the forum html response header, e.g. using View HTTP Request and Response Header and paste in a forum URL you get the following below. Note the no-cache setting, which is meant to disable page caching.

     

    GET /community/developer/english/server_%26_storage_systems/linux/generic_linux HTTP/1.1[CRLF]

    Host: forums.oracle.com[CRLF]

    Connection: close[CRLF]

    User-Agent: Web-sniffer/1.0.46 (+http://web-sniffer.net/)[CRLF]

    Accept-Encoding: gzip[CRLF]

    Accept-Charset: ISO-8859-1,UTF-8;q=0.7,*;q=0.7[CRLF]

    Cache-Control: no-cache[CRLF]

    Accept-Language: de,en;q=0.7,en-us;q=0.3[CRLF]

    Referer: http://web-sniffer.net/[CRLF]


    So I think the no caching of the browser back button is by design according to the HTTP header, which allows the following:

    Cache-Control: no-cache, max-age=0, must-revalidate, no-store

    So if, as I understand that, the forum itself is disabling caching and causing a simple 'browser back' to request a refresh from the forum servers, could that in itself be part of the performance issue ... by causing a multiplication of requests to the server ... a sort of crude DDoS attack?

  • 9. Re: When will it be faster
    BrianW-Oracle Guru Moderator
    Currently Being Moderated

    Current performance issues are raised as a top priority internally being addressed.

  • 10. Re: When will it be faster
    jgarry Guru
    Currently Being Moderated

    Perhaps more in the land of Sisyphus, he did ask Oracle how to kill his brother without consequence, ending up doing a lot of pointless interminable work.

  • 11. Re: When will it be faster
    Paul M. Oracle ACE
    Currently Being Moderated

    BrianW-OracleMod ha scritto:

     

    Current performance issues are raised as a top priority internally being addressed.

    Then I confirm what I said in my first post in this thread, because it's more than a week that performances are unacceptable.

  • 12. Re: When will it be faster
    jgarry Guru
    Currently Being Moderated

    I don't think we have enough data to adjudge incompetence.   We don't know if they are awaiting hardware to throw at the problem, performing pointless index rebuilds, guessing at code changes, performing a proper load analysis, or already have a fix and are testing it.

     

    Not sure when patience should turn to judgement.  Wouldn't want to be wussies, now, would we?

  • 13. Re: When will it be faster
    Paul M. Oracle ACE
    Currently Being Moderated
    I don't think we have enough data to adjudge incompetence.

     

    Read carefully my post : I said "probably they don't care enough", incompetence is only the alternative.

  • 14. Re: When will it be faster
    Paul M. Oracle ACE
    Currently Being Moderated

    Wouldn't want to be wussies, now, would we?

    wussies ? what's that ?

1 2 Previous Next

Legend

  • Correct Answers - 10 points
  • Helpful Answers - 5 points