Only one of two browsers finds physicsforums

[symbolipoint]

Google Chrome will not let me visit www.physicsforums.com, instead showing me:
This site can’t be reached
www.www.physicsforums.com refused to connect.Try:

• Checking the connection
• https://www.physicsforums.com/chrome-error%3A//chromewebdata/#buttons

ERR_CONNECTION_REFUSED

When I check with downforeveryoneorjustme site, I am told, "Its Just You" (meaning, "me"); that physicsforums is up.Firefox let's me visit and sign into physicsforums just fine.

 

[Greg Bernhardt]

Try to access PF in Chrome in incognito mode
https://support.google.com/chrome/answer/95464

 

[symbolipoint]

Try to access PF in Chrome in incognito mode
https://support.google.com/chrome/answer/95464

That's is what I did (and forgot to say).
I will TRY to visit physicsforums in Google Chrome's regular mode.

 

[symbolipoint]

Chrome(Google) is now letting me into physicsforums, either (BOTH) regular mode and incognito mode.

Completely unknown problem. Completely unknown return to good functionality of visiting physicsforums. Mystery.

 

[Borek]

www.www.physicsforums.com refused to connect.

Too many wwwwwwwwws?

 

[DaveE]

OK, I just had to try it. I've never done "www.www." before. It didn't work, of course.

• Try running Windows Network Diagnostics.

DNS_PROBE_FINISHED_NXDOMAIN

 

[symbolipoint]

Too many wwwwwwwwws?

That was a typing mistake in the forum post.

 

[symbolipoint]

OK, I just had to try it. I've never done "www.www." before. It didn't work, of course.

This site can’t be reached
www.www.physicsforum.com’s server IP address could not be found.

• https://www.physicsforums.com/javascript%3AdiagnoseErrors().

DNS_PROBE_FINISHED_NXDOMAIN

That was just my typing mistake in my forum posting.

 

[Klystron]

I had the same problem as @symbolipoint a few days ago while using Chrome incognito to login. For whatever reason the partial name matching on the browser address line appended www.physicsforums.com to a 'leftover' www. I too thought the PF servers were down until 1) loging in using Firefox showed no problem with PF, then 2) finally noticing the www.www.phy... error.

{Edit 04012020: typo: the 'extra' www. appends to the correct www.physicsforums.com address.
When I first posted this I suspected a buffer flushing error but now think the domain software does not realize the browser already (properly) appended www. to the PF address under some conditions but irrespective of browser choice.}

 

[symbolipoint]

I had the same problem as @symbolipoint a few days ago while using Chrome incognito to login. For whatever reason the partial name matching on the browser address line appended www.physicsforums.com to a 'leftover' www. I too thought the PF servers were down until 1) loging in using Firefox showed no problem with PF, then 2) finally noticing the www.www.phy... error.

I am not sure why any of this would involve a doubling of the "www" part. This was, for me, only a forum posting mistake - NOT what I put into the browser to try to reach www.physicsforums.com. Whatever the previous site failure I was having today, the problem has now stopped. ( I wonder if the trouble involved use of this item in "Bookmarks" for the browser or the password manager combined with "Bookmarks"?)

 

[Klystron]

I had the same problem as @symbolipoint a few days ago while using Chrome incognito to login. For whatever reason the partial name matching on the browser address line appended www.physicsforums.com to a 'leftover' www. I too thought the PF servers were down until 1) loging in using Firefox showed no problem with PF, then 2) finally noticing the www.www.phy... error.

I am not sure why any of this would involve a doubling of the "www" part. This was, for me, only a forum posting mistake - NOT what I put into the browser to try to reach www.physicsforums.com. Whatever the previous site failure I was having today, the problem has now stopped. ( I wonder if the trouble involved use of this item in "Bookmarks" for the browser or the password manager combined with "Bookmarks"?)

Sounds logical as I have PF bookmarked. I assumed some Chrome string buffer did not flush or overwrite correctly, perhaps after an automatic update (?). The problem has not repeated.

 

[symbolipoint]

Sounds logical as I have PF bookmarked. I assumed some Chrome string buffer did not flush or overwrite correctly, perhaps after an automatic update (?). The problem has not repeated.

