1. Advertising
    y u no do it?

    Advertising (learn more)

    Advertise virtually anything here, with CPM banner ads, CPM email ads and CPC contextual links. You can target relevant areas of the site and show ads based on geographical location of the user if you wish.

    Starts at just $1 per CPM or $0.10 per CPC.

302 Hijacks returning?

Discussion in 'Google' started by NetMidWest, Jan 11, 2007.

  1. #1
    I am seeing some disturbing serps, it seems that Google is following 302 redirects and indexing the resulting content as belonging to the domain the 302 redirect status code is generated from, which is a problem that has lead to 302 hijack problems in the past.

    Let me be clear: I have not found a full hijack at this time.

    What I am seeing is this:
    http://209.85.165.104/search?q=site:imdb.com/r/&hl=en&lr=&rls=GGLD,GGLD:2005-15,GGLD:en
    which is changing in number of results on various d/c's as I post here... this one shows 19,500.

    http://209.85.165.104/search?q=amazon.com+rachael+ray
    brings up an imdb.com 302 redirect url (check the status codes), and the cache is this:
    http://209.85.165.104/search?q=cach...azon.com+rachael+ray&hl=en&gl=us&ct=clnk&cd=4

    And it has PR:
    http://livepr.raketforskning.com/?u...01036a02057370000001037a0403786f6070000001047

    The actual page on amazon.com seems to be a search result, and not something that amazon.com is linking and promoting, so there is no real page to be hijacked...

    From the pr update thread, there have been a number of posts about amazon.com and other biggies going to PR 0 - and I think this is the cause. Watching amazon.com's pr this morning, I could definitely see that certain d/c's were assigning PR to redirects and taking it from somewhere - there was some correlation at times to the d/c's showing PR 0 for amazon.com:
    http://livepr.raketforskning.com/?u=http://www.amazon.com/

    I think that if you check, you may find that some of the other sites that are getting PR 0 or watching backlinks go to zip currently in this update may have a similar redirect problem...

    Which leads me to believe that there may be some sites fully hijacked out there... :(
     
    NetMidWest, Jan 11, 2007 IP
    debunked likes this.
  2. MrMOJO

    MrMOJO Well-Known Member Affiliate Manager

    Messages:
    817
    Likes Received:
    32
    Best Answers:
    0
    Trophy Points:
    100
    #2
    God, I hope not.
     
    MrMOJO, Jan 11, 2007 IP
  3. NetMidWest

    NetMidWest Peon

    Messages:
    1,677
    Likes Received:
    151
    Best Answers:
    0
    Trophy Points:
    0
    #3
    http://www.google.com/search?q=link:http://www.bluehost.com/
    brings up:
    http//www.webhostdir.com/banners/banman.asp?ZoneID=55&Task=Click&Mode=HTML&SiteID=1&PageID=1297
    at #4. (replace the colon)
    Running it through a status code / server header checker produces a 302 status code.
    On the bluehost site, it 301's, which is a good thing.

    But you really have to wonder what is going on when you see a 302 redirect being counted as a backlink in Google.

    I am not worried about getting hijacked, I know how to avoid it, and have tried to get others to follow the same advice.

    But when you think about what happens to a site when it gets hijacked, you realize the links from it are no longer any good.

    Think about the ripple effect of such a problem, how that can effect serps down the line. :(

    Has anybody that is linking relatively had their backlinks reduced to 0, PR reduced to 0 in this last update? I'd sure like to check it out.
     
    NetMidWest, Jan 17, 2007 IP
  4. hhheng

    hhheng Banned

    Messages:
    2,633
    Likes Received:
    37
    Best Answers:
    0
    Trophy Points:
    0
    #4
    I don't think this is a hijack. Maybe that's the problem of google. Many people reported problem of their website during this update. Let's wait until next update and see.
     
    hhheng, Jan 17, 2007 IP
    NetMidWest likes this.
  5. NetMidWest

    NetMidWest Peon

    Messages:
    1,677
    Likes Received:
    151
    Best Answers:
    0
    Trophy Points:
    0
    #5
    No, it's not - and I have yet to find one, though I have several likely situations to watch.

