You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@manifoldcf.apache.org by Jeff Potts <je...@gmail.com> on 2016/11/03 22:58:55 UTC

SharePoint Connector: "Root site is unreachable"

I am using apache-manifoldcf-2.5. I am trying to get the SharePoint
Connector to work against SharePoint 2003.

My SharePoint site URLs are like:
http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx

I have set up a Connector with:

Server protocol: http
Server name: somedomain.com
Server port: 80
Site path: /SomeSite

The connection status says, "Connection working".

However, when I go to set up a job, and I click the "Paths" tab, I see the
message, "Root site is unreachable, or user has no permissions".

I know I have permissions to that site because I can read/write to its
document library through the web interface. So I feel like I may be missing
something regarding the correct value for site path.

The docs say, "...type in the portion of the root site URL that includes
everything after the server and port, except for the final "aspx" file,"
which is why I am using "/SomeSite" as the site path.

But another snippet from the docs says this:

"The SharePoint connection type is designed so that one SharePoint
repository connection can access all SharePoint sites from a specific root
site though its explicit subsites. It is the case that it is desirable in
some very large SharePoint installations to access *all* SharePoint sites
using a single connection. But the ManifoldCF SharePoint connection type
does not support that model as of yet."

That confused me because it isn't clear if I should be specifying some sort
of root site (which I have no idea how to determine the path for) or if
creating one connection for each site should work.

Jeff

RE: SharePoint Connector: "Root site is unreachable"

Posted by as...@wipro.com.
Hi Jeff,

I agree with Karl. My suggestions were just to help you make the connector work temporarily (and I should have specified that in my email - apologies).

In our case, getting the plugin installed is a long drawn process and our priority was to get mcf up and running in a short span of time with the ability to crawl sharepoint documents (just basic crawl).

Regards
-Ashutosh




From: Karl Wright<ma...@gmail.com>
Sent: 04 November 2016 15:19
To: Jeff Potts<ma...@gmail.com>
Cc: user@manifoldcf.apache.org<ma...@manifoldcf.apache.org>
Subject: Re: SharePoint Connector: "Root site is unreachable"


** This mail has been sent from an external source **

Hi Jeff,

Despite what Ashutosh says it is highly recommended that you use the plugin unless you truly are on SharePoint 2003.  Without it a number of services won't work properly on later versions of SharePoint.

If everything is there, I suggest enabling httpcomponents/httpclient logging in the logging.ini file and seeing what actually is going back and forth.  That should give a strong hint as to what is going wrong.

Thanks,
Karl


On Fri, Nov 4, 2016 at 11:06 AM, Jeff Potts <je...@gmail.com>> wrote:
It looks like the MCPermissions.asmx is indeed missing so I assume the MCF plugin has not been installed.

I'm connecting as a user with admin rights and I am setting the connection as SharePoint 2003.

Jeff

On Thu, Nov 3, 2016 at 9:36 PM, <as...@wipro.com>> wrote:
Hi Jeff,

Based on my experience with the Sharepoint connector in MCF 2.4, this error may indicate that the user id you are using to connect to Sharepoint, only has read privilege on the site.

I think you can crawl sharepoint sites without installing the mcf plugin, provided the sharepoint user id you are using has contributor permissions at a minimum.

Also, in the above case, you need to set up the repo connection as sharepoint 2003/2004 not higher.

At least that's the configuration that worked for us without installing the MCF plugin.

Hope this helps.
-Ashutosh


From: Karl Wright<ma...@gmail.com>
Sent: 04 November 2016 00:28
To: user@manifoldcf.apache.org<ma...@manifoldcf.apache.org>
Subject: Re: SharePoint Connector: "Root site is unreachable"


** This mail has been sent from an external source **

Hi Jeff,

The meaning of that error message is that the data coming back from SharePoint when MCF inquires as to the subsites, child libraries, etc. is not valid XML.  It is not clear precisely what is wrong.

