Forgot your password?
typodupeerror
Chrome Cloud Bug Google The Internet Technology

Google Sync Clobbers Chrome Browsers 102

Posted by timothy
from the browser-is-the-network-is-the-computer dept.
If you use Chrome along with Google's Sync, you may have noticed something strange Monday: normally stable Chrome crashing. An article at Wired (excerpt below) explains why: "Late Monday, Google engineer Tim Steele confirmed what developers had been suspecting. The crashes were affecting Chrome users who were using another Google web service known as Sync, and that Sync and other Google services — presumably Gmail too — were clobbered Monday when Google misconfigured its load-balancing servers. ... Steele wrote in a developer discussion forum, a problem with Google's Sync servers kicked off an error on the browser, which made Chrome abruptly shut down on the desktop. 'It's due to a backend service that sync servers depend on becoming overwhelmed, and sync servers responding to that by telling all clients to throttle all data types,' Steele said. That 'throttling' messed up things in the browser, causing it to crash."
This discussion has been archived. No new comments can be posted.

Google Sync Clobbers Chrome Browsers

Comments Filter:
  • by smitty_one_each (243267) * on Tuesday December 11, 2012 @12:02PM (#42251659) Homepage Journal
    Sync was sunk
    By a bristly punk.
    Having less facial drag,
    He'd've more clearly thunk.
    Burma Shave
  • by crazyjj (2598719) * on Tuesday December 11, 2012 @12:02PM (#42251665)

    Fine as *one* backup location, fine for non-critical data and apps, fine for anything that won't be particularly missed if it goes offline for a while.

    Shit for anything important.

  • Firefox bookmarks sync is much better than Chrome bookmarks sync. Firefox stored your bookmarks locally and updated them periodically from the cloud. Chrome appears to have to download everything when I start the browser. I get a blank bookmarks bar for a few seconds when the internet is slow and I open Chrome. This is one place where Firefox got the design right and Chrome has it wrong.
    • by The MAZZTer (911996) <.megazzt. .at. .gmail.com.> on Tuesday December 11, 2012 @12:13PM (#42251799) Homepage
      What you are seeing has nothing to do with sync and would happen even if you turned sync off. Chrome is just starting up as quickly as it can and only loading bookmarks when it can get to it.
      • It has everything to do with Sync but I guess you didn't read the material. I use Chrome all the time and experienced no problems whatsoever. Why? I don't use Sync.

    • Firefox bookmarks sync is much better than Chrome bookmarks sync.

      Chrome doesn't have bookmark sync, it has a fairly deep browser state sync which happens to also include bookmarks.

      Firefox stored your bookmarks locally and updated them periodically from the cloud.

      Which, if all you want is bookmark sync, is a fairly great way to do it. That's not the focus of Chrome's browser sync, so its not surprising that Chrome's sync isn't optimized for that use case.

      • by robmv (855035)

        Firefox doesn't have only bookmark sync, it can sync settings, bookmarks, add-ons, passwords, history and tabs

        The real advantage of Firefox sync is that is encrypted on the client side, so Mozilla is unable to read your data, not the same with Chrome

        • by Anonymous Coward on Tuesday December 11, 2012 @12:34PM (#42252011)

          The real advantage of Firefox sync is that is encrypted on the client side, so Mozilla is unable to read your data, not the same with Chrome

          That's what I thought too, but apparently Chrome can do that too - it's just not on by default. Go to Settings > Advanced sync settings > Encrypt all synced data.

          • I like firefox though. They tell you you are SOL without the passkey. I have no idea how Chrome encrypts. It looks like it is linked to your google account. Google could easily be holding all the keys.
            • by CTachyon (412849)

              I like firefox though. They tell you you are SOL without the passkey. I have no idea how Chrome encrypts. It looks like it is linked to your google account. Google could easily be holding all the keys.

              Chrome uses a passphrase to encrypt sync data. By default it will use your Google account password, but you can change it to use any passphrase. If the Chrome devs are doing it right, they should be running the passphrase through PBKDF2 to derive an AES symmetric key. It's worth noting, though, that the Dashboard [google.com] for "Chrome sync" shows counts for the number of synced items of each type. Assuming they're doing the crypto correctly, I see only two ways the Dashboard could know those numbers: (a) if Chome

              • Thanks, I just found that. I read the page wrong. Great to know. Also the fact you pick the key is arguably nicer.
          • by caspy7 (117545)

            The real advantage of Firefox sync is that is encrypted on the client side, so Mozilla is unable to read your data, not the same with Chrome

            That's what I thought too, but apparently Chrome can do that too - it's just not on by default. Go to Settings > Advanced sync settings > Encrypt all synced data.

            The problem is, though, that default is the reality for the vast majority of users.

            • The reality is, that most users don't really care if someone sees their bookmarks. It's only a problem for privacy-obsessed nerds who project their mania onto other people.

              • by russotto (537200)

                The reality is, that most users don't really care if someone sees their bookmarks. It's only a problem for privacy-obsessed nerds who project their mania onto other people.