    I don't see any reason it could not happen as things stand, unless there have been internal changes at Google to keep these links from causing harm. The next update, if things remain as they are, will likely show a few.

    Before this last update, there have been infrastructure changes, supplemental pages are processed differently, there were problems with some dropping of index pages of .com domains hosted outside the U.S.
    Now there are crazy PR reports - going PR0 on the homepage or PR0 on interior pages, and strange PR distribution within sites. These really make me wonder if there isn't one hiding out there, unfound at present.

    I am looking for an example of a site going PR0 throughout and losing all backlinks in this update as that would possibly indicate a full hijack. Finding a dynamic url that redirects to your site and is listed in the serps at a position held long-term for a specific keyword or phrase would be another situation I would like to hear about. Quite often, the same keywords you use in the title of the homepage would be the phrase that would show. (A full hijack will show the page the dynamic url is published on.) Sometimes that phrase is a company name or catch phrase that is not generic.
     
    NetMidWest, Jan 17, 2007 IP
  6. Paz

    Paz Well-Known Member

    Messages:
    587
    Likes Received:
    36
    Best Answers:
    0
    Trophy Points:
    170
    #6
    I know this is quite an old thread but I have an interesting site Capita L&D. It lost all it's rankings, PR and later backlinks during January. Google completely ignored the homepage for another 3 months and searches for the domain name were pointing to base href tags, the 404 Not Found page... anywhere but the real homepage, which was last crawled 7th Jan according to Google tools.

    The Capita group have dozens of domains on the same IP, interlinked with plain links, 302 redirects and js redirects within frames. Also they didn't have a canonical, non www redirect until after the missing homepage problem showed up. Capita L&D's PR and backlinks came back in mid March, but it was (and still is) the PR and backlinks of another Capita domain.

    If you look at the cache of capita-ld.co.uk you'll see capita.co.uk content;
    http://209.85.135.104/search?q=cache:http://www.capita-ld.co.uk/

    and there's a similar story with the links. Funnily enough if you look at the cached Capita L&D version via the site operator, you'll see the bona fide content, you only see the two domains "crossed" when you use the cache:example.com rather than, say cache:F-6lUpHxv3YJ:example.com



    Cheers,
    Paz.
     
    Paz, Apr 8, 2007 IP
    NetMidWest likes this.
  7. NetMidWest

    NetMidWest Peon

    Messages:
    1,677
    Likes Received:
    151
    Best Answers:
    0
    Trophy Points:
    0
    #7
    I think that is the point of the extra junk in those queries, to help sort them out, or make it harder to find the problems.

    Sure enough, the links are the same:
    http://www.google.com/search?q=link:http://www.capita-ld.co.uk/
    http://www.google.com/search?q=link:http://www.capita.co.uk/

    Another query that bring up the problem:
    http://www.google.com/search?q=info:http://www.capita-ld.co.uk/
    (shows capita.co.uk info, note where the cache, similar, and link point!)

    If you can find the anchor text for the link that produces a 302 status code, chances are a Google query will show the wrong site as well - which if it is a major keyword phrase, will kill the ranking of the phrase for that site. It may be easier to spot by searching in quotes.

    I did find a few hijacked pages on other sites, but they were pretty new and spammy, one had a high percentage of duplicate content. I have not gone back to check them and see if they turned into domain killer 302 hijacks. Sometimes new pages and sites will be temporarily hijacked, and straighten out on their own.

    But this is a pretty good example - if nothing has changed, PR for capita-ld.co.uk will probably go to zero in the next update.

    These sites did this to themselves, but it does show that it can still happen, and I see no reason that a site owned by someone else could do the same...
     
    NetMidWest, Apr 8, 2007 IP
    Paz likes this.
  8. Paz

    Paz Well-Known Member

    Messages:
    587
    Likes Received:
    36
    Best Answers:
    0
    Trophy Points:
    170
    #8
    Actually I'm quite heavily involved with this site and was looking for background info, which is how I stumbled across this thread.