I think the best way forward is to turn on wire httpcomponents/httpclient wire debugging in logging.ini, and see what is going back and forth.  It should be obvious pretty quickly what is wrong.  FWIW, in a properly deployed SharePoint system, you should see a _vti_bin directory with a bunch of .asmx files in there for each SharePoint site or subsite.  That is, there should be: http://somedomain.com/SomeSite/<http://somedomain.com/SomeSite/_layouts/viewlsts.aspx>_vti_bin/lists.asmx, etc.  The asmx files I would expect are:


Permissions.asmx
MCPermissions.asmx
usergroup.asmx
dspsts.asmx
lists.asmx
versions.asmx
webs.asmx

It is possible that you are missing some of these because of how you installed sharepoint (and because of what you enabled).  Or, you didn't install the MCF SharePoint plugin that is appropriate for your SharePoint version (which supplies MCPermissions.asmx).  Any missing files will *also* cause the kind of problem you are seeing.

Thanks,
Karl



On Thu, Nov 3, 2016 at 8:01 PM, Jeff Potts <je...@gmail.com>> wrote:
In my example those are all virtual sites, but for some reason they are not accessible via /site. Is that an indication that something is misconfigured on the Sharepoint side?

Why would the connector report a working connection but the job be unable to navigate the path structure?

Jeff

On Nov 3, 2016, at 6:17 PM, Karl Wright <da...@gmail.com>> wrote:

Hi Jeff,

SharePoint's site model is as follows:

-- there is a root site, usually with the URL http://somedomain.com
-- there are "virtual" sites, which are also full-fledge sites, usually with the URL http://somedomain.com/site/site_name
-- there are subsites of the root site, usually with URLs like http://somedomain.com/subsiteX
-- there are subsites of virtual sites, usually with URLs like http://somedomain.com/site/site_name/subsiteY

Your MCF sharepoint connections should not point to subsites; they should only point to the root site or a single virtual site.  There is no way to deal with crawling multiple virtual sites with the same connection, which is what the documentation is trying to say.

I hope this helps.

Karl


On Thu, Nov 3, 2016 at 6:58 PM, Jeff Potts <je...@gmail.com>> wrote:
I am using apache-manifoldcf-2.5. I am trying to get the SharePoint Connector to work against SharePoint 2003.

My SharePoint site URLs are like:
http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx

I have set up a Connector with:

Server protocol: http
Server name: somedomain.com<http://somedomain.com>
Server port: 80
Site path: /SomeSite

The connection status says, "Connection working".

However, when I go to set up a job, and I click the "Paths" tab, I see the message, "Root site is unreachable, or user has no permissions".

I know I have permissions to that site because I can read/write to its document library through the web interface. So I feel like I may be missing something regarding the correct value for site path.

The docs say, "...type in the portion of the root site URL that includes everything after the server and port, except for the final "aspx" file," which is why I am using "/SomeSite" as the site path.

But another snippet from the docs says this:

"The SharePoint connection type is designed so that one SharePoint repository connection can access all SharePoint sites from a specific root site though its explicit subsites. It is the case that it is desirable in some very large SharePoint installations to access all SharePoint sites using a single connection. But the ManifoldCF SharePoint connection type does not support that model as of yet."

That confused me because it isn't clear if I should be specifying some sort of root site (which I have no idea how to determine the path for) or if creating one connection for each site should work.

Jeff




The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com<http://www.wipro.com>


The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com

Re: SharePoint Connector: "Root site is unreachable"

Posted by Karl Wright <da...@gmail.com>.
Hi Jeff,

Despite what Ashutosh says it is highly recommended that you use the plugin
unless you truly are on SharePoint 2003.  Without it a number of services
won't work properly on later versions of SharePoint.

If everything is there, I suggest enabling httpcomponents/httpclient
logging in the logging.ini file and seeing what actually is going back and
forth.  That should give a strong hint as to what is going wrong.

Thanks,
Karl


On Fri, Nov 4, 2016 at 11:06 AM, Jeff Potts <je...@gmail.com> wrote:

> It looks like the MCPermissions.asmx is indeed missing so I assume the MCF
> plugin has not been installed.
>
> I'm connecting as a user with admin rights and I am setting the connection
> as SharePoint 2003.
>
> Jeff
>
> On Thu, Nov 3, 2016 at 9:36 PM, <as...@wipro.com> wrote:
>
>> Hi Jeff,
>>
>>
>>
>> Based on my experience with the Sharepoint connector in MCF 2.4, this
>> error may indicate that the user id you are using to connect to Sharepoint,
>> only has read privilege on the site.
>>
>>
>>
>> I think you can crawl sharepoint sites without installing the mcf plugin,
>> provided the sharepoint user id you are using has contributor permissions
>> at a minimum.
>>
>>
>>
>> Also, in the above case, you need to set up the repo connection as
>> sharepoint 2003/2004 not higher.
>>
>>
>>
>> At least that's the configuration that worked for us without installing
>> the MCF plugin.
>>
>>
>>
>> Hope this helps.
>>
>> -Ashutosh
>>
>>
>>
>>
>>
>> *From: *Karl Wright <da...@gmail.com>
>> *Sent: *04 November 2016 00:28
>> *To: *user@manifoldcf.apache.org
>> *Subject: *Re: SharePoint Connector: "Root site is unreachable"
>>
>>
>>
>> ** This mail has been sent from an external source **
>> Hi Jeff,
>>
>> The meaning of that error message is that the data coming back from
>> SharePoint when MCF inquires as to the subsites, child libraries, etc. is
>> not valid XML.  It is not clear precisely what is wrong.
>>
>> I think the best way forward is to turn on wire httpcomponents/httpclient
>> wire debugging in logging.ini, and see what is going back and forth.  It
>> should be obvious pretty quickly what is wrong.  FWIW, in a properly
>> deployed SharePoint system, you should see a _vti_bin directory with a
>> bunch of .asmx files in there for each SharePoint site or subsite.  That
>> is, there should be: http://somedomain.com/SomeSite/
>> <http://somedomain.com/SomeSite/_layouts/viewlsts.aspx>_vti_bin/lists.asmx,
>> etc.  The asmx files I would expect are:
>>
>>
>> Permissions.asmx
>> MCPermissions.asmx
>> usergroup.asmx
>> dspsts.asmx
>> lists.asmx
>> versions.asmx
>> webs.asmx
>>
>> It is possible that you are missing some of these because of how you
>> installed sharepoint (and because of what you enabled).  Or, you didn't
>> install the MCF SharePoint plugin that is appropriate for your SharePoint
>> version (which supplies MCPermissions.asmx).  Any missing files will *also*
>> cause the kind of problem you are seeing.
>>
>> Thanks,
>> Karl
>>
>>
>>
>> On Thu, Nov 3, 2016 at 8:01 PM, Jeff Potts <je...@gmail.com> wrote:
>>
>>> In my example those are all virtual sites, but for some reason they are
>>> not accessible via /site. Is that an indication that something is
>>> misconfigured on the Sharepoint side?
>>>
>>> Why would the connector report a working connection but the job be
>>> unable to navigate the path structure?
>>>
>>> Jeff
>>>
>>> On Nov 3, 2016, at 6:17 PM, Karl Wright <da...@gmail.com> wrote:
>>>
>>> Hi Jeff,
>>>
>>> SharePoint's site model is as follows:
>>>
>>> -- there is a root site, usually with the URL http://somedomain.com
>>> -- there are "virtual" sites, which are also full-fledge sites, usually
>>> with the URL http://somedomain.com/site/site_name
>>> -- there are subsites of the root site, usually with URLs like
>>> http://somedomain.com/subsiteX
>>> -- there are subsites of virtual sites, usually with URLs like
>>> http://somedomain.com/site/site_name/subsiteY
>>>
>>> Your MCF sharepoint connections should not point to subsites; they
>>> should only point to the root site or a single virtual site.  There is no
>>> way to deal with crawling multiple virtual sites with the same connection,
>>> which is what the documentation is trying to say.
>>>
>>> I hope this helps.
>>>
>>> Karl
>>>
>>>
>>> On Thu, Nov 3, 2016 at 6:58 PM, Jeff Potts <je...@gmail.com>
>>> wrote:
>>>
>>>> I am using apache-manifoldcf-2.5. I am trying to get the SharePoint
>>>> Connector to work against SharePoint 2003.
>>>>
>>>> My SharePoint site URLs are like:
>>>> http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
>>>> http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
>>>> http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx
>>>>
>>>> I have set up a Connector with:
>>>>
>>>> Server protocol: http
>>>> Server name: somedomain.com
>>>> Server port: 80
>>>> Site path: /SomeSite
>>>>
>>>> The connection status says, "Connection working".
>>>>
>>>> However, when I go to set up a job, and I click the "Paths" tab, I see
>>>> the message, "Root site is unreachable, or user has no permissions".
>>>>
>>>> I know I have permissions to that site because I can read/write to its
>>>> document library through the web interface. So I feel like I may be missing
>>>> something regarding the correct value for site path.
>>>>
>>>> The docs say, "...type in the portion of the root site URL that
>>>> includes everything after the server and port, except for the final "aspx"
>>>> file," which is why I am using "/SomeSite" as the site path.
>>>>
>>>> But another snippet from the docs says this:
>>>>
>>>> "The SharePoint connection type is designed so that one SharePoint
>>>> repository connection can access all SharePoint sites from a specific root
>>>> site though its explicit subsites. It is the case that it is desirable in
>>>> some very large SharePoint installations to access *all* SharePoint
>>>> sites using a single connection. But the ManifoldCF SharePoint connection
>>>> type does not support that model as of yet."
>>>>
>>>> That confused me because it isn't clear if I should be specifying some
>>>> sort of root site (which I have no idea how to determine the path for) or
>>>> if creating one connection for each site should work.
>>>>
>>>> Jeff
>>>>
>>>>
>>>>
>>>
>> The information contained in this electronic message and any attachments
>> to this message are intended for the exclusive use of the addressee(s) and
>> may contain proprietary, confidential or privileged information. If you are
>> not the intended recipient, you should not disseminate, distribute or copy
>> this e-mail. Please notify the sender immediately and destroy all copies of
>> this message and any attachments. WARNING: Computer viruses can be
>> transmitted via email. The recipient should check this email and any
>> attachments for the presence of viruses. The company accepts no liability
>> for any damage caused by any virus transmitted by this email.
>> www.wipro.com
>>
>
>