No way to know for certain, but the problem WAS only several minutes after finishing Windows 10 Updates; and the problem for only a small number of hours in length - and ONLY with physicsforums.

 

[sysprog]

Before the standard DNS (plural) started not requiring 'www' in the address, some browsers would auto-prepend it. If you included 'www' in a bookmark, you could have encountered the 'www' being prepended to a URL that already had it, whence the 'www.www', which obviously the DNS would not find.

On a side note, 'w' is the only letter in the alphabet that requires more than 1 syllable to say, and Marilyn (Mach Vos Savant) asked her readers (of her Ask Marilyn Q&A in the 'Parade' newspaper magazine) to suggest something better than having to say 3x3 prefatory syllables every time we say a website address. Getting rid of the requirement of the 'www' prefix somewhat rendered moot her concern, but the obnoxiousness of the 3-syllable name for the letter remains.

If we were to call the letter 'w', 'vay', it would bother the French, because they already call the letter 'v', 'vay' -- and if we call it 'we', it would ' make 'www' sound like part of 'this little piggy' -- but I think we should be able to say 'w' with 1 syllable, like all the other letters, so I propose 'way' -- even though that's a word in English -- that's not unprecedented in the alphabet -- some people who are texting (SMS) use 'c' to mean 'see', and 'u' to mean 'you', and 'r' to mean 'are', for example . . .

 

[mfb]

I have heard people calling them "the double-u-s", 4 syllables in total.

If we were to call the letter 'w', 'vay', it would bother the French, because they already call the letter 'v', 'vay'

That doesn't seem to be a problem. English "e" is pronounced like the French and German "i".
French "v" sounds like German "w".

(That also means www is very short in German)

 

[sysprog]

I have heard people calling them "the double-u-s", 4 syllables in total.That doesn't seem to be a problem. English "e" is pronounced like the French and German "i".
French "v" sounds like German "w".

(That also means www is very short in German)

Well, in English, WC is an abbreviation for 'water closet', and the same abbreviation is used in France, and it's pronounced something like 'doo-bla-vay-say' -- still 3 syllables for the 'w' -- I remember a sign that said (skipping the correct diacritals) 'priere de laissez les WC dans l'etat de propretre ou vous souhaitez les trouvez en entrant' -- which I translate as 'please leave the restroom in the state of propriety in which you would seek to find it upon entering' -- I still think no letter's name should require more than 1 syllable -- I thank that 'w' should not be an exeption -- I think that we should fix that. I reiterate my suggestion that we adopt 'way' as the name for 'w'.

 

[symbolipoint]

Posts #13, 14, 15 have drifted off into language/letters/linguistics. My original question was about my browser's temporary failure to reach the physicsforums site. I am still interested in any other information or guidance about the original trouble I reported.

 

[sysprog]

Posts #13, 14, 15 have drifted off into language/letters/linguistics. My original question was about my browser's temporary failure to reach the physicsforums site. I am still interested in any other information or guidance about the original trouble I reported.

I apologize for the inadvertent hijack, which I began with "On a side note,"; I didn't mean to derail your thread -- I'll reiterate my hypothesis regarding your concern:

Before the standard DNS (plural) started not requiring 'www' in the address, some browsers would auto-prepend it. If you included 'www' in a bookmark, you could have encountered the 'www' being prepended to a URL that already had it, whence the 'www.www', which obviously the DNS would not find.

 

[symbolipoint]

Happening Again.
Chrome(Google) Incognito Mode, laptop Windows 10.
Type into browser to test once more instead of the password manager:

Fine - the normal forum page loads.

Test again using the BOOKMARK for PF:
Fine - the normal forum page loads.

I do not understand, but as before, the problem was only temporary.

---EDIT---- wrote to soon. Problem returned. This seems to be something about the browser, Google Chrome. I will have to trace my exact steps and then list their sequence.

