treebase-devel

[Treebase-devel] Can we do a new build? and proxy timeout?

[William P. <wil...@ya...>]

Hi all,

Can we do a new build on production? (and dev or stage, if you have time)

Youjun recently fixed the delete-tree function, and that will be useful for this guy's issue (see below). Also Youjun fixed the MacFirefox-PhyloWidget problem on the browse side of the equation.  For that, we now pass the tree to the latest version of phylowidget stored on the phylowidget.org website (which doesn't have a problem with Mac-Firefox). 

We still have a problem with MacFirefox-PhyloWidget on the Submit side, but that can't be solved just by pointing to the latest copy on phylowidget.org because we have made custom changes to the code (for saving rerooted trees back the database). 

Finally, Michael Pirie's problem (below) seems to be the result of a time-out on headless-Mesquite ingest. Can we increase the time allotted to this process so as to avoid the time-out? (And, I should mention that when I deleted his extra matrix, I also got a proxy-timeout problem -- but when I came back to the submission, the matrix had been deleted. So again, the functions appear to go to completion eventually even though the browser shows a proxy error. If we could avoid the timeout, it would be better).

Bill



On Mar 31, 2010, at 3:18 PM, William Piel wrote:

> Dear Michael Pirie:
> 
> I think that the proxy error came from a time-out because the parser was taking too long to commit the inserts (e.g. 298 x 11810 = 3,519,380 inserts, not including all the other bits) -- but the inserts happened anyway despite the time-out. And it looks like you backed-up and re-clicked, so there were two matrices and two sets of trees. I deleted one of the matrices -- but I don't want to delete the extra set of trees until we've done a rebuild on our deployment (we recently fixed some code having to do with deleting trees). By tomorrow, feel free to delete the extra set of trees. 
> 
> So everything looks okay, despite the fact that the ingest did  a time-out on you. I'm going to suggest that our server gets reset to allow for longer ingest times. 
> 
> regards,
> 
> Bill
> 
> 
> On Mar 31, 2010, at 10:54 AM, Michael D. Pirie via RT wrote:
> 
>> 
>> Wed Mar 31 10:54:10 2010: Request 7708 was acted upon.
>> Transaction: Ticket created by mp...@su...
>>      Queue: Treebase-help
>>    Subject: File upload problem submission 10367
>>      Owner: Nobody
>> Requestors: mp...@su...
>>     Status: new
>> Ticket <URL: https://help.nescent.org/Ticket/Display.html?id=7708 >
>> 
>> 
>> Dear Sir/Madam,
>> I get the following error when attempting to upload a file to my 
>> submission 10367:
>> 
>> 
>> Proxy Error
>> 
>> The proxy server received an invalid response from an upstream server.
>> The proxy server could not handle the request POST 
>> /treebase-web/user/uploadFile.html.
>> 
>> Reason: Error reading from remote server
>> 
>> Apache/2.2.3 (CentOS) Server at www.treebase.org Port 80
>> 
>> 
>> I'm attaching a zipped copy of the file - would it be possible to add 
>> this for me?
>> Thanks and best wishes,
>> 
>> -- 
>> Dr Michael D. Pirie
>> Department of Biochemistry
>> University of Stellenbosch
>> Private Bag X1, MATIELAND 7602
>> Stellenbosch, Western Cape
>> South Africa
>> 
>> Tel: +27-21-8085841
>> Fax: +27-21-8085863
>> 
>> 
>> 
>> <TreeBASE 10367.zip>
> 

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Jon A. <jon...@du...>]

I've redeployed and bumped the timeout to 15 minutes.
-jon

On Mar 31, 2010, at 3:29 PM, William Piel wrote:

> 
> Hi all,
> 
> Can we do a new build on production? (and dev or stage, if you have time)
> 
> Youjun recently fixed the delete-tree function, and that will be useful for this guy's issue (see below). Also Youjun fixed the MacFirefox-PhyloWidget problem on the browse side of the equation.  For that, we now pass the tree to the latest version of phylowidget stored on the phylowidget.org website (which doesn't have a problem with Mac-Firefox). 
> 
> We still have a problem with MacFirefox-PhyloWidget on the Submit side, but that can't be solved just by pointing to the latest copy on phylowidget.org because we have made custom changes to the code (for saving rerooted trees back the database). 
> 
> Finally, Michael Pirie's problem (below) seems to be the result of a time-out on headless-Mesquite ingest. Can we increase the time allotted to this process so as to avoid the time-out? (And, I should mention that when I deleted his extra matrix, I also got a proxy-timeout problem -- but when I came back to the submission, the matrix had been deleted. So again, the functions appear to go to completion eventually even though the browser shows a proxy error. If we could avoid the timeout, it would be better).
> 
> Bill
> 
> 
> 
> On Mar 31, 2010, at 3:18 PM, William Piel wrote:
> 
>> Dear Michael Pirie:
>> 
>> I think that the proxy error came from a time-out because the parser was taking too long to commit the inserts (e.g. 298 x 11810 = 3,519,380 inserts, not including all the other bits) -- but the inserts happened anyway despite the time-out. And it looks like you backed-up and re-clicked, so there were two matrices and two sets of trees. I deleted one of the matrices -- but I don't want to delete the extra set of trees until we've done a rebuild on our deployment (we recently fixed some code having to do with deleting trees). By tomorrow, feel free to delete the extra set of trees. 
>> 
>> So everything looks okay, despite the fact that the ingest did  a time-out on you. I'm going to suggest that our server gets reset to allow for longer ingest times. 
>> 
>> regards,
>> 
>> Bill
>> 
>> 
>> On Mar 31, 2010, at 10:54 AM, Michael D. Pirie via RT wrote:
>> 
>>> 
>>> Wed Mar 31 10:54:10 2010: Request 7708 was acted upon.
>>> Transaction: Ticket created by mp...@su...
>>>     Queue: Treebase-help
>>>   Subject: File upload problem submission 10367
>>>     Owner: Nobody
>>> Requestors: mp...@su...
>>>    Status: new
>>> Ticket <URL: https://help.nescent.org/Ticket/Display.html?id=7708 >
>>> 
>>> 
>>> Dear Sir/Madam,
>>> I get the following error when attempting to upload a file to my 
>>> submission 10367:
>>> 
>>> 
>>> Proxy Error
>>> 
>>> The proxy server received an invalid response from an upstream server.
>>> The proxy server could not handle the request POST 
>>> /treebase-web/user/uploadFile.html.
>>> 
>>> Reason: Error reading from remote server
>>> 
>>> Apache/2.2.3 (CentOS) Server at www.treebase.org Port 80
>>> 
>>> 
>>> I'm attaching a zipped copy of the file - would it be possible to add 
>>> this for me?
>>> Thanks and best wishes,
>>> 
>>> -- 
>>> Dr Michael D. Pirie
>>> Department of Biochemistry
>>> University of Stellenbosch
>>> Private Bag X1, MATIELAND 7602
>>> Stellenbosch, Western Cape
>>> South Africa
>>> 
>>> Tel: +27-21-8085841
>>> Fax: +27-21-8085863
>>> 
>>> 
>>> 
>>> <TreeBASE 10367.zip>
>> 
> 
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel

-------------------------------------------------------
Jon Auman
Systems Administrator
National Evolutionary Synthesis Center
Duke University
http:www.nescent.org
jon...@ne...
------------------------------------------------------

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[youjun g. <you...@ya...>]

Bill,

In treebase.org the server name in the phylowidget redirect url return
"localhost:8080" because we have a  Apache server in front of Tomcat, and
Apache addressed the Tomact work this way.

I think we have to use purl this time.