Re: SharePoint Connector: "Root site is unreachable"

Posted by Jeff Potts <je...@gmail.com>.
It looks like the MCPermissions.asmx is indeed missing so I assume the MCF
plugin has not been installed.

I'm connecting as a user with admin rights and I am setting the connection
as SharePoint 2003.

Jeff

On Thu, Nov 3, 2016 at 9:36 PM, <as...@wipro.com> wrote:

> Hi Jeff,
>
>
>
> Based on my experience with the Sharepoint connector in MCF 2.4, this
> error may indicate that the user id you are using to connect to Sharepoint,
> only has read privilege on the site.
>
>
>
> I think you can crawl sharepoint sites without installing the mcf plugin,
> provided the sharepoint user id you are using has contributor permissions
> at a minimum.
>
>
>
> Also, in the above case, you need to set up the repo connection as
> sharepoint 2003/2004 not higher.
>
>
>
> At least that's the configuration that worked for us without installing
> the MCF plugin.
>
>
>
> Hope this helps.
>
> -Ashutosh
>
>
>
>
>
> *From: *Karl Wright <da...@gmail.com>
> *Sent: *04 November 2016 00:28
> *To: *user@manifoldcf.apache.org
> *Subject: *Re: SharePoint Connector: "Root site is unreachable"
>
>
>
> ** This mail has been sent from an external source **
> Hi Jeff,
>
> The meaning of that error message is that the data coming back from
> SharePoint when MCF inquires as to the subsites, child libraries, etc. is
> not valid XML.  It is not clear precisely what is wrong.
>
> I think the best way forward is to turn on wire httpcomponents/httpclient
> wire debugging in logging.ini, and see what is going back and forth.  It
> should be obvious pretty quickly what is wrong.  FWIW, in a properly
> deployed SharePoint system, you should see a _vti_bin directory with a
> bunch of .asmx files in there for each SharePoint site or subsite.  That
> is, there should be: http://somedomain.com/SomeSite/
> <http://somedomain.com/SomeSite/_layouts/viewlsts.aspx>_vti_bin/lists.asmx,
> etc.  The asmx files I would expect are:
>
>
> Permissions.asmx
> MCPermissions.asmx
> usergroup.asmx
> dspsts.asmx
> lists.asmx
> versions.asmx
> webs.asmx
>
> It is possible that you are missing some of these because of how you
> installed sharepoint (and because of what you enabled).  Or, you didn't
> install the MCF SharePoint plugin that is appropriate for your SharePoint
> version (which supplies MCPermissions.asmx).  Any missing files will *also*
> cause the kind of problem you are seeing.
>
> Thanks,
> Karl
>
>
>
> On Thu, Nov 3, 2016 at 8:01 PM, Jeff Potts <je...@gmail.com> wrote:
>
>> In my example those are all virtual sites, but for some reason they are
>> not accessible via /site. Is that an indication that something is
>> misconfigured on the Sharepoint side?
>>
>> Why would the connector report a working connection but the job be unable
>> to navigate the path structure?
>>
>> Jeff
>>
>> On Nov 3, 2016, at 6:17 PM, Karl Wright <da...@gmail.com> wrote:
>>
>> Hi Jeff,
>>
>> SharePoint's site model is as follows:
>>
>> -- there is a root site, usually with the URL http://somedomain.com
>> -- there are "virtual" sites, which are also full-fledge sites, usually
>> with the URL http://somedomain.com/site/site_name
>> -- there are subsites of the root site, usually with URLs like
>> http://somedomain.com/subsiteX
>> -- there are subsites of virtual sites, usually with URLs like
>> http://somedomain.com/site/site_name/subsiteY
>>
>> Your MCF sharepoint connections should not point to subsites; they should
>> only point to the root site or a single virtual site.  There is no way to
>> deal with crawling multiple virtual sites with the same connection, which
>> is what the documentation is trying to say.
>>
>> I hope this helps.
>>
>> Karl
>>