 

[symbolipoint]

Continuing from post #18, not sure why, but Incognito Chrome is adding an extra "www." whenever I try to reach PF using the stored BOOKMARK. When I look at the actual bookmark information in Bookmarks Manager, the address is exactly as it should be, h t t p : // www.physicsforums.com

I may try to delete the bookmark and then visit PF and add this as bookmark anew.

Not have the problem in Firefox.

 

[symbolipoint]

I may try to delete the bookmark and then visit PF and add this as bookmark anew.

That seems to work; FOR NOW.
(Let me see what happens later.)

 

[Klystron]

Continuing from post #18, not sure why, but Incognito Chrome is adding an extra "www." whenever I try to reach PF using the stored BOOKMARK. When I look at the actual bookmark information in Bookmarks Manager, the address is exactly as it should be, h t t p : // www.physicsforums.com

I may try to delete the bookmark and then visit PF and add this as bookmark anew.

Not have the problem in Firefox.

Does the 'extra' www append only after starting Chrome browser for first time, everything else being as you describe?

I have not seen the problem again but instead of using Bookmark PF, I let Chrome 'remember' the sign in data. Unlike Firefox, Chrome must be prompted with one character of the username then displays an elusive username+passwd fill. I use my touch pad (mouse) to precisely click the field and then Chrome auto-fills username and passwd. Click the 'stay logged in button', press return and "Voila!": PF.

When I try Chrome auto-fill using touch screen, instead of the touch pad, Chrome opens a useless Settings window. This begs the question: why use Chrome instead of Firefox for PF?

No specific answer but Chrome Incognito seems to misbehave less with the latest PF release. Perhaps @symbolipoint has better data. As things stand I use Chrome for PF and Firefox for everything else.

 

[symbolipoint]

Klystron,
I have not yet checked what you said, but I use (when in W10), Google Chrome incognito mode, never with any touchscreen; laptop for my W10 computer. I never sign into Chrome.

 

[Klystron]

Wait one. I was able to duplicate the problem using Mozilla Firefox browser, also in private mode.

I brought up www.physicsforums.com in the address window by entering search string 'phy' into my search engine, currently duck duck do go (whatever ;-). Pressed return on the above (correct) string in the address window. Connection timed out with address error www.www.physicsforums.com

I am running Windows 8.2 on an Intel-7 Quad-core laptop but do not think that matters. Somewhere an extra www is appended to the stored address not related to the platform or browser. Some earlier posters may be correct. I note an earlier thread on DNS errors?

My point about touch screen was that the Chrome PF login form is either ultra sensitive to position else misbehaving. A finger touch encompasses more pixels than pointing with a mouse arrow. Neither the mouse/touch-pad nor the touch-screen on my laptop have any errors or problems. This is an adjoint issue to logging into PF. Thanks.

 

[sysprog]

Klystron,
I have not yet checked what you said, but I use (when in W10), Google Chrome incognito mode, never with any touchscreen; laptop for my W10 computer. I never sign into Chrome.

(underlining added)

I tend to keep 2 gmail sign-ins active. I don't worry about my primary and secondary gmail email addresses being known -- they're, respectively, 24x7 and 7x24 in binary 1 and 0 numerals -- pretty much everyone on PF could find me with that description, and spammer bots probably won't get it.

I've not encountered the described problem; however, I think that the idea of deleting the bookmark and then reinstating it is a good idea, but I think that, and I'm just guessing here, in between the deletion and reinstatement you might try adding a different bookmark, so that Windows or Chrome isn't as likely to simply revert, but instead is more likely to emplace the bookmark anew.

Here's a site that is in my view worth bookmarking -- some of the bookmarklets on this site are also in my view worthy of being among a person's bookmarks: https://www.squarefree.com/bookmarklets/zap.html

 

[Tom.G]

To see if it is intrenal or external to your computer you could use a Protocol Analyzer to look at/log all incoming and outgoing network traffic.

Wireshark is a popular and free one: https://www.wireshark.org/

 

[symbolipoint]

I tend to keep 2 gmail sign-ins active. I don't worry about my primary and secondary gmail email addresses being known -- they're, respectively, 24x7 and 7x24 in binary 1 and 0 numerals -- pretty much everyone on PF could find me with that description, and spammer bots probably won't get it.