    They had a lot of redirects but I think the strongest and most frequently occurring anchor text in their 302's was "Capıta". The way the sites were interlinked meant that they had to have a common 404 Not Found page, which until February was actually a 302 redirect from the capita-ld.co.uk to the capita.co.uk domain. The C' L&D site had an upgrade in December and there were lots of pages with strong, natural backlinks that went AWOL and pointed to the 404 NF page.

    Even now, if you do a search for "Capita" you'll see the capita-ld.co.uk domain in "n" and "n+1"th place and the "n+1" result points to various "We're sorry, the page you requested cannot be found" urls that change every 72 hours.

    But there were lots of other problems with these sites, I sent them a 13 page Word document about their duplicate content problems in February (interesting that you found it a factor in the sites you looked at) but the point is that these sites had been like this for years and that something changed in mid January.

    Anyway, thanks for listening... LOL

    Cheers,
    Paz.
     
    Paz, Apr 8, 2007 IP
  9. NetMidWest

    NetMidWest Peon

    Messages:
    1,677
    Likes Received:
    151
    Best Answers:
    0
    Trophy Points:
    0
    #9
    Googleguy (Matt Cutts) indicated some time back that there was a trigger to 302 hijacks, usually a spam penalty (such as dupe content).

    But I have also seen them occur due to downtime...

    http://clsc.net/research/google-302-page-hijack.htm has good advice, but I do not see that a dedicated IP is necessary. Linking absolutely is probably the best advice for avoiding and fixing the problem...

    Something else to look at is this:
    http://forums.digitalpoint.com/showthread.php?p=982752#post982752
    (You may wish to read a bit further back, we got off topic there...)
    Several posts have to do with 302's and error pages, and how they can affect a site.

    Fact is, a good header checker is a webmaster's best friend... you cannot build a skyscraper on shifting sand.
     
    NetMidWest, Apr 8, 2007 IP
  10. jason@strangelogic.com

    jason@strangelogic.com Peon

    Messages:
    36
    Likes Received:
    1
    Best Answers:
    0
    Trophy Points:
    0
    #10
    it's been about for a while. I mentioned it in some detail a few months back.

    http://widgetlogic.com/2007_02_14/has-the-302-hijack-returned/

    Is it pretty? Nope
    Is it happening? Well I haven't seen *many* implementations of true hijacking but seen lots and lots of potential hijacks.
    What can you do about it? Not a lot, other than be prepared to act if you are the victim of a true hijack rather than a Google algo problem.

    Possible answers are to hijack your own page with one of obscene pages. Your own hijack will always win in that scenario
     
    jason@strangelogic.com, Apr 12, 2007 IP
    NetMidWest likes this.
  11. NetMidWest

    NetMidWest Peon

    Messages:
    1,677
    Likes Received:
    151
    Best Answers:
    0
    Trophy Points:
    0
    #11
    Check this out, this is what looks to be a single page hijacked:
    http://www.google.com/search?q=cache:http://www.build-reciprocal-links.com/jump.asp?id=1218
    The real url's cache:
    http://www.google.com/search?q=cache:http://find-hotels-in-canada.com/niagara-falls-hotels/
    For me, it shows nothing... but there have been some bugs reported in cache queries.

    http://www.google.com/search?hl=en&...iagara+Falls+|+Niagara+Falls+-+online+booking
    Look around #10...

    The use of info: searches is extremely handy... see my post above, and check this out:
    http://www.google.com/search?hl=en&...w.build-reciprocal-links.com/jump.asp?id=1218
    and then this:
    http://www.google.com/search?q=info...nada.com/niagara-falls-hotels/&hl=en&safe=off
    Nothing for me...

    And now, for the kicker:
    http://www.google.com/search?q=site:http://find-hotels-in-canada.com&hl=en&safe=off

    Sure looks like a domain killer...

    I've been watching this one, hinted at it in one of my January 17th posts.

    What is really scary... the hijacked site links absolutely. :eek:
     
    NetMidWest, Apr 12, 2007 IP