Youjun




On Wed, Mar 31, 2010 at 4:46 PM, Jon Auman <jon...@du...> wrote:

> I've redeployed and bumped the timeout to 15 minutes.
> -jon
>
> On Mar 31, 2010, at 3:29 PM, William Piel wrote:
>
>
> Hi all,
>
> Can we do a new build on production? (and dev or stage, if you have time)
>
> Youjun recently fixed the delete-tree function, and that will be useful for
> this guy's issue (see below). Also Youjun fixed the MacFirefox-PhyloWidget
> problem on the browse side of the equation.  For that, we now pass the tree
> to the latest version of phylowidget stored on the phylowidget.org website
> (which doesn't have a problem with Mac-Firefox).
>
> We still have a problem with MacFirefox-PhyloWidget on the Submit side, but
> that can't be solved just by pointing to the latest copy on
> phylowidget.org because we have made custom changes to the code (for
> saving rerooted trees back the database).
>
> Finally, Michael Pirie's problem (below) seems to be the result of a
> time-out on headless-Mesquite ingest. Can we increase the time allotted to
> this process so as to avoid the time-out? (And, I should mention that when I
> deleted his extra matrix, I also got a proxy-timeout problem -- but when I
> came back to the submission, the matrix had been deleted. So again, the
> functions appear to go to completion eventually even though the browser
> shows a proxy error. If we could avoid the timeout, it would be better).
>
> Bill
>
>
>
> On Mar 31, 2010, at 3:18 PM, William Piel wrote:
>
> Dear Michael Pirie:
>
>
> I think that the proxy error came from a time-out because the parser was
> taking too long to commit the inserts (e.g. 298 x 11810 = 3,519,380 inserts,
> not including all the other bits) -- but the inserts happened anyway despite
> the time-out. And it looks like you backed-up and re-clicked, so there were
> two matrices and two sets of trees. I deleted one of the matrices -- but I
> don't want to delete the extra set of trees until we've done a rebuild on
> our deployment (we recently fixed some code having to do with deleting
> trees). By tomorrow, feel free to delete the extra set of trees.
>
>
> So everything looks okay, despite the fact that the ingest did  a time-out
> on you. I'm going to suggest that our server gets reset to allow for longer
> ingest times.
>
>
> regards,
>
>
> Bill
>
>
>
> On Mar 31, 2010, at 10:54 AM, Michael D. Pirie via RT wrote:
>
>
>
> Wed Mar 31 10:54:10 2010: Request 7708 was acted upon.
>
> Transaction: Ticket created by mp...@su...
>
>     Queue: Treebase-help
>
>   Subject: File upload problem submission 10367
>
>     Owner: Nobody
>
> Requestors: mp...@su...
>
>    Status: new
>
> Ticket <URL: https://help.nescent.org/Ticket/Display.html?id=7708 >
>
>
>
> Dear Sir/Madam,
>
> I get the following error when attempting to upload a file to my
>
> submission 10367:
>
>
>
> Proxy Error
>
>
> The proxy server received an invalid response from an upstream server.
>
> The proxy server could not handle the request POST
>
> /treebase-web/user/uploadFile.html.
>
>
> Reason: Error reading from remote server
>
>
> Apache/2.2.3 (CentOS) Server at www.treebase.org Port 80
>
>
>
> I'm attaching a zipped copy of the file - would it be possible to add
>
> this for me?
>
> Thanks and best wishes,
>
>
> --
>
> Dr Michael D. Pirie
>
> Department of Biochemistry
>
> University of Stellenbosch
>
> Private Bag X1, MATIELAND 7602
>
> Stellenbosch, Western Cape
>
> South Africa
>
>
> Tel: +27-21-8085841
>
> Fax: +27-21-8085863
>
>
>
>
> <TreeBASE 10367.zip>
>
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>
>
> -------------------------------------------------------
> Jon Auman
> Systems Administrator
> National Evolutionary Synthesis Center
> Duke University
> http:www.nescent.org
> jon...@ne...
> ------------------------------------------------------
>
>
>
>
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>
>

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Hilmar L. <hl...@ne...>]

Do you need an absolute path that includes the server name, i.e., does  
a relative path not suffice? If you do need a full path, does that  
have to be generated server-side or can that also be done in  
JavaScript (which will yield the correct server name).