I've not encountered the described problem; however, I think that the idea of deleting the bookmark and then reinstating it is a good idea, but I think that, and I'm just guessing here, in between the deletion and reinstatement you might try adding a different bookmark, so that Windows or Chrome isn't as likely to simply revert, but instead is more likely to emplace the bookmark anew.

I have no understanding what you said there, especially the first paragraph. Maybe at least you could re-word the second paragraph for clarity.

 

[Not anonymous]

I had the same problem as @symbolipoint a few days ago while using Chrome incognito to login. For whatever reason the partial name matching on the browser address line appended www.physicsforums.com to a 'leftover' www. I too thought the PF servers were down until 1) loging in using Firefox showed no problem with PF, then 2) finally noticing the www.www.phy... error.

I noticed a similar problem in Firefox for a few weeks but the problem has vanished now. I was using private mode and Physicsforums was not in bookmarks. I would type physicsforums in location bar and then CTRL+ENTER, a shortcut that is supposed to take care of prepending "www" and appending ".com" to create the correct address. Strangely, Firefox would convert physicsforums to www.www.physicsforums.com (notice the extra "www") and result in "can’t connect to the server at .." error. CTRL+ENTER continued to work correctly for other websites during the same period, so I suspect there was some problem in website configuration or in some domain service settings that specifically affected Physicforums.

 

[symbolipoint]

Strangely, Firefox would convert physicsforums to www.www.physicsforums.com (notice the extra "www") and result in "can’t connect to the server at .." error. CTRL+ENTER continued to work correctly for other websites during the same period, so I suspect there was some problem in website configuration or in some domain service settings that specifically affected Physicforums.

That was what I was finding, but using Google Chrome icognito mode. Other websites were fine.
(notice what I bolded in your quote.)

 

[sysprog]

I have no understanding what you said there, especially the first paragraph. Maybe at least you could re-word the second paragraph for clarity.

I'm guessing at the logical operational content of proprietary code for which I don't have source code access rather than examining the machine language and figuring out for certain what it's doing. I think it's probably doing something like caching, as in oh, you deleted this bookmark, and now you want it back -- no I don't; I want it to be be newly created, because my reason for deleting it was that it was created in a screwed-up manner previously. If I ask for it back right after deleting it, I think that I'll probably get restoration of the same screwup, but if I do a different bookmark in between the deletion and the re-creation, then maybe when I ask for re-creation of the screwed-up bookmark, I'll get a newly-created version of it that doesn't include the same screwup. instead of just getting a reinstated copy of the bookmark that still includes the same screwup.

 

[symbolipoint]

I'm guessing at the logical operational content of proprietary code for which I don't have source code access rather than examining the machine language and figuring out for certain what it's doing. I think it's probably doing something like caching, as in oh, you deleted this bookmark, and now you want it back -- no I don't; I want it to be be newly created, because my reason for deleting it was that it was created in a screwed-up manner previously. If I ask for it back right after deleting it, I think that I'll probably get restoration of the same screwup, but if I do a different bookmark in between the deletion and the re-creation, then maybe when I ask for re-creation of the screwed-up bookmark, I'll get a newly-created version of it that doesn't include the same screwup. instead of just getting a reinstated copy of the bookmark that still includes the same screwup.

I will recheck what I newly bookmarked when I operate that computer again. As a reminder, I did try deleting the bookmark; and then visited www physics forums com again going through the address bar, and set the new bookmark; and upon checking the use of this new bookmark, visiting physicforums was successful. NEXT is to try to check a couple more times. ON THAT MACHINE.

 

[sysprog]

I wanted to suggest also doing a setting of another bookmark, to another site, after deleting the PF bookmark, and before re-creating your PF bookmark, but hey, bookmarking the PFs is, in my view, a very good idea, because, at least in my view, it's a great site, and I think that I'm not just being a kiss-up in saying that -- I'm genuinely happy to be allowed to be here. I also suggest that you should omit the www from your bookmark -- it's no longer required, and if the code somehow prepends it, you're less likely to get redundant inclusion of it if you haven't typed it in, so maybe set your bookmark like this: http://physicsforums.com

 

[Klystron]

Updated and corrected my original post in this thread. I think we agree despite the thread title that this problem is browser independent. I have not had opportunity to try all the tools suggested by members but classify this as a general auto-fill problem:

Under some conditions an extraneous www. is appended to the fully formed PF address.