>>
>> On Thu, Nov 3, 2016 at 6:58 PM, Jeff Potts <je...@gmail.com> wrote:
>>
>>> I am using apache-manifoldcf-2.5. I am trying to get the SharePoint
>>> Connector to work against SharePoint 2003.
>>>
>>> My SharePoint site URLs are like:
>>> http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
>>> http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
>>> http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx
>>>
>>> I have set up a Connector with:
>>>
>>> Server protocol: http
>>> Server name: somedomain.com
>>> Server port: 80
>>> Site path: /SomeSite
>>>
>>> The connection status says, "Connection working".
>>>
>>> However, when I go to set up a job, and I click the "Paths" tab, I see
>>> the message, "Root site is unreachable, or user has no permissions".
>>>
>>> I know I have permissions to that site because I can read/write to its
>>> document library through the web interface. So I feel like I may be missing
>>> something regarding the correct value for site path.
>>>
>>> The docs say, "...type in the portion of the root site URL that includes
>>> everything after the server and port, except for the final "aspx" file,"
>>> which is why I am using "/SomeSite" as the site path.
>>>
>>> But another snippet from the docs says this:
>>>
>>> "The SharePoint connection type is designed so that one SharePoint
>>> repository connection can access all SharePoint sites from a specific root
>>> site though its explicit subsites. It is the case that it is desirable in
>>> some very large SharePoint installations to access *all* SharePoint
>>> sites using a single connection. But the ManifoldCF SharePoint connection
>>> type does not support that model as of yet."
>>>
>>> That confused me because it isn't clear if I should be specifying some
>>> sort of root site (which I have no idea how to determine the path for) or
>>> if creating one connection for each site should work.
>>>
>>> Jeff
>>>
>>>
>>>
>>
> The information contained in this electronic message and any attachments
> to this message are intended for the exclusive use of the addressee(s) and
> may contain proprietary, confidential or privileged information. If you are
> not the intended recipient, you should not disseminate, distribute or copy
> this e-mail. Please notify the sender immediately and destroy all copies of
> this message and any attachments. WARNING: Computer viruses can be
> transmitted via email. The recipient should check this email and any
> attachments for the presence of viruses. The company accepts no liability
> for any damage caused by any virus transmitted by this email.
> www.wipro.com
>

RE: SharePoint Connector: "Root site is unreachable"

Posted by as...@wipro.com.
Hi Jeff,

Based on my experience with the Sharepoint connector in MCF 2.4, this error may indicate that the user id you are using to connect to Sharepoint, only has read privilege on the site.

I think you can crawl sharepoint sites without installing the mcf plugin, provided the sharepoint user id you are using has contributor permissions at a minimum.

Also, in the above case, you need to set up the repo connection as sharepoint 2003/2004 not higher.

At least that's the configuration that worked for us without installing the MCF plugin.

Hope this helps.
-Ashutosh


From: Karl Wright<ma...@gmail.com>
Sent: 04 November 2016 00:28
To: user@manifoldcf.apache.org<ma...@manifoldcf.apache.org>
Subject: Re: SharePoint Connector: "Root site is unreachable"


** This mail has been sent from an external source **

Hi Jeff,

The meaning of that error message is that the data coming back from SharePoint when MCF inquires as to the subsites, child libraries, etc. is not valid XML.  It is not clear precisely what is wrong.

