Please navigate to chrome:crashes and list the report IDs that correspond with the crashes you're experiencing.
If none are listed you may need to enable "Automatically send usage statistics and crash reports to Google" in chrome:settings.

I have just enabled chrash reports and sending usage statistics.

I'll see if something shows up the next time I observe the issue, but I don't think there is any crash involved at all. 

As I already mentioned in my report, the contents of the tab vanish, but then they become visible again, and it doesn't seem like anything is reloaded. Certainly not in the case when what vanished was the content of the Style panel (unless that's rendered by a separate process that may have crashed and then restarted)

Thank you for providing more feedback. Adding requester "alancutter@chromium.org" for another review and adding "Needs-Review" label for tracking.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Sorry, I mistakenly interpreted tab vanishing as a crash, it's unlikely that a crash is occurring.

With the unreliability of being able to repro this it seems very difficult to investigate. Given that it seems to be triggered by scroll I'm assigning to scroll team in case they have any ideas for debugging this issue.

I also have this problem. It happens often when I am browsing Instagram.

Another page where I experienced the issue was this:
https://support.google.com/chrome/answer/96817

For me, it is easy to reproduce the issue by visiting the page indicated above and rapidly scrolling up and down several times.

Hi,
I jus ran google-chrome (51.0.2704.79) in a terminal and noticed I am getting the following message roughly at the same time as the page content goes blank:
[1:1:0604/031838:ERROR:PlatformKeyboardEvent.cpp(117)] Not implemented reached in static PlatformEvent::Modifiers blink::PlatformKeyboardEvent::getCurrentModifierState()

I am getting this problem extensively on chromium 51.0.2704.79 (64-bit) on Arch. I can pretty much cause it on demand within a few secs or so just by scrolling up/down quickly on my touchpad (using libinput). Finding this bug and seeing comment #6 just above, I ran chromium from the command line and find I get exactly the same message as reported in that comment.

I have swapped between xf86-video-intel and native xf86-video-modesetting but bug occurs with both. It even occurs in tabs displaying PDF.

Further to my comment #7 above, since I can cause this problem on demand just by moving my touchpad quickly it makes chromium essentially unusable for me. So I re-installed each previous chromium package I had cached on my Arch notebook until I hit the last version that works:

chromium-51.0.2704.79-1-x86_64 = BAD
chromium-51.0.2704.63-1-x86_64 = BAD
chromium-50.0.2661.94-1-x86_64 = GOOD

I.e. this problem appears to have been introduced since chromium v51. I reinstalled back and forward a few times to ensure this is the case. The problem only occurs in v51.

I can see this happening on 53.0.2760.0 canary (64-bit)on Mac 10.9.5
In my case this happens several times (~20) per day when the developer tools is open.

To notice that even if the page is blank by moving the mouse around the cursor appears and disappears, so probably it can still interprets the page and links but nothing is visible.

Actually, after another day I have to report that this problem even occurs on chromium v50, although much less often and I certainly can't reproduce it on demand.

OK, I have worked out how to stop this bug occurring. Open chrome://flags and disable smooth scrolling. I went back to current chromium-51.0.2704.79-1 where I can immediately initiate this problem by rapid scrolling via my touchpad but now the bug does not occur.

OP and others, please confirm this bug disappears when you disable smooth scrolling.

Note that the message in comment #6 above is not related to this problem and is benign as per https://bugs.chromium.org/p/chromium/issues/detail?id=538289.

I can confirm that the bug does not appear when smooth scrolling is disabled
(using 51.0.2704.79 on debian sid amd64).

With smooth scrolling enabled the bug appears constantly (approx. every other minute), rendering the browser useless.

Stock Arch chromium updated to 51.0.2704.84-1 today but bug still there.

 Issue 609528  has been merged into this issue.

Is it possible we're somehow not updating some compositor state while doing animated scrolls so that it doesn't know that new tiles are in the viewport?

It's seems this is not happening anymore (or maybe not so often) on  53.0.2763.0 canary (64-bit) WITHOUT disabling the smooth scrolling

Possibly related to  issue 616308 ,  issue 616086 , or  issue 617264 ?

Does it only happen with touchpads, or also with real mouse wheels?

Does it help to pass --disable-main-frame-before-activation?

> Does it only happen with touchpads, or also with real mouse wheels?

In my case I'm pretty sure it happens with real mouse wheels, meaning when I am using a real mouse with a real mouse wheel and not using the touchpad (I almost always use external mice), though I always do have the touchpad enabled, so it is theoretically possible, though extremely unlikely, that the touchpad could be triggering it.


I should perhaps also note that, unlike others, I've always observed this quite rarely (less than daily), and I have never been able to reproduce it at will.

I tested chromium 51.0.2704.79 with --disable-main-frame-before-activation (and smooth scrolling enabled), and it didnt seem to make much of a difference.

Whats really weird though is that afterwards, running chromium normally (and with smooth scrolling enabled), I havent been able to reproduce the issue

Update for comment #20:

After a while the bug began appearing again, with the same frequency (approx. once every other minute) as before trying --disable-main-frame-before-activation.

I can initiate this problem within a few seconds so I just tried --disable-main-frame-before-activation using my touchpad and the problem still occurred immediately. I also tried with a mouse using the wheel and the problem still occurs immediately, both with and without --disable-main-frame-before-activation.

Chromium updated from version 51.0.2704.84-1 to 51.0.2704.103-1 on Arch today but this bug still exists.

Is this happening because somehow having the finger on the touchpad(while scrolling) at all times  keeps the current view position which interferes with the scrolling somehow  as some kind of  position-(re)set race between the two? (this is just a guess from a position of not knowing anything, so safe to ignore)

Thanks for mark in Comment 11 - disabling smooth scrolling as a workaround!

@24 I don't think so:
I observe the issue even using a regular mouse (and with my fingers nowhere near the touchpad)

Updated to 52.0.2743.82 on debian sid amd64 a couple of days ago.
This version has not exhitibed the tab blanking behaviour so far (with smooth scrolling turned on)

Given I have reported here that I can reproduce this bug immediately on call, I just tried with current Arch chromium version 52.0.2743.82-1 and see that the bug does not occur anymore. However I suspect it was a hacky fix because I see occasional screen flashes as I rapidly scroll. They occur at about the same rate I used to see the tab completely blank. I am guessing some kind of sub-optimal hammer fix was applied?