Domain name services cannot recognize or handle the www.www. Strange this gets passed through.

 

[symbolipoint]

I wanted to suggest also doing a setting of another bookmark, to another site, after deleting the PF bookmark, and before re-creating your PF bookmark, but hey, bookmarking the PFs is, in my view, a very good idea, because, at least in my view, it's a great site, and I think that I'm not just being a kiss-up in saying that -- I'm genuinely happy to be allowed to be here. I also suggest that you should omit the www from your bookmark -- it's no longer required, and if the code somehow prepends it, you're less likely to get redundant inclusion of it if you haven't typed it in, so maybe set your bookmark like this: http://physicsforums.com

That would require some editing but is worth trying.

 

[mathman]

This may not help, but I had no problem (using one www). My browser is Avant via Chrome, but my test was with Chrome directly.

 

[DaveC426913]

Fun fact:

The World Wide Web has the dubiously unique distinction of (probably) being the only word in human language whose abbreviated form is (much) longer than its expanded form.

double-yoo double-yoo double-yoo (9 syllables)
World Wide Web (3 syllables)

 

[OCR]

This works for me. . .

Physics Forums | Science Articles, Homework Help, DiscussionThis works, also. . .

Physics Forums | Science Articles, Homework Help, Discussion
Lol, this doesn't seem to work. . .

Physics Forums | Science Articles, Homework Help, DiscussionOh, almost forgot. . .

.

 

[symbolipoint]

So far the remedy seems still to have fixed the trouble.

This remedy was as I described:

1. Deleted the physicforums bookmark from Google Chrome
2. I believe I did visit some other site or two, and then
3. Made new bookmark upon another visit to physicsforums

Now, visits to physicsforums still seem to be working.

 

[MikeeMiracle]

This all stems because people are lazy to write the www part and so browser manufacturers started to put it automatically in front of anything we type in the browser bar. I wish technology would just do what it's asked and not try and "Guess" what it thinks we wanted it to do.

 

[DaveC426913]

This all stems because people are lazy to write the www part and so browser manufacturers started to put it automatically in front of anything we type in the browser bar. I wish technology would just do what it's asked and not try and "Guess" what it thinks we wanted it to do.

Well, what technology is asked to do is to take on tasks that are onerous for humans to do. That's the raison d'etre of most technology.
(Removing steps that are repetitive is one such onerous task).

Such time-saving tasks are always a compromise.
To take the initiative helps most of time and hinders only rarely.
To not take the initiative hinders most of the time and helps only rarely.

On average, more time (a lot more) is saved by taking the initiative to eliminate repetitive steps.

The rule in software design is "design for the 80% use case".If designers weren't continually trying to make a better experience for humans, we'd still be coding in machine language.

 

[rcgldr]

I had the same problem here, Chrome in incognito mode or IE in private mode, Windows 7 Pro 64 bit. Using links that I haven't changed in years, and no recent updates to Chrome that I'm aware of. I edited the link, removing the trailing "/", from "https://www.physicsforums.com/" to "https://www.physicsforums.com", which solved the problem. I haven't encountered this issue with any other web sites (yet).

 

[Greg Bernhardt]

I had the same problem here, Chrome in incognito mode or IE in private mode, Windows 7 Pro 64 bit. Using links that I haven't changed in years, and no recent updates to Chrome that I'm aware of. I edited the link, removing the trailing "/", from "https://www.physicsforums.com/" to "https://www.physicsforums.com", which solved the problem. I haven't encountered this issue with any other web sites (yet).

There is no slash in the domain name URL. Add the slash and it's technically a different page. This is as expected. However, it should redirect to non-slash.

 

[OCR]

There is no slash in the domain name URL.

Interesting. . .

More interesting. . .