I think the best way forward is to turn on wire httpcomponents/httpclient wire debugging in logging.ini, and see what is going back and forth.  It should be obvious pretty quickly what is wrong.  FWIW, in a properly deployed SharePoint system, you should see a _vti_bin directory with a bunch of .asmx files in there for each SharePoint site or subsite.  That is, there should be: http://somedomain.com/SomeSite/<http://somedomain.com/SomeSite/_layouts/viewlsts.aspx>_vti_bin/lists.asmx, etc.  The asmx files I would expect are:


Permissions.asmx
MCPermissions.asmx
usergroup.asmx
dspsts.asmx
lists.asmx
versions.asmx
webs.asmx

It is possible that you are missing some of these because of how you installed sharepoint (and because of what you enabled).  Or, you didn't install the MCF SharePoint plugin that is appropriate for your SharePoint version (which supplies MCPermissions.asmx).  Any missing files will *also* cause the kind of problem you are seeing.

Thanks,
Karl



On Thu, Nov 3, 2016 at 8:01 PM, Jeff Potts <je...@gmail.com>> wrote:
In my example those are all virtual sites, but for some reason they are not accessible via /site. Is that an indication that something is misconfigured on the Sharepoint side?

Why would the connector report a working connection but the job be unable to navigate the path structure?

Jeff

On Nov 3, 2016, at 6:17 PM, Karl Wright <da...@gmail.com>> wrote:

Hi Jeff,

SharePoint's site model is as follows:

-- there is a root site, usually with the URL http://somedomain.com
-- there are "virtual" sites, which are also full-fledge sites, usually with the URL http://somedomain.com/site/site_name
-- there are subsites of the root site, usually with URLs like http://somedomain.com/subsiteX
-- there are subsites of virtual sites, usually with URLs like http://somedomain.com/site/site_name/subsiteY

Your MCF sharepoint connections should not point to subsites; they should only point to the root site or a single virtual site.  There is no way to deal with crawling multiple virtual sites with the same connection, which is what the documentation is trying to say.

I hope this helps.

Karl


On Thu, Nov 3, 2016 at 6:58 PM, Jeff Potts <je...@gmail.com>> wrote:
I am using apache-manifoldcf-2.5. I am trying to get the SharePoint Connector to work against SharePoint 2003.

My SharePoint site URLs are like:
http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx

I have set up a Connector with:

Server protocol: http
Server name: somedomain.com<http://somedomain.com>
Server port: 80
Site path: /SomeSite

The connection status says, "Connection working".

However, when I go to set up a job, and I click the "Paths" tab, I see the message, "Root site is unreachable, or user has no permissions".

I know I have permissions to that site because I can read/write to its document library through the web interface. So I feel like I may be missing something regarding the correct value for site path.

The docs say, "...type in the portion of the root site URL that includes everything after the server and port, except for the final "aspx" file," which is why I am using "/SomeSite" as the site path.

But another snippet from the docs says this:

"The SharePoint connection type is designed so that one SharePoint repository connection can access all SharePoint sites from a specific root site though its explicit subsites. It is the case that it is desirable in some very large SharePoint installations to access all SharePoint sites using a single connection. But the ManifoldCF SharePoint connection type does not support that model as of yet."

That confused me because it isn't clear if I should be specifying some sort of root site (which I have no idea how to determine the path for) or if creating one connection for each site should work.

Jeff




The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com

Re: SharePoint Connector: "Root site is unreachable"

Posted by Karl Wright <da...@gmail.com>.
Hi Jeff,

The meaning of that error message is that the data coming back from
SharePoint when MCF inquires as to the subsites, child libraries, etc. is
not valid XML.  It is not clear precisely what is wrong.

