citibank site seems broken on chromebooks (chrome 68.0.3440.87) |
||||||||||||
Issue descriptionReported by Paul (on cc:) on chromeos-discuss. ---------- Any one tried logging into https://online.citi.com from a Chromebook? I can log in from other systems in Chrome just fine, but on a Pixelbook I get "Error accessing your account information" and a warning about Flash being blocked, even though Flash is set to "always allow" for online.citi.com. Clicking on the "Allow Flash" in the popup has no effect. If I poke around in the console logs I do find the message "Cross-origin plugin content from must have a visible size larger than 400 x 300 pixels", but looking at the source of the webpage I don't see any <object> tags. ---------- confirmed by me (reporter) on Eve beta channel with chrome 68.0.3440.87 (doesn't work) and Linux chrome 68.0.3440.84 (Official Build) (64-bit).
,
Aug 9
This doesn't appear to be a Citibank issue, or at least not beyond the use of flash on the website. The problem is specific to the combination of Chrome & a Chromebook. Other browsers and other platforms work fine, and the error message in the logs reeks of a abuse-detection heuristic gone wrong. Paul J. Ste. Marie | Bug Collector | pstemari@google.com | (425) 272-5757
,
Aug 9
Thank you Paul. Albert, do we need to escalate this?
,
Aug 9
Adding more Flash-knowledgeable people to the conversation. A major bank still using Flash as a primary component on their site is troubling, esp. given the effort Chrome team has been putting into getting people off of Flash. laforge@ do you know if anyone has reached out to Citibank? Even without this bug Flash dependencies mean that their customers on Chrome are going to have a bad time.
,
Aug 13
abodenha@, I am not a flash knowledgeable person - please find a more suitable owner. (I assume because I temporarily owned issue 866645 you may have thought so.)
,
Aug 13
Due to tracking/ abuse, we affirmatively block invisible/ tiny off origin Flash content (~90% of Flash usage was attributable to tracking and viewability). This behavior has been present since June (1) of last year. If this issue is specific to Chromebooks, it's likely that Citi is doing incorrect UA detection and putting ChromeOS down the Flash path. Beyond outreach, which was done in comment #1, I don't foresee any additional changes to Chrome's behavior, particularly in light of the impending deprecation (2). (1) - https://www.chromium.org/flash-roadmap#TOC-PPS-Tiny---Remove-Un-sized-0x0-or-hidden-content-Exceptions-Target:-Chrome-59---June-2017- (2) - https://blogs.adobe.com/conversations/2017/07/adobe-flash-update.html
,
Aug 13
#6 thanks. I have tried to reach Citi in three ways (by voice just now) but I am not confident that this information will be made available to their technical staff. (Citi: if you see this, please leave a note.)
,
Aug 13
The issue is definitely Chromebook-specific. When I load the site on Linux the page doesn't appear to contain any Flash, although the center section of the page is loaded dynamically. Paul J. Ste. Marie | Bug Collector | pstemari@google.com | (425) 272-5757
,
Sep 12
I'm able to consistently reproduce this on m70.x and also called Citibank to notify.
,
Sep 13
I have no idea if this is related in any way, but I found #enable-experimental-web-platform-features breaks citi with the same error. https://bugs.chromium.org/p/chromium/issues/detail?id=883725
,
Sep 26
I have tried multiple Chromebook models and got the same error described by semenzato@. Chrome running on Macbook worked fine. Incognito mode does not help either. I called Citi support couple times and they have been blaming on Chromebooks. This is a serious issue for me that deserves some attention. How do I escalate?
,
Sep 26
A workaround might be to install a UA spoofer to pretend to not be a Chromebook: https://chrome.google.com/webstore/detail/user-agent-switcher-for-c/djflhoibgkdhkhhcedjiklpkjnoahfmg?hl=en-US
,
Sep 26
I've attempted to spoof the UA using the devtools but the website is still broken. I think the problem is more than just the UA string (ex. enable-experimental-web-platform-features is unrelated to the UA string.)