                And for anyone who has gotten burned by someone seeing the wrong bookmark/history entry. The difference with privacy-obsessed nerds is
                1) They see the problem before it actually happens to them and
                2) They have nobody to actually object to their viewing horse porn.

                • And that's cool - there's the option to turn it on for thems that wants it. Where it crosses the line is where they insist (like the GP) that having the option isn't enough, and that paranoia must be the default.

        • by kllrnohj (2626947)

          Chrome always encrypts your passwords on the client side - the other stuff isn't encrypted by default but can be encrypted as well if you choose to.

          • by robmv (855035)

            Great, I have old information of the first Chrome sync implementation. I will try it on my Chrome instances (my secondary browser)

      • by Bryansix (761547)
        Ya, I mean why would their focus be on features that users actually want?
    • by vlm (69642)

      The best part about firefox sync is it always resulted in duplicates duplicates so even if you "lost" one set of bookmarks bookmarks then you'd have the other copy copy right there. That's why I stopped using firefox sync. xmarks on FF actually worked flawlessly, but FF wanted to put them out of business by shipping something built in that claimed to do the same thing for free but actually didn't work at all. Then I switched to chrome and never looked back. GOOG can F up quite a few more times until the

    • by ArhcAngel (247594)
      Why limit yourself? I have been using X-Marks [xmarks.com] (originally Foxmarks) since 2006. Long before FireFox had sync or Google even thought of creating Chrome. It doesn't matter what system I log into or what browser I am using (well...between Chrome, FireFox, IE, and Safari) so I have a local copy on numerous machines and a copy in the cloud. You can even create bookmark profiles so those "mature" links don't show up on the work PC unless YOU want them to. If you add their premium service (US $12 annually) you can
  • Software crashes due to complex online dependency, film at 11.
  • by Baldrson (78598) * on Tuesday December 11, 2012 @12:17PM (#42251843) Homepage Journal
    It seems there are some significant problems with Chrome on Windows that go beyond Windows' standard brain damage. The problems I've run into have been numerous enough that I've had to drop Chrome on my Windows machine and go to Firefox. Everyone is familiar with the usual disk pounding that Windows considers more important than servicing user events such as mouse clicks, etc. However, in the case of Chrome it seems very much worse. Firefox -- no problem. (Yes, all the usual suspects such as extensions, plugins, malware/virus scans etc. have been dealt with.)
    • Firefox is no longer the bloated piece of crap it once was while Chrome was new. It uses the least amount of ram of any browser. It no longer requires 4 gigs of ram and a quad core like version 4. Plugins no longer break with the latest release either between versions. Chrome has gotten buggy and much slower in comparison. In 2011 Chrome would the only browser besides old IE that could run on 5 year old hardware. Now firefox runs as fast as 2.0 on these systems.

      • Firefox is no longer the bloated piece of crap it once was

        Yeah, every time a new version of Firefox comes out (what is it - every 8 hours or so?) we hear how the NEW version is all shiny and efficient, unlike those bloated OLD versions.

        The cycle's been going on for several years now. I'm sure we'll eventually be hearing how Firefox 129 isn't bloated, unlike that gosh-awful memory hog Firefox 128.

        • by Bryansix (761547)
          Memory isn't the only important factor in browser performance. Ya, its important on 32-bit machines but on newer machines what is more important is disk activity and cpu time.
          • There are many old machines out there.

            Look at the popularity of XP? Something insane like 300,000,000 still are using it today! with these machines that probably only have 512- 1 gb of ram ram usage is performance. XP swaps like a mofo too so less hard drive thrashing by less memory usage will help on such ancient platforms.

            On mine and probably your machine yeah we have 8 gigs so who the fuck cares. But on these the companies and users see no reason to upgrade and Firefox 3.6 and 4 were getting near unusabl

            • by Bryansix (761547)
              I use XP at home and I have 4GB of RAM and I keep enough tabs open all the time that I have to scroll through them.
        • Go use it rather than hold on to your beliefs? Go under help about FF and update and give it a try for 10 minutes and tell me if it is now much better?

          I submitted a story last july with Firefox 13 gets benchmarked here. You can do a search and look it up? I wouldn't touch before last summer but if you have Chrome Firefox is certainly an alternative again. It is prone to bad releases every now and then. Also the quality of the code is up now too so the old update jokes are invalid.

  • I can tell you this for certain first hand. I was wondering if there was something wrong with my phone itself, had it not been a busy day there's a serious chance I would have devoted some time finding a better ROM even though Mean ROM has been been pretty good so far, other than annoyances with the Android browser - which is why I put mobile Chrome on despite them being so similar.

  • This is still happening today.

  • by Anonymous Coward

    Here I thought it was crashing because I installed a plugin to make it more like Firefox.

  • I've noticed lately my Windows 7 x64 machine has been having issues with a fraction of my normal "Tab load" on Chrome, crashing or freezing pages, unable to show YouTube pages properly, etc... I believe this might be related to the Sync problem, since I use that.

    These problems started a few weeks ago.