	-hilmar

On Mar 31, 2010, at 8:02 PM, youjun guo wrote:

> Bill,
>
> In treebase.org the server name in the phylowidget redirect url  
> return "localhost:8080" because we have a  Apache server in front of  
> Tomcat, and Apache addressed the Tomact work this way.
>
> I think we have to use purl this time.
>
> Youjun
>
>
>
>
> On Wed, Mar 31, 2010 at 4:46 PM, Jon Auman <jon...@du...> wrote:
> I've redeployed and bumped the timeout to 15 minutes.
> -jon
>
> On Mar 31, 2010, at 3:29 PM, William Piel wrote:
>
>>
>> Hi all,
>>
>> Can we do a new build on production? (and dev or stage, if you have  
>> time)
>>
>> Youjun recently fixed the delete-tree function, and that will be  
>> useful for this guy's issue (see below). Also Youjun fixed the  
>> MacFirefox-PhyloWidget problem on the browse side of the equation.   
>> For that, we now pass the tree to the latest version of phylowidget  
>> stored on the phylowidget.org website (which doesn't have a problem  
>> with Mac-Firefox).
>>
>> We still have a problem with MacFirefox-PhyloWidget on the Submit  
>> side, but that can't be solved just by pointing to the latest copy  
>> on phylowidget.org because we have made custom changes to the code  
>> (for saving rerooted trees back the database).
>>
>> Finally, Michael Pirie's problem (below) seems to be the result of  
>> a time-out on headless-Mesquite ingest. Can we increase the time  
>> allotted to this process so as to avoid the time-out? (And, I  
>> should mention that when I deleted his extra matrix, I also got a  
>> proxy-timeout problem -- but when I came back to the submission,  
>> the matrix had been deleted. So again, the functions appear to go  
>> to completion eventually even though the browser shows a proxy  
>> error. If we could avoid the timeout, it would be better).
>>
>> Bill
>>
>>
>>
>> On Mar 31, 2010, at 3:18 PM, William Piel wrote:
>>
>>> Dear Michael Pirie:
>>>
>>> I think that the proxy error came from a time-out because the  
>>> parser was taking too long to commit the inserts (e.g. 298 x 11810  
>>> = 3,519,380 inserts, not including all the other bits) -- but the  
>>> inserts happened anyway despite the time-out. And it looks like  
>>> you backed-up and re-clicked, so there were two matrices and two  
>>> sets of trees. I deleted one of the matrices -- but I don't want  
>>> to delete the extra set of trees until we've done a rebuild on our  
>>> deployment (we recently fixed some code having to do with deleting  
>>> trees). By tomorrow, feel free to delete the extra set of trees.
>>>
>>> So everything looks okay, despite the fact that the ingest did  a  
>>> time-out on you. I'm going to suggest that our server gets reset  
>>> to allow for longer ingest times.
>>>
>>> regards,
>>>
>>> Bill
>>>
>>>
>>> On Mar 31, 2010, at 10:54 AM, Michael D. Pirie via RT wrote:
>>>
>>>>
>>>> Wed Mar 31 10:54:10 2010: Request 7708 was acted upon.
>>>> Transaction: Ticket created by mp...@su...
>>>>     Queue: Treebase-help
>>>>   Subject: File upload problem submission 10367
>>>>     Owner: Nobody
>>>> Requestors: mp...@su...
>>>>    Status: new
>>>> Ticket <URL: https://help.nescent.org/Ticket/Display.html?id=7708 >
>>>>
>>>>
>>>> Dear Sir/Madam,
>>>> I get the following error when attempting to upload a file to my
>>>> submission 10367:
>>>>
>>>>
>>>> Proxy Error
>>>>
>>>> The proxy server received an invalid response from an upstream  
>>>> server.
>>>> The proxy server could not handle the request POST
>>>> /treebase-web/user/uploadFile.html.
>>>>
>>>> Reason: Error reading from remote server
>>>>
>>>> Apache/2.2.3 (CentOS) Server at www.treebase.org Port 80
>>>>
>>>>
>>>> I'm attaching a zipped copy of the file - would it be possible to  
>>>> add
>>>> this for me?
>>>> Thanks and best wishes,
>>>>
>>>> -- 
>>>> Dr Michael D. Pirie
>>>> Department of Biochemistry
>>>> University of Stellenbosch
>>>> Private Bag X1, MATIELAND 7602
>>>> Stellenbosch, Western Cape
>>>> South Africa
>>>>
>>>> Tel: +27-21-8085841
>>>> Fax: +27-21-8085863
>>>>
>>>>
>>>>
>>>> <TreeBASE 10367.zip>
>>>
>>
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Download Intel&#174; Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Treebase-devel mailing list
>> Tre...@li...
>> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>
> -------------------------------------------------------
> Jon Auman
> Systems Administrator
> National Evolutionary Synthesis Center
> Duke University
> http:www.nescent.org
> jon...@ne...
> ------------------------------------------------------
>
>
>
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev_______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel

-- 
===========================================================
: Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
===========================================================

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[William P. <wil...@ya...>]

Whatever works is fine by me... let's fix and redeploy, seeing as the production is broken. 

The path needs to be full, not relative, and while Javascript can do just about anything -- it wouldn't be an easy solution. 

Rather than a PURL, how about just hard-coding "www.treebase.org"? (do we have a parameter init file where this can be stored?)

bp


On Mar 31, 2010, at 8:07 PM, Hilmar Lapp wrote:

> Do you need an absolute path that includes the server name, i.e., does a relative path not suffice? If you do need a full path, does that have to be generated server-side or can that also be done in JavaScript (which will yield the correct server name).
> 
> 	-hilmar
>  
> On Mar 31, 2010, at 8:02 PM, youjun guo wrote:
> 
>> Bill,
>> 
>> In treebase.org the server name in the phylowidget redirect url return "localhost:8080" because we have a  Apache server in front of Tomcat, and Apache addressed the Tomact work this way. 
>> 
>> I think we have to use purl this time.
>> 
>> Youjun

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Hilmar L. <hl...@ne...>]

On Apr 1, 2010, at 9:11 AM, William Piel wrote:

> Rather than a PURL, how about just hard-coding "www.treebase.org"?  
> (do we have a parameter init file where this can be stored?)


That's possible, but I'd do that only as a last resort, because it  
would be different between dev/staging/production deployments.

	-hilmar
-- 
===========================================================
: Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
===========================================================

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Vladimir G. <vla...@du...>]

I currently work on moving the PURL prefix from treebase.properties  
into the Tomcat config, so that we can have back a properly  
functioning development instance.  Since I am already into that, I  
will as well create a new JNDI parameter, say tb2/HostUrl, to hold the  
server name (like "www.treebase.org").

I hope to be done by the end of the day.  Meanwhile, would you prefer  
to revert production to the previous build, or leave it as is?

--VG


On Apr 1, 2010, at 9:11 AM, William Piel wrote:

>
> Whatever works is fine by me... let's fix and redeploy, seeing as  
> the production is broken.
>
> The path needs to be full, not relative, and while Javascript can do  
> just about anything -- it wouldn't be an easy solution.
>
> Rather than a PURL, how about just hard-coding "www.treebase.org"?  
> (do we have a parameter init file where this can be stored?)
>
> bp
>
>
> On Mar 31, 2010, at 8:07 PM, Hilmar Lapp wrote:
>
>> Do you need an absolute path that includes the server name, i.e.,  
>> does a relative path not suffice? If you do need a full path, does  
>> that have to be generated server-side or can that also be done in  
>> JavaScript (which will yield the correct server name).
>>
>> 	-hilmar
>>
>> On Mar 31, 2010, at 8:02 PM, youjun guo wrote:
>>
>>> Bill,
>>>
>>> In treebase.org the server name in the phylowidget redirect url  
>>> return "localhost:8080" because we have a  Apache server in front  
>>> of Tomcat, and Apache addressed the Tomact work this way.
>>>
>>> I think we have to use purl this time.
>>>
>>> Youjun
>
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev_______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[William P. <wil...@ya...>]

On Apr 1, 2010, at 8:51 AM, Vladimir Gapeyev wrote:

>  Meanwhile, would you prefer to revert production to the previous build, or leave it as is?  

uhmm... well, I don't want to lose Youjun's fix that allows for deletion of trees.  

