Re: [WebDNA] TCPConnect example doesn't work

This WebDNA talk-list message is from

2017


It keeps the original formatting.
numero = 113466
interpreted = N
texte = 1061 Ken, Most likely its the Host header that you need to populate. Without it= , the remote server won't be able to route the request to a host based virtu= al host. If you are communicating with a highly secure service such as bank= or payment gateway, you will also likely need to include the User-Agent and= maybe a few others. =20 Mike > On Mar 10, 2017, at 5:27 AM, Kenneth Grome wrote= : >=20 > Mike, >=20 > My problem with GET tcpconnect is not that it doesn't follow > redirects, it's that it returns a 404 error when the specified > page exists. >=20 > POST tcpconnect does not have this problem, at least in the way I > coded mine. So now I believe that GET tcpconnect might work if I > ignore the bad example in the docs, and instead use the correct > but undocumented syntax. >=20 > Your example contains several parameters inside the tcpsend > context, whereas the GET example on webdna.us does not include > any. So maybe some of your tcpsend parameters are required in > order for GET tcpconnect to work properly? >=20 > Regards, > Kenneth Grome > WebDNA Solutions > http://www.webdnasolutions.com > Web Database Systems and Linux Server Administration >=20 >=20 >=20 >> On 03/09/2017 11:55 PM, Michael Davis wrote: >> Ken, >>=20 >> I've run into this problem in the past. You can use cURL to >> pre-flight your endpoint. I have a basic function that works for >> the situations I need it for and you're welcome to grab it and use it >> http://www.network13.net/webdna_code_curl_redirects.dna >>=20 >> Maybe WebDNA TCPConnect will be improved to follow redirects in >> the future, but for now, we DIY, which I don't really have a >> problem with, so long as it works. >>=20 >>=20 >> Mike >>=20 >>=20 >>> On Mar 9, 2017, at 11:10 AM, Kenneth Grome >>> > wrote: >>>=20 >>> I haven't used GET tcpconnect in a very long time so maybe you're >>> right. I don't really remember having any problems with it >>> before, but something is definitely wrong when the file clearly >>> exists in the exact location specified yet it won't GET it. >>>=20 >>> I do know that POST tcpconnects often require the host to be >>> specified in two places: in the tcpconnect tag and also inside the >>> tcpsend context. >>>=20 >>> =46rom what I recall when I learned about this 'two hosts' issue a >>> long time ago, this was a 'sometimes' requirement that depended >>> upon the server receiving the request. It's been a while since I >>> ran into it, so maybe putting the host in both places is a >>> requirement in all POST tcpconnects now. >>>=20 >>> Unfortunately there is almost nothing in the online docs that >>> mentions any of this. I would hate to be a new user who doesn't >>> know all these undocumented requirements because after trying and >>> failing to get the published examples to work properly I might >>> conclude that the software is not that great. >>>=20 >>> P.S. I didn't use cURL here because my POST tcpconnects are >>> working fine, but I've used it before and it seems to work well. >>>=20 >>> Regards, >>> Kenneth Grome >>> WebDNA Solutions >>> http://www.webdnasolutions.com >>> Web Database Systems and Linux Server Administration >>>=20 >>>=20 >>>=20 >>>=20 >>>=20 >>>=20 >>>=20 >>>> On 03/09/2017 11:45 AM, Donovan Brooke wrote: >>>> Ahh=E2=80=A6 well, If memory serves.. I think that=E2=80=99s the way it= =E2=80=99s >>>> always functioned=E2=80=A6 and I believe there has been some feature >>>> requests to allow tcpconnect to follow redirects. >>>>=20 >>>> I think in PHP curl you have to specify to follow redirects. >>>>=20 >>>> So, a workaround could be to use curl I guess. >>>>=20 >>>> Donovan >>>>=20 >>>>=20 >>>>=20 >>>>=20 >>>> On Mar 9, 2017, at 11:36 AM, Kenneth Grome >>>> wrote: >>>>=20 >>>>> It doesn't follow the redirect like a POST tcpconnect does, >>>>> that's why I said it's not working properly. >>>>>=20 >>>>> Do tcpconnects only follow redirects when using POST but not >>>>> when using GET? IF so, this should be documented. >>>>>=20 >>>>> Or is there actually something wrong with the example I >>>>> posted that's preventing it from following the redirect? >>>>>=20 >>>>> You can easily compare the two examples below. I would >>>>> expect them to return the same results since they are >>>>> requesting the same page, but they do not: >>>>>=20 >>>>>=20 >>>>> [tcpconnect host=3Dwww.webdna.us&port=3D80] [tcpsend]GET / >>>>> HTTP/1.0[unurl]%0D%0A%0D%0A[/unurl][/tcpsend] [/tcpconnect] >>>>>=20 >>>>>=20 >>>>> [text]host=3Dwww.webdna.us[/text] [text]path=3D/[/text] >>>>> [text]n=3D[unurl]%0D%0A[/unurl][/text] [text]content=3D[/text] >>>>> [tcpconnect host=3D[host]&port=3D80] [tcpsend]POST [path] >>>>> HTTP/1.0[n][!] [/!]Host: [host][n][!] [/!]User-Agent: >>>>> Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)[n][!] >>>>> [/!]Content-Type: text/namevalue[n][!] [/!]Content-Length: >>>>> [countchars][content][/countchars][n][n][!] >>>>> [/!][content][n][!] [/!][/tcpsend] [/tcpconnect] >>>>>=20 >>>>>=20 >>>>> I had the same problem when using both types of tcpconnects >>>>> to request pages from two of my sites, each of which is on a >>>>> different server. The POST versions worked fine every time, >>>>> but the GET versions always failed. >>>>>=20 >>>>> And in my tests I specifically requested an existing file, >>>>> but instead of receiving it in the GET versions I always got >>>>> a 404 error while the POST versions received the specified >>>>> file. >>>>>=20 >>>>> Bottom line: I still think something's wrong, either with >>>>> the internal code that interprets GET tcpconnects or with the >>>>> WebDNA syntax itself. >>>>>=20 >>>>> Regards, Kenneth Grome WebDNA Solutions >>>>> http://www.webdnasolutions.com Web Database Systems and Linux >>>>> Server Administration >>>>>=20 >>>>>=20 >>>>>=20 >>>>>> On 03/09/2017 10:39 AM, Donovan Brooke wrote: >>>>>> Looks like it=E2=80=99s working to me.. a 302 is a (temp) >>>>>> redirect. >>>>>>=20 >>>>>> Donovan >>>>>>=20 >>>>>>=20 >>>>>>=20 >>>>>> On Mar 9, 2017, at 9:59 AM, Kenneth Grome >>>>>> wrote: >>>>>>=20 >>>>>>> This sample code (from the webdna.us website) doesn't >>>>>>> work: >>>>>>>=20 >>>>>>> [tcpconnect host=3Dwww.webdna.us&port=3D80] [tcpsend]GET / >>>>>>> HTTP/1.0[unurl]%0D%0A%0D%0A[/unurl][/tcpsend] >>>>>>> [/tcpconnect] >>>>>>>=20 >>>>>>> What's missing from this code ... or what's incorrect >>>>>>> about? Or don't tcpconnects work with method=3DGET any >>>>>>> more? >>>>>>>=20 >>>>>>> Something's wrong with it because this is what it >>>>>>> produces: >>>>>>>=20 >>>>>>> HTTP/1.1 302 Moved Temporarily Date: Thu, 09 Mar 2017 >>>>>>> 15:59:16 GMT Server: Apache/2.2.15 (CentOS) Location: >>>>>>> page.dna?numero=3D27 Content-Length: 1 Vary: >>>>>>> Accept-Encoding,User-Agent Connection: close >>>>>>> Content-Type: text/html >>>>>>>=20 >>>>>>> Regards, Kenneth Grome WebDNA Solutions >>>>>>> http://www.webdnasolutions.com Web Database Systems and >>>>>>> Linux Server Administration >>>>>>>=20 >>>>>>>=20 >>>>>>> --------------------------------------------------------- >>>>>>>=20 >>>>>>>=20 >>> This message is sent to you because you are subscribed to >>>>>>> the mailing list . To unsubscribe, E-mail >>>>>>> to: archives: >>>>>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>>>>> support@webdna.us >>>>>>=20 >>>>>> --------------------------------------------------------- >>>>>> This message is sent to you because you are subscribed to >>>>>> the mailing list . To unsubscribe, E-mail >>>>>> to: archives: >>>>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>>>> support@webdna.us >>>>>>=20 >>>>> --------------------------------------------------------- >>>>> This message is sent to you because you are subscribed to the >>>>> mailing list . To unsubscribe, E-mail to: >>>>> archives: >>>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>>> support@webdna.us >>>>=20 >>>> --------------------------------------------------------- This >>>> message is sent to you because you are subscribed to the >>>> mailing list . To unsubscribe, E-mail to: >>>> archives: >>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>> support@webdna.us >>>>=20 >>>=20 >>> Sincerely, >>> Kenneth Grome >>>=20 >>> --------------------------------------------------------- >>> This message is sent to you because you are subscribed to >>> the mailing list . >>> To unsubscribe, E-mail to: >>> archives: http://mail.webdna.us/list/talk@webdna.us >>> Bug Reporting: support@webdna.us >>=20 > --------------------------------------------------------- > This message is sent to you because you are subscribed to > the mailing list . > To unsubscribe, E-mail to: > archives: http://mail.webdna.us/list/talk@webdna.us > Bug Reporting: support@webdna.us --------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: archives: http://mail.webdna.us/list/talk@webdna.us Bug Reporting: support@webdna.us . Associated Messages, from the most recent to the oldest:

    
  1. Re: [WebDNA] TCPConnect example doesn't work (christophe.billiottet@webdna.us 2017)
  2. Re: [WebDNA] TCPConnect example doesn't work (Kenneth Grome 2017)
  3. Re: [WebDNA] TCPConnect example doesn't work (Michael Davis 2017)
  4. Re: [WebDNA] TCPConnect example doesn't work (Kenneth Grome 2017)
  5. Re: [WebDNA] TCPConnect example doesn't work (Michael Davis 2017)
  6. Re: [WebDNA] TCPConnect example doesn't work (Kenneth Grome 2017)
  7. Re: [WebDNA] TCPConnect example doesn't work (Donovan Brooke 2017)
  8. Re: [WebDNA] TCPConnect example doesn't work (Kenneth Grome 2017)
  9. Re: [WebDNA] TCPConnect example doesn't work (Donovan Brooke 2017)
  10. [WebDNA] TCPConnect example doesn't work (Kenneth Grome 2017)
1061 Ken, Most likely its the Host header that you need to populate. Without it= , the remote server won't be able to route the request to a host based virtu= al host. If you are communicating with a highly secure service such as bank= or payment gateway, you will also likely need to include the User-Agent and= maybe a few others. =20 Mike > On Mar 10, 2017, at 5:27 AM, Kenneth Grome wrote= : >=20 > Mike, >=20 > My problem with GET tcpconnect is not that it doesn't follow > redirects, it's that it returns a 404 error when the specified > page exists. >=20 > POST tcpconnect does not have this problem, at least in the way I > coded mine. So now I believe that GET tcpconnect might work if I > ignore the bad example in the docs, and instead use the correct > but undocumented syntax. >=20 > Your example contains several parameters inside the tcpsend > context, whereas the GET example on webdna.us does not include > any. So maybe some of your tcpsend parameters are required in > order for GET tcpconnect to work properly? >=20 > Regards, > Kenneth Grome > WebDNA Solutions > http://www.webdnasolutions.com > Web Database Systems and Linux Server Administration >=20 >=20 >=20 >> On 03/09/2017 11:55 PM, Michael Davis wrote: >> Ken, >>=20 >> I've run into this problem in the past. You can use cURL to >> pre-flight your endpoint. I have a basic function that works for >> the situations I need it for and you're welcome to grab it and use it >> http://www.network13.net/webdna_code_curl_redirects.dna >>=20 >> Maybe WebDNA TCPConnect will be improved to follow redirects in >> the future, but for now, we DIY, which I don't really have a >> problem with, so long as it works. >>=20 >>=20 >> Mike >>=20 >>=20 >>> On Mar 9, 2017, at 11:10 AM, Kenneth Grome >>> > wrote: >>>=20 >>> I haven't used GET tcpconnect in a very long time so maybe you're >>> right. I don't really remember having any problems with it >>> before, but something is definitely wrong when the file clearly >>> exists in the exact location specified yet it won't GET it. >>>=20 >>> I do know that POST tcpconnects often require the host to be >>> specified in two places: in the tcpconnect tag and also inside the >>> tcpsend context. >>>=20 >>> =46rom what I recall when I learned about this 'two hosts' issue a >>> long time ago, this was a 'sometimes' requirement that depended >>> upon the server receiving the request. It's been a while since I >>> ran into it, so maybe putting the host in both places is a >>> requirement in all POST tcpconnects now. >>>=20 >>> Unfortunately there is almost nothing in the online docs that >>> mentions any of this. I would hate to be a new user who doesn't >>> know all these undocumented requirements because after trying and >>> failing to