I think the best way forward is to turn on wire httpcomponents/httpclient
wire debugging in logging.ini, and see what is going back and forth.  It
should be obvious pretty quickly what is wrong.  FWIW, in a properly
deployed SharePoint system, you should see a _vti_bin directory with a
bunch of .asmx files in there for each SharePoint site or subsite.  That
is, there should be: http://somedomain.com/SomeSite/
<http://somedomain.com/SomeSite/_layouts/viewlsts.aspx>_vti_bin/lists.asmx,
etc.  The asmx files I would expect are:


Permissions.asmx
MCPermissions.asmx
usergroup.asmx
dspsts.asmx
lists.asmx
versions.asmx
webs.asmx

It is possible that you are missing some of these because of how you
installed sharepoint (and because of what you enabled).  Or, you didn't
install the MCF SharePoint plugin that is appropriate for your SharePoint
version (which supplies MCPermissions.asmx).  Any missing files will *also*
cause the kind of problem you are seeing.

Thanks,
Karl



On Thu, Nov 3, 2016 at 8:01 PM, Jeff Potts <je...@gmail.com> wrote:

> In my example those are all virtual sites, but for some reason they are
> not accessible via /site. Is that an indication that something is
> misconfigured on the Sharepoint side?
>
> Why would the connector report a working connection but the job be unable
> to navigate the path structure?
>
> Jeff
>
> On Nov 3, 2016, at 6:17 PM, Karl Wright <da...@gmail.com> wrote:
>
> Hi Jeff,
>
> SharePoint's site model is as follows:
>
> -- there is a root site, usually with the URL http://somedomain.com
> -- there are "virtual" sites, which are also full-fledge sites, usually
> with the URL http://somedomain.com/site/site_name
> -- there are subsites of the root site, usually with URLs like
> http://somedomain.com/subsiteX
> -- there are subsites of virtual sites, usually with URLs like
> http://somedomain.com/site/site_name/subsiteY
>
> Your MCF sharepoint connections should not point to subsites; they should
> only point to the root site or a single virtual site.  There is no way to
> deal with crawling multiple virtual sites with the same connection, which
> is what the documentation is trying to say.
>
> I hope this helps.
>
> Karl
>
>
> On Thu, Nov 3, 2016 at 6:58 PM, Jeff Potts <je...@gmail.com> wrote:
>
>> I am using apache-manifoldcf-2.5. I am trying to get the SharePoint
>> Connector to work against SharePoint 2003.
>>
>> My SharePoint site URLs are like:
>> http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
>> http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
>> http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx
>>
>> I have set up a Connector with:
>>
>> Server protocol: http
>> Server name: somedomain.com
>> Server port: 80
>> Site path: /SomeSite
>>
>> The connection status says, "Connection working".
>>
>> However, when I go to set up a job, and I click the "Paths" tab, I see
>> the message, "Root site is unreachable, or user has no permissions".
>>
>> I know I have permissions to that site because I can read/write to its
>> document library through the web interface. So I feel like I may be missing
>> something regarding the correct value for site path.
>>
>> The docs say, "...type in the portion of the root site URL that includes
>> everything after the server and port, except for the final "aspx" file,"
>> which is why I am using "/SomeSite" as the site path.
>>
>> But another snippet from the docs says this:
>>
>> "The SharePoint connection type is designed so that one SharePoint
>> repository connection can access all SharePoint sites from a specific root
>> site though its explicit subsites. It is the case that it is desirable in
>> some very large SharePoint installations to access *all* SharePoint
>> sites using a single connection. But the ManifoldCF SharePoint connection
>> type does not support that model as of yet."
>>
>> That confused me because it isn't clear if I should be specifying some
>> sort of root site (which I have no idea how to determine the path for) or
>> if creating one connection for each site should work.
>>
>> Jeff
>>
>>
>>
>

Re: SharePoint Connector: "Root site is unreachable"

Posted by Jeff Potts <je...@gmail.com>.
In my example those are all virtual sites, but for some reason they are not accessible via /site. Is that an indication that something is misconfigured on the Sharepoint side?

Why would the connector report a working connection but the job be unable to navigate the path structure?