Why the slash on that one ?No slash in the address bar, either. . .

OMG. . . you'll get my passwords. .

Lol...[/size]

.

 

[symbolipoint]

Looking in Google Chrome and in Bookmarks Manager, and picking physicsforu site, Edit Bookmark, what is now listed while AFTER I had newly set this bookmark as described recently and been working well:

https://www.physicsforums.com/

I now also wonder, if the possible lack of the "s" for "secure" was important before resetting the new bookmark? I have no way to see what was stored before since I did not look carefully at that detail.

 

[darth boozer]

I have just tried using Chrome by typing in www.physicsforums.com into the address bar and got the same result as in post #6. Trying it again in a new tab loaded normally. Copying and pasting the displayed address into Notepad showed the added www. in the first case. Subsequent attempts loaded normally until I deleted the browsing history and tried again. Trying with the https:// included loaded normally.

 

[pbuk]

@Greg Bernhardt there is a problem in the configuration of your HTTPS redirect. An HTTP request to http://www.physicsforums.com/ (see note below) is being responded to with the following (which is all OK apart from the additional www. in the last line):

HTTP/1.1 301 Moved Permanently
Connection: Keep-Alive
Content-Type: text/html
Content-Length: 706
Date: Wed, 08 Apr 2020 15:05:43 GMT
Server: LiteSpeed
Location: https://www.www.physicsforums.com/

(see also Firefox developer tools capture at the bottom of this message).

Note that a terminating slash is ALWAYS present in a valid HTTP request to an HTTP host regardless of what is displayed in the browser bar. This is because the underlying (HTTP/1.1) message has the following structure:

GET / HTTP/1.1
Host: www.physicsforums.com

See https://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.2.1 "... If the abs_path is not present in the URL, it MUST be given as "/" when used as a Request-URI for a resource".

 

[Greg Bernhardt]

Note that a terminating slash is ALWAYS present in a valid HTTP request to an HTTP host regardless of what is displayed in the browser bar. This is because the underlying (HTTP/1.1) message has the following structure:

It is present, but the browser strips it out.

I see the redirect adding the second www. Not sure why that is happening, need to test.

Here is my code:

RewriteCond %{HTTPS} off [OR]

RewriteCond %{HTTP_HOST} !^www\. [NC]

RewriteCond %{HTTP_HOST} ^(.*)$  [NC]

RewriteRule (.*) https://www.%1/$1 [R=301,L]

%1 must be returning the www too

 

[Greg Bernhardt]

I removed %1 and manually entered physicsforums.com in and now the header is fixed. Can everyone else test and see the problem is resolved?

 

[pbuk]

I removed %1 and manually entered physicsforums.com in and now the header is fixed. Can everyone else test and see the problem is resolved?

Yes that should work, and you can now remove the third line too as you are no longer using that capture.

 

[darth boozer]

Yes, working normally now,

 

[Klystron]

I removed %1 and manually entered physicsforums.com in and now the header is fixed. Can everyone else test and see the problem is resolved?

Hi Greg,
I have waited to reply until network issues at my end were fixed. All my browsers -- Chrome, Firefox, and Explorer -- work fine logging into PF in private / incognito mode. The extra www prepend has not reoccurred even without deleting old bookmarks.
Good fix!

My network problems involved extremely slow Inet downloads, glacial handshaking and ultra slow data transfers at most common websites such as digital newspapers, weather maps and initial Youtube requests. PF performed rather well by comparison. My carrier, Cox Communications, claimed that increased Inet usage due to more users at home was overwhelming their systems nationwide.

Under social distancing guidelines, the Cox techs could not enter my home. I took pictures of the inside wiring and equipment. We troubleshot on the front porch. Cox replaced external parts and exchanged my network router with their latest 5G model.

After updating my wireless devices to use the new router and rebooting my laptop with the latest Microsoft patches; my computer, digital TV and cell phone internet connections are as fast as I have seen even with increased traffic. Not certain if this is because my old router was deteriorating or a case of "squeaky wheel gets the grease" but I can finally appreciate the 5G network.

PF runs super-fast today. I will essay some empirical tests tomorrow.