get the published examples to work properly I might >>> conclude that the software is not that great. >>>=20 >>> P.S. I didn't use cURL here because my POST tcpconnects are >>> working fine, but I've used it before and it seems to work well. >>>=20 >>> Regards, >>> Kenneth Grome >>> WebDNA Solutions >>> http://www.webdnasolutions.com >>> Web Database Systems and Linux Server Administration >>>=20 >>>=20 >>>=20 >>>=20 >>>=20 >>>=20 >>>=20 >>>> On 03/09/2017 11:45 AM, Donovan Brooke wrote: >>>> Ahh=E2=80=A6 well, If memory serves.. I think that=E2=80=99s the way it= =E2=80=99s >>>> always functioned=E2=80=A6 and I believe there has been some feature >>>> requests to allow tcpconnect to follow redirects. >>>>=20 >>>> I think in PHP curl you have to specify to follow redirects. >>>>=20 >>>> So, a workaround could be to use curl I guess. >>>>=20 >>>> Donovan >>>>=20 >>>>=20 >>>>=20 >>>>=20 >>>> On Mar 9, 2017, at 11:36 AM, Kenneth Grome >>>> wrote: >>>>=20 >>>>> It doesn't follow the redirect like a POST tcpconnect does, >>>>> that's why I said it's not working properly. >>>>>=20 >>>>> Do tcpconnects only follow redirects when using POST but not >>>>> when using GET? IF so, this should be documented. >>>>>=20 >>>>> Or is there actually something wrong with the example I >>>>> posted that's preventing it from following the redirect? >>>>>=20 >>>>> You can easily compare the two examples below. I would >>>>> expect them to return the same results since they are >>>>> requesting the same page, but they do not: >>>>>=20 >>>>>=20 >>>>> [tcpconnect host=3Dwww.webdna.us&port=3D80] [tcpsend]GET / >>>>> HTTP/1.0[unurl]%0D%0A%0D%0A[/unurl][/tcpsend] [/tcpconnect] >>>>>=20 >>>>>=20 >>>>> [text]host=3Dwww.webdna.us[/text] [text]path=3D/[/text] >>>>> [text]n=3D[unurl]%0D%0A[/unurl][/text] [text]content=3D[/text] >>>>> [tcpconnect host=3D[host]&port=3D80] [tcpsend]POST [path] >>>>> HTTP/1.0[n][!] [/!]Host: [host][n][!] [/!]User-Agent: >>>>> Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)[n][!] >>>>> [/!]Content-Type: text/namevalue[n][!] [/!]Content-Length: >>>>> [countchars][content][/countchars][n][n][!] >>>>> [/!][content][n][!] [/!][/tcpsend] [/tcpconnect] >>>>>=20 >>>>>=20 >>>>> I had the same problem when using both types of tcpconnects >>>>> to request pages from two of my sites, each of which is on a >>>>> different server. The POST versions worked fine every time, >>>>> but the GET versions always failed. >>>>>=20 >>>>> And in my tests I specifically requested an existing file, >>>>> but instead of receiving it in the GET versions I always got >>>>> a 404 error while the POST versions received the specified >>>>> file. >>>>>=20 >>>>> Bottom line: I still think something's wrong, either with >>>>> the internal code that interprets GET tcpconnects or with the >>>>> WebDNA syntax itself. >>>>>=20 >>>>> Regards, Kenneth Grome WebDNA Solutions >>>>> http://www.webdnasolutions.com Web Database Systems and Linux >>>>> Server Administration >>>>>=20 >>>>>=20 >>>>>=20 >>>>>> On 03/09/2017 10:39 AM, Donovan Brooke wrote: >>>>>> Looks like it=E2=80=99s working to me.. a 302 is a (temp) >>>>>> redirect. >>>>>>=20 >>>>>> Donovan >>>>>>=20 >>>>>>=20 >>>>>>=20 >>>>>> On Mar 9, 2017, at 9:59 AM, Kenneth Grome >>>>>> wrote: >>>>>>=20 >>>>>>> This sample code (from the webdna.us website) doesn't >>>>>>> work: >>>>>>>=20 >>>>>>> [tcpconnect host=3Dwww.webdna.us&port=3D80] [tcpsend]GET / >>>>>>> HTTP/1.0[unurl]%0D%0A%0D%0A[/unurl][/tcpsend] >>>>>>> [/tcpconnect] >>>>>>>=20 >>>>>>> What's missing from this code ... or what's incorrect >>>>>>> about? Or don't tcpconnects work with method=3DGET any >>>>>>> more? >>>>>>>=20 >>>>>>> Something's wrong with it because this is what it >>>>>>> produces: >>>>>>>=20 >>>>>>> HTTP/1.1 302 Moved Temporarily Date: Thu, 09 Mar 2017 >>>>>>> 15:59:16 GMT Server: Apache/2.2.15 (CentOS) Location: >>>>>>> page.dna?numero=3D27 Content-Length: 1 Vary: >>>>>>> Accept-Encoding,User-Agent Connection: close >>>>>>> Content-Type: text/html >>>>>>>=20 >>>>>>> Regards, Kenneth Grome WebDNA Solutions >>>>>>> http://www.webdnasolutions.com Web Database Systems and >>>>>>> Linux Server Administration >>>>>>>=20 >>>>>>>=20 >>>>>>> --------------------------------------------------------- >>>>>>>=20 >>>>>>>=20 >>> This message is sent to you because you are subscribed to >>>>>>> the mailing list . To unsubscribe, E-mail >>>>>>> to: archives: >>>>>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>>>>> support@webdna.us >>>>>>=20 >>>>>> --------------------------------------------------------- >>>>>> This message is sent to you because you are subscribed to >>>>>> the mailing list . To unsubscribe, E-mail >>>>>> to: archives: >>>>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>>>> support@webdna.us >>>>>>=20 >>>>> --------------------------------------------------------- >>>>> This message is sent to you because you are subscribed to the >>>>> mailing list . To unsubscribe, E-mail to: >>>>> archives: >>>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>>> support@webdna.us >>>>=20 >>>> --------------------------------------------------------- This >>>> message is sent to you because you are subscribed to the >>>> mailing list . To unsubscribe, E-mail to: >>>> archives: >>>> http://mail.webdna.us/list/talk@webdna.us Bug Reporting: >>>> support@webdna.us >>>>=20 >>>=20 >>> Sincerely, >>> Kenneth Grome >>>=20 >>> --------------------------------------------------------- >>> This message is sent to you because you are subscribed to >>> the mailing list . >>> To unsubscribe, E-mail to: >>> archives: http://mail.webdna.us/list/talk@webdna.us >>> Bug Reporting: support@webdna.us >>=20 > --------------------------------------------------------- > This message is sent to you because you are subscribed to > the mailing list . > To unsubscribe, E-mail to: > archives: http://mail.webdna.us/list/talk@webdna.us > Bug Reporting: support@webdna.us --------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: archives: http://mail.webdna.us/list/talk@webdna.us Bug Reporting: support@webdna.us . Michael Davis

DOWNLOAD WEBDNA NOW!

Top Articles:

Talk List

The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...

Related Readings:

company names not begin with a letter (2005) WebCatalog and Macromedia Flash (1998) [Sum] function? (1997) [WebDNA] showif/sendmail question (2008) frames & carts (1997) [searchString] (1997) still facing truncated [texta] problem (2000) Help formatting search results w/ table (1997) [WebDNA] Anyone using CODA ?? (2012) Database Options (1997) Rollovers (1999) SpamCop Blues (was Sending emails before it (went[ot] )) (2002) Country & Ship-to address & other fields ? (1997) WebDNA Solutions ... sorry! (1997) UPDATE PROBLEM (1997) searching multiple databases in single search (1997) Running _every_ page through WebCat ? (1997) Search/sort in URL Was: GuestBook example (1997) [TCPConnect] questions/problems (2001) Security Question (1997)