Jeff

> On Nov 3, 2016, at 6:17 PM, Karl Wright <da...@gmail.com> wrote:
> 
> Hi Jeff,
> 
> SharePoint's site model is as follows:
> 
> -- there is a root site, usually with the URL http://somedomain.com
> -- there are "virtual" sites, which are also full-fledge sites, usually with the URL http://somedomain.com/site/site_name
> -- there are subsites of the root site, usually with URLs like http://somedomain.com/subsiteX
> -- there are subsites of virtual sites, usually with URLs like http://somedomain.com/site/site_name/subsiteY
> 
> Your MCF sharepoint connections should not point to subsites; they should only point to the root site or a single virtual site.  There is no way to deal with crawling multiple virtual sites with the same connection, which is what the documentation is trying to say.
> 
> I hope this helps.
> 
> Karl
> 
> 
>> On Thu, Nov 3, 2016 at 6:58 PM, Jeff Potts <je...@gmail.com> wrote:
>> I am using apache-manifoldcf-2.5. I am trying to get the SharePoint Connector to work against SharePoint 2003.
>> 
>> My SharePoint site URLs are like:
>> http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
>> http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
>> http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx
>> 
>> I have set up a Connector with:
>> 
>> Server protocol: http
>> Server name: somedomain.com
>> Server port: 80
>> Site path: /SomeSite
>> 
>> The connection status says, "Connection working".
>> 
>> However, when I go to set up a job, and I click the "Paths" tab, I see the message, "Root site is unreachable, or user has no permissions".
>> 
>> I know I have permissions to that site because I can read/write to its document library through the web interface. So I feel like I may be missing something regarding the correct value for site path.
>> 
>> The docs say, "...type in the portion of the root site URL that includes everything after the server and port, except for the final "aspx" file," which is why I am using "/SomeSite" as the site path.
>> 
>> But another snippet from the docs says this:
>> 
>> "The SharePoint connection type is designed so that one SharePoint repository connection can access all SharePoint sites from a specific root site though its explicit subsites. It is the case that it is desirable in some very large SharePoint installations to access all SharePoint sites using a single connection. But the ManifoldCF SharePoint connection type does not support that model as of yet."
>> 
>> That confused me because it isn't clear if I should be specifying some sort of root site (which I have no idea how to determine the path for) or if creating one connection for each site should work.
>> 
>> Jeff
>> 
>> 
> 

Re: SharePoint Connector: "Root site is unreachable"

Posted by Karl Wright <da...@gmail.com>.
Hi Jeff,

SharePoint's site model is as follows:

-- there is a root site, usually with the URL http://somedomain.com
-- there are "virtual" sites, which are also full-fledge sites, usually
with the URL http://somedomain.com/site/site_name
-- there are subsites of the root site, usually with URLs like
http://somedomain.com/subsiteX
-- there are subsites of virtual sites, usually with URLs like
http://somedomain.com/site/site_name/subsiteY

Your MCF sharepoint connections should not point to subsites; they should
only point to the root site or a single virtual site.  There is no way to
deal with crawling multiple virtual sites with the same connection, which
is what the documentation is trying to say.

I hope this helps.

Karl


On Thu, Nov 3, 2016 at 6:58 PM, Jeff Potts <je...@gmail.com> wrote:

> I am using apache-manifoldcf-2.5. I am trying to get the SharePoint
> Connector to work against SharePoint 2003.
>
> My SharePoint site URLs are like:
> http://somedomain.com/SomeSite/_layouts/viewlsts.aspx
> http://somedomain.com/SomeOtherSite/_layouts/viewlsts.aspx
> http://somedomain.com/YetAnotherSite/_layouts/viewlsts.aspx
>
> I have set up a Connector with:
>
> Server protocol: http
> Server name: somedomain.com
> Server port: 80
> Site path: /SomeSite
>
> The connection status says, "Connection working".
>
> However, when I go to set up a job, and I click the "Paths" tab, I see the
> message, "Root site is unreachable, or user has no permissions".
>
> I know I have permissions to that site because I can read/write to its
> document library through the web interface. So I feel like I may be missing
> something regarding the correct value for site path.
>
> The docs say, "...type in the portion of the root site URL that includes
> everything after the server and port, except for the final "aspx" file,"
> which is why I am using "/SomeSite" as the site path.
>
> But another snippet from the docs says this:
>
> "The SharePoint connection type is designed so that one SharePoint
> repository connection can access all SharePoint sites from a specific root
> site though its explicit subsites. It is the case that it is desirable in
> some very large SharePoint installations to access *all* SharePoint sites
> using a single connection. But the ManifoldCF SharePoint connection type
> does not support that model as of yet."
>
> That confused me because it isn't clear if I should be specifying some
> sort of root site (which I have no idea how to determine the path for) or
> if creating one connection for each site should work.
>
> Jeff
>
>
>