Could anyone tell me why Facebook comments are not working properly?
50% of the time no comments show up initially, with multiple errors in the response ajax from facebook.
99% of the time it's impossible to delete a comment without getting a "Bad Parameter" message.
50% of the time it's impossible to post a comment without getting a "Database Down" message.
Is there something I am doing wrong? I have tried copying example code exactly and other methods with no luck...
About: http://wiki.developers.facebook.com/index.php/Comments_Box
Unfortunately the Facebook API and comments box are both unreliable. fb:comments I've found to be particularly bad (often not loading at all) whether using the FBML or the XFBML versions. I think if it works some of the time for you then it's unlikely that it's anything in your configuration or setup.
Sorry.
Related
Is anyone else having this issue with Heroku?
Something went wrong. If you're using an ad-blocker, please try disabling it. You can also check the Heroku Status site to see if there are any incidents impacting our service.
And in the top corner shows Unknown error
Screenshot of the issue
I was using my dashboard normally until 2 days ago, then I started getting this splash and even disabling my adblocker this still remains. And in Firefox I get an eternal loading page (after authentication).
I'm pretty new using Heroku, but it seems that a free user just can't create a ticket or email them anyway... LOL
I'm being able to do what I need through terminal/cmd by now, but eventually a dashboard would be great!
One possibility is that the site is experiencing some sort of problem. It is the website itself and not you.
You can check in Status to see if there is a real problem reporting (but if the problem just happened, it may not happen so quickly)
I suggest that if you really have a question, you write an email to ask.
outage-support#heroku.com
Their reply is very fast (my experience is that they reply after 2 minutes), here is their reply
We're seeing similar reports and looking into this, possibly an incident brewing. We should have a status post shortly: https://status.heroku.com/.
In case a (say login) form POST submission fails and Firefox displays "Try Again" message.
Is there any way to click this "Try Again" automatically or through any settings in Firefox about:config that it clicks it?
Related
"Clicking" the Try Again button is relatively easy. There is an extension that does just that, and lets you set the number of seconds between retries.
The real rub here is that you want to "blindly" retry form POSTs. As we all know, just because you didn't get a response, that doesn't necessarily imply that nothing was changed on the server.
Re-submitting a login form sounds harmless enough, and usually is. But if you imagine forms that result in orders being placed or money being moved, it's easy to understand why browsers have implemented this kind of warning:
This is what you'll see if you enable an extension like TryAgain and a form post fails. It's the same behavior you'd get by pressing F5 yourself. The extension will dutifully try to POST again, but the browser is going to intervene with an alert, and refuse to send the POST until "Resend" is clicked.
This kind of safety feature does a fair amount to protect end-users and developers from poor implementations and network hiccups. However, it's really going to work against what you're trying to accomplish.
That said, if you could figure out a way to modify the extension to detect the alert and somehow click "Resend", you'd be in business. I can't say for sure that this is impossible, put it kind of looks that way, at least for now: this issue was marked as "won't fix", and this issue is still open.
Here is an extension for firefox:
auto reload
but i would warn you. because you could auto send any sensitive data. usually web browsers ask reload because the dont want any sensitive data to be submitted without user discretion.
This is a very strange one for me and I've been battling with it for a while now. I really hope someone can help.
I have a fairly typical MVC 3 Website and I only seem to be getting this problem in IE and Firefox. Chrome plays along nicely. Lucky for me, the majority of our company's clients uses Chrome at the moment.
Problem is at a seemingly random point in time, the browser will automatically redirect me to the Account/LogOff action when I click on a link, and from there it will obviously go back to the Login page. This link will then continue with the same behavior.
I say "seemingly random" because today that link will work, tomorrow it won't and all other (or the majority - I have never had more than on problem link giving this problem at a time) links will be fine. Sometimes restarting the server/dev environment will take care of the problem, other times it won't. The browser will just keep redirecting to LogOff.
I have tried looking at the Referrer URL, but the controller/action being referred to will never be reached. (If I place a breakpoint in the action, it is missed and the next point reached would be the LogOff action)
If I look at the stacktrace when in the LogOff Action, I can't see any info from where the application has come from. I have also tried what was suggested in this page: Posting the Stack Trace on ASP.NET MVC, but I cannot see why I am being redirected to the LogOff action.
The only place I seem to be able to have a breakpoint get hit before hitting LogOff, is Application_BeginRequest in the Global.asax, but can't see where it is going from there on.
My guess is that somewhere along the line, ASPNET Auth decides the user isn't authenticated any more and redirects to the LogOff action. Problem is that the cookies associated with ASPNET Auth all still exist, have data in them and they haven't expired yet.
Anyway, I hope I have given enough info on the problem.
Thanks in advance.
[Edit]
OK, so I might have gotten a step closer. I came across this link and looked to see what is happening in my Application_AuthenticateRequest in my global.asax.
I am not quite sure why, by when I click on a link, Application_AuthenticateRequest gets accessed 3 times. When a link works (as in I can follow it and it doesn't log me out), the value of the .ASPAUTH cookie stays the same. I checked this by adding a breakpoint and a watch over
HttpContext.Current.Request.Cookies[".ASPXAUTH"].Value
When the link does not work, the first time the cookie has a value, then the other two times it is null. Thus, because the ASPXAUTH cookie is null, the system automatically redirects to the LogOut action.
If I consider the solution they posted in the link, I am not sure if this applies to me. As far as I can tell, the encrypted cookie is still small (as in a few hundred characters long) and not close to 4096 bytes. Also, I have only 3 cookies going at the time I when tested the broken link and I have a maximum of 5 cookies at any given time.
Any idea?
OK, so I had a hunch about cookies expiring. So I looked at whether there is a way to keep (force) cookies in Forms Authentication alive and that led me to http://www.codeproject.com/Articles/221889/How-to-Generate-Machine-Key-in-IIS7
The only way I could test this theory was to keep working and debugging the site as normal. (And that's why it took me so long to post this answer.) Since I introduced this solution it seems that the problem has been solved.
Interestingly I spoke an Architect the other day, with 20 years dev experience, about my problem. He looked at my code and is convinced this is a bug in the Forms Authentication code.
I hope this helps some people who are experiencing the same problem I have.
I log all visits to detect bots. Got sometimes visit with ip from another country and user agent like "Java/1.6.0_32". Is it probably a bot?
It's definately not one of the major browsers. It's probably a screen-scraper - you could probably take it as a compliment that somebody finds the content provided by your site to be useful.
"You could probably take it as a compliment that somebody finds the content provided by your site to be useful..."
You might be right, but I`m not 100% sure.
An unrecognized bot could be trying to steal you content, attempting to spam links or scanning for vulnerabilities...
I've checked Botopedia.org and found no reference for this user-agent.
Might be fine, but it seems fishy to me.
Installed a captcha on my blog, been good up until now.
There have recently been a bunch of legit-at-first-glance-but-actually-spam entries along with stuff like this:
message: IDevY7 sdbgztbczgpj
from: fmfwls
The IP changes per submission and they must be correctly filling in the captcha. Is my only option manual approval of comments?
The thing is all captchas can be beaten by bots now, even reCaptcha which is a really great solution (Google) Try reCaptcha, you might have better results with that then what you are using now. I found it kept most things out when I was using it, but stuff still did get through.
Other than that look at some other non-captcha spam fighting solutions. Have you looked into akismet? Its a great server side solution that looks at the content and attempts to identify it as spam. Perhaps if you combine the two you may catch the majority of it.
There are various other tricks you can try too. I'd definitely recommend checking out akismet though.
Which captcha system are you using? Has it been broken? Have you tried recaptcha?