rseiler

Microsoft forum feeds (cURL error 47: Maximum (5) redirects followed)

Recommended Posts

Microsoft has once again revamped certain forums, namely their "Answers" forum and ones that go by a different name but look the same (e.g. PartnerSupport). Their Technet forum, which is based on a different system, is not affected AFAIK.

The change kicked in on Aug 1, and that's when the above errors started in the crawl. Examples:

https://answers.microsoft.com/en-us/feed/f/msoffice/msoffice_officeinsider-mso_win10?tab=Threads&status=all&threadType=All
https://partnersupport.microsoft.com/en-us/feed/f/par_servplat/par_winserv?tab=Threads&threadType=All

 

Share this post


Link to post
Share on other sites

Since I noticed the feeds work if you visit them in Chrome I poked around for a bit: 

I'm not sure exactly why they're doing this, but if you add  

-H "cookie: enablePreview=true"

to the curl command the server will give you the XML. Without that it's just the redirects. 

Hopefully this points you in the right direction. 

Share this post


Link to post
Share on other sites
2 hours ago, wesson said:

Unfortunately we can't workaround this on our side. 

See the feedvalidator check for both feeds here and here

That's a little confusing. Don't many (most?  nearly all?) feeds show some kind of issue in the validator?  They did work in Inoreader before this month, and there's a good chance the validator said the same thing then, though I'm not sure.

The feeds shows up fine in Opera. They also seem to work in Feedly.

Share this post


Link to post
Share on other sites

After a little more poking around I noticed the response with the redirect has this header: 

Set-Cookie: enablePreview=true; expires=Fri, 31-Dec-9999 23:59:59 GMT; path=/; secure 

So if Inoreader's crawler accepts that cookie it will work, no more endless redirects. 

Share this post


Link to post
Share on other sites
On 8/4/2017 at 5:28 PM, Firestone said:

So if Inoreader's crawler accepts that cookie it will work, no more endless redirects. 

Unfortunately our crawlers can't save cookies. They are not browsers. 

Share this post


Link to post
Share on other sites

As of three days ago, only one of the two examples above returns the cURL error (not to imply that the one not showing the error is behaving normally).

The output from the validator for both of them is completely different than it was when we first looked at it, further reducing my confidence in anything the validator has to say.

I'm not sure cookies are an actual requirement if Feedly can handle the feeds.

 

Share this post


Link to post
Share on other sites

It looks like Microsoft removed the cookie requirement.

So both work now if I manually use curl without cookies, they're also fine in Inoreader. The validator output changed for the same reason, it's able to fetch the feeds now. The issues it reports are not showstopping bugs,  it seems Inoreader is able to work around them. 

Share this post


Link to post
Share on other sites

5-10 minutes after I posted that last message, the history for the one feed switched off the cURL error. Weird!

We'll have to see if messages actually start to show up normally though. The one that started working again three days ago is not really its old self in that very long stretches go by with nothing, followed by a big burst of messages, many of which are not new. Part of this may have to do with MS still working things out.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now