  • for me in the exact way described, so maybe the problem's not fixed yet?
  • And Yet (Score:5, Informative)

    by JustOK (667959) on Tuesday December 11, 2012 @12:56PM (#42252231) Journal

    No one is talking about slashdot being down last night.

    • It was mentioned in another thread today and was modded 'off topic' which to be fair it was. /. Wasn't 'down' though. Th front page was visible, the ads were loading, but, all clicks were ignored except the log in window which you could fill in then a 503 error this morning around 8.30 uk time. /. Was up, but unresponsive. Just like me at 8.30 in the morning in the office

      • by JustOK (667959)

        baloney. couldn't get anything. sourceforge, slashdot and freecode all down etc were down. front page was not visible

        • by Pope (17780)

          baloney. couldn't get anything. sourceforge, slashdot and freecode all down etc were down. front page was not visible

          Turn on your computer next time.

      • It was mentioned in another thread today and was modded 'off topic' which to be fair it was. /. Wasn't 'down' though. Th front page was visible, the ads were loading, but, all clicks were ignored except the log in window which you could fill in then a 503 error this morning around 8.30 uk time. /. Was up, but unresponsive. Just like me at 8.30 in the morning in the office

        Actually the clicks weren't ignored, as the URL bar showed. It's just that the complete URL was ignored, and Slashdot just showed the front page for any and every URL.

        • You're right and my mistake. The URL showed correctly, but the page didn't load.

          The front page was definitely visible at around 8.30am UK. How else was i able to attempt to log in and get a 503?

    • by Waccoon (1186667)

      Slashdot being down doesn't cause my browser to crash.

    • by antdude (79039)

      Yeah, what happened? I noticed new stories were still posted and its uptime did not get reset.

      • Dice was doing some DB server maintenance. They also took down Sourceforge for the same reason (since it shares some of the hardware).

  • One more good reason NOT to use the bookmarks sync function the browser supplies, and instead use a third party alternative.
  • Not in my experience. It crashes daily on me, often more than once a day. And the pseudo-Macintosh sad face "Aw snap..." message wasn't cute the first time. Now it's beyond irritating. Don't hearken back to a 20 year old theme in a vain attempt to try to be cute when you've imploded; at the very least give me some kind of error message. (I suppose that's verboten in the Google "less is more" universe, which is why you also strip down the browser interface so much that not only are semi-relevant buttons and

    • It sounds to me like you might have a corrupt profile? I've never had any of these issues with Chrome (on Linux) and I've been using it since the very first Chromium releases. Switched to the official Google version of Chrome a few years ago when it was released. It's pretty stable for me, but this bug did cause my profile to become corrupted and it would basically crash on startup. I just restored from my weekend system backup and everything was fine again.

  • I'm more surprised that every time I BOOTED Windows there was a Google Chrome crash message box presented. I can assure you that I was never given an option about having Chrome start with Windows and I most definitely did NOT added Chrome to any of my start up stuff. So in addition to showing that actual humans work at Google (well, at least a few) this also exposed the fact that installing Chrome installs something (that may claim to BE Chrome) that normally runs silently every time Windows is started.

  • by Sasayaki (1096761) on Tuesday December 11, 2012 @01:24PM (#42252559)
    //try
    //{
    tabs.sync();
    //}
    //catch
    //{
    //printf("Oops, cloud sync failed. Terribly sorry, Captain. We'll fail gracefully and just make do without.");
    //}
    // James, I fucking told you not to use try-catch statements, they're too slow. The code works and a cloud failure is basically impossible (five nines, baby) so just chill, will you?
    // P.S. /* comment */ is for chumps.
    • by lgw (121541)

      Ha! I think I worke with that guy. I love the C guys stuck in the 80s, who think exceptions are somehow slower than checking a return code and conditionally branching manually at every layer in the stack, and even better the guys who just never check for errors - faster that way.

      At least Chrome is hitting an assertion and not a crash.

      • Maybe you can explain why when you search for "C++ exceptions are NOT slow" on the internet at large, you will only find discussion threads about how slow C++ exception handling is?

        The null hypothesis is that fancy tech would have an overhead. If you're going to claim it doesn't, then you need to provide the independent studies verifying that claim.
        • by lgw (121541)

          C++ error checking was horribly slow when the feature was first added to compilers (late 80's, or early 90's, depending on the compiler), but that got fixed soon thereafter.

          Basically, a catch block pushes a marker on the stack, which is pretty fast. On a throw, the compiler looks for that marker on each stack frame it unwinds, which is not free but still pretty fast - and since it's completely under the compiler's control, it's quite well optimized these days.

          So, it's stil slower than not doing error check

  • So for one day Internet Explorer was actually superior to Chrome?

  • Normally stable? What Chrome are they talking about because 25 is not stable?! They must not be talking about laying out pages wrong because you include that and it's right up there with Safari, lol. And forget super-complicated, code-heavy interactive pages. It will destroy them.
  • This is why I use IE 6!! Rock solid.

The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov

Working...