Very frequently official YouTube website errors out. It works in incognito mode if I try. Which tells me the issue is due to one of cookies / cache / localstorage / application-data on the browser.
If I delete the data, it will start working again, but, I will need to sign-in again in all the gmail-accounts that I use.
I've asked Google (in feedback and over twitter) to look into this but they can't help it.
So, questions:
If anyone has faced this issue, have you figured out why and how to fix it without signing out from accounts?
How do I go about debugging this?
I've taken a memory heap snapshot on chrome but have no idea how to analyze it.
Is sending a memory heap snapshot as part of feedback or to anyone else secure?
So, I figured out the solution. Tried looking at patterns in general in the memory of the browser (localstorage, etc) and found the following in cookies and here's what fixed it (without signing me out of my accounts).
In the image below - there are lines which have gsScrollPos-xyz with a past expiry date. I deleted all of those. I understand that only Google will be able to figure out why that caused the issue and I will send them the feedback.
Related
We're currently using Okta for SSO for our IIS web app and it seems to work fine 99% of the time. However, there is a single user that, when attempting to login, gets this exception about a missing nonce. I've tried reducing all of the variables as much as possible, and I've gotten to the point where two users are trying to login to the same application with the same Okta credentials from the same machine and using the same browser (default chrome freshly installed with no plugins or browsing history). The only differences are where they are RDP'd into the machine from and which windows domain account they're logged in as. Neither of these seem like they should make any difference whatsoever.
However, one user successfully logs in and the other gets this obtuse Exception about a missing nonce.
I've seen several other questions regarding solving this error (IDX21323), and I'm not seeing any actual solutions or explanations that make sense. I've tried a couple of hacks like adding in a new challenge when authentication fails and the auth failed notification exception contains the text "IDX21323", but it doesn't have any effect.
I don't understand the problem well enough to ask a more detailed question because I can't, for the life of me, understand how it could be happening in one case but not the other. I'm not even sure what to investigate at this point.
Morning all.
We have a problem with one of our websites on azure.
It has been running fine but then last night at midnight (exactly midnight) the site started getting HTTP errors and all things went up (requests, cpu time, etc) which is really weird. It is only happening to this site out of the many we have.
I am not sure what is causing it; I am thinking ddos attacks, but I can think of no reason why anyone would target our site.
Here is a screenshot of the monitor board.
Has anyone had experiences like this before? Or may know what is causing it?
There is a very handy tool via the Kudu console for Azure Websites that can help you in figuring out what may be going on with your site. To access it type in a url like this, https://mywebsite.scm.azurewebsites.net/support
You can see live stats for your site. There is also an Analyze tab that you can use to help diagnose issues with your site. Click the Diagnose button and give the tool some time to collect and analyze your logs.
hope that helps.
Mark
I've looked at the various questions on this topic but none of them QUITE fit the problem I'm having.
I've developed an MVC4 app which utilizes DNOA to call into a particular provider (Intuit). All worked perfectly on my local IIS (testing) but when I deployed to Windows Azure I get the proverbial wonderful "strange, intermittent" behavior. Specifically, 99% of the time, the initial sign-in request results in the "No OpenID Endpoint Found" error; however, SUBSEQUENT sign-ins go through without a hitch.
I've added the code referred to here: ServiceManagerCode, to no avail. I've checked and the OpenID URL is correct. I've also attempted to add log4net to see what might be occurring but have been unable to do this correctly, some other answers seem to suggest this returns nothing anyway. I've also asked Intuit but, so far, no responses.
Again, if this wasn't occurring on just the first attempt then there would be numerous relevant posts but with this peculiar behavior I am wary of wasting inordinate amounts of time on a wild goose chase.
Any suggestions, however slight, would be very much appreciated.
I am not familiar with OpenID. Is the OpenID sign in service hosted by you in Windows Azure as well? Please make sure the sign in service has started without any problems, one suggestion is to check the federation configuration. Most federation providers require you to configure the realm and return URL. If they’re not properly configured, the application won’t work.
Best Regards,
Ming Xu.
Since you say that your Azure relying party works reliably after the first failed attempt, perhaps you can workaround it by having your app_start event in your Azure web role call DotNetOpenAuth's OpenIdRelyingParty.CreateRequest method, not doing anything with its result, just to 'prime the pump'?
I am using a UIwebview in my iphone app. It is working fine. No memory leaks when you browse websites, google, news etc. But when you start a video in youtube then it shows me many memory leaks (Under AudioToolBox library). How do I fix them. I imported AVFoundation.h and audioToolBox.h and added these frameworks but I am still getting the same problem.
One more thing, I know apple checks no connectivity condition. Means when there is no internet available or connection problem then user must get some message about connection issues. How do you do that? How do i check internet is connected? what kind of message do you show? Alert or something else?
Do I also need to show activity indicator? How do you show that? can you please reply with sample code?
Apple's designated way of checking connectivity is "try it first". If you get connection problems, you can diagnose with their Reachability suite, available here. But always try first, because Reachability can give false negatives in some situations.
As far as UIWebView's memory problems, I don't think you're in a position to do anything about them. Hope they get fixed in iOS5, I guess. If you want to use AVMediaPlayer for rich media, like Safari and UIWebView do, use it directly, rather than going through WebKit.
We are looking at Windows Azure, but getting a token appears to be hard now, at least that's what I'm seeing in web searches. Anyone tried it or know how to accelerate that process?
Any idea how long this typically takes?
Any help appreciated // :)
You can get one if you are a US resident from CodeProject directly. Go to http://www.codeproject.com and you get your token instantly. They only have 500 so be quick.