,
Sep 26
Does this deserve to be escalated? I'll tentatively do that, but please downgrade if I am wrong. Comment #6 contains the root cause, which suggests that this needs to be fixed by Citi. Do we have other ways of reaching them? My approach was earnest but weenie and probably not effective.
,
Sep 26
+1 - i'm not able to login to Citi account from Chromebook. Have no issues from Mac.
,
Sep 26
,
Sep 26
Are you sure this problem is on Citi's side and related to flash? I set the UA string to be: "Mozilla/5.0 (X11; CrOS x86_64 10066.0.0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36" and it still works on OS X and fails on Chrome OS. This might be a bug in Chrome.
,
Sep 27
The breakage appears to be the result of the use of a Flash widget that's smaller than what the security provisions in Chrome find acceptable. On the one side, you have Citibank's use of Flash, and on the other, Chrome's refusal to display it. Paul J. Ste. Marie | Bug Collector | pstemari@google.com | (425) 272-5757
,
Sep 27
pstemari, you keep blaming Flash but what is the evidence? Flash is no different on a Chromebook from Win, Mac, Linux.
,
Sep 27
The javascript console contains error messages stating that the flash widget is too small, and the status bar contains a message stating that Flash was blocked. In the meantime the actual display has a big empty frame that appears to want to use flash as its display area. Paul J. Ste. Marie | Bug Collector | pstemari@google.com | (425) 272-5757
,
Sep 27
The only experiment I can think of is to use an old ChromeOS image (that used to work with Citibank) and see if it still works. But absent any Citibank test account nobody is going to debug this on the Chromium side. (And I personally recommend not to use a real account if you depend on the money in your account.)
,
Oct 26
it's not chromebook specific, same thing happens in a linux browser : Google Chrome 70.0.3538.45 (Official Build) beta (64-bit) Revision cbdc32e4334458954e9def214d7e5fa1ca1960eb-refs/branch-heads/3538@{#830} OS Linux JavaScript V8 7.0.276.25 It was working fine a few months ago but I guess citibank changed something. I even complained hard enough that I wasn't able to make a payment online for my costco visa card and they gave me $30 credit.
,
Oct 26
or the constant push to upgrade caused a breaking change in some version of chrome, I want to say it was working pre v65, but not 100% sure
,
Oct 26
so for chrome i keep getting a flash plugin blocked warning and even if i enable it, it still keeps popping up. But the strange thing is that the website works fine in firefox 60.2.0esr (linux) and I don't have flash installed in that at all. So somehow citibank is detecting and trying to use flash when it shouldn't
,
Oct 26
i completely disabled flash in chrome, according to detectmybrowser.com flash isn't detected so now more puzzled
,
Oct 26
Tried it again today and it worked for me. Chrome version is: Google Chrome 71.0.3578.8 (Official Build) dev (64-bit) Revision a2c2b8f674ee3b7372bc87891e04a14a0e51e812-refs/branch-heads/3578@{#14} Platform 11151.4.0 (Official Build) dev-channel eve Firmware Version Google_Eve.9584.174.0 Customization ID GOOGLE-EVE ARC 5067914 JavaScript V8 7.1.302.2 Flash 31.0.0.122 /opt/google/chrome/pepper/libpepflashplayer.so User Agent Mozilla/5.0 (X11; CrOS x86_64 11151.4.0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.8 Safari/537.36 Don't know what changed, but maybe it's fixed or getting closer?
,
Oct 27
confirmed, it's fixed in the latest linux beta : Google Chrome 71.0.3578.20 (Official Build) beta (64-bit) Revision be2b6ea800808b865cd07ab76acf9e2a009c045a-refs/branch-heads/3578@{#254} OS Linux JavaScript V8 7.1.302.4 Flash 31.0.0.122 /home/sabujp/.config/google-chrome-beta/PepperFlash/31.0.0.122/libpepflashplayer.so User Agent Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.20 Safari/537.36
,
Oct 27
|
||||||||||||
►
Sign in to add a comment |
||||||||||||
Comment 1 by semenzato@chromium.org
, Aug 9