Can we revert to just the changes to the phylowidget links without disturbing other advances?

(or just comment out the phylowidget links with a "temporarily unavailable" hover-over message)

bp

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[William P. <wil...@ya...>]

On Apr 1, 2010, at 8:40 AM, Hilmar Lapp wrote:

> On Apr 1, 2010, at 9:11 AM, William Piel wrote:
> 
>> Rather than a PURL, how about just hard-coding "www.treebase.org"?  
>> (do we have a parameter init file where this can be stored?)
> 
> 
> That's possible, but I'd do that only as a last resort, because it  
> would be different between dev/staging/production deployments.
> 
> 	-hilmar

If using javascript, it would have to change links like this:

http://www.phylowidget.org/full/?tree='http://localhost:8080/treebase-web/tree_for_phylowidget/TB2:Tr300'

to this:

http://www.phylowidget.org/full/?tree='http://www.treebase.org/treebase-web/tree_for_phylowidget/TB2:Tr300'

on two different pages. This page:

http://www.treebase.org/treebase-web/search/treeSearch.html

and this page:

http://www.treebase.org/treebase-web/search/study/trees.html?id=2344

bp

[Treebase-devel] SiteUrl parameter in Tomcat config - for Phylowidget

[Vladimir G. <vla...@du...>]

On Apr 1, 2010, at 10:05 AM, William Piel wrote:

>> On Apr 1, 2010, at 9:11 AM, William Piel wrote:
>>
>>> Rather than a PURL, how about just hard-coding "www.treebase.org"?
>>> (do we have a parameter init file where this can be stored?)


As of rev 664, I introduced a JNDI parameter tb2/SiteUrl for this  
purpose.  See src/main/webapp/META-INF/context.xml.example for usage.   
These is also a convenience method,  
org.cipres.treebase.TreebaseUtil.getSiteUrl() to look it up.

--Vladimir 

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Jon A. <jon...@du...>]

A little php code inserted in the javascript could get the host dynamically:

Of course, that means we've now added yet another framework to the app, but php is already enabled on the apache server.

<script language="Javascript"> 
var hostName = <?= $_SERVER['SERVER_NAME'] ?> 
</script>

-Jon


On Apr 1, 2010, at 10:05 AM, William Piel wrote:

> 
> On Apr 1, 2010, at 8:40 AM, Hilmar Lapp wrote:
> 
>> On Apr 1, 2010, at 9:11 AM, William Piel wrote:
>> 
>>> Rather than a PURL, how about just hard-coding "www.treebase.org"?  
>>> (do we have a parameter init file where this can be stored?)
>> 
>> 
>> That's possible, but I'd do that only as a last resort, because it  
>> would be different between dev/staging/production deployments.
>> 
>> 	-hilmar
> 
> If using javascript, it would have to change links like this:
> 
> http://www.phylowidget.org/full/?tree='http://localhost:8080/treebase-web/tree_for_phylowidget/TB2:Tr300'
> 
> to this:
> 
> http://www.phylowidget.org/full/?tree='http://www.treebase.org/treebase-web/tree_for_phylowidget/TB2:Tr300'
> 
> on two different pages. This page:
> 
> http://www.treebase.org/treebase-web/search/treeSearch.html
> 
> and this page:
> 
> http://www.treebase.org/treebase-web/search/study/trees.html?id=2344
> 
> bp
> 
> 
> 
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel

-------------------------------------------------------
Jon Auman
Systems Administrator
National Evolutionary Synthesis Center
Duke University
http:www.nescent.org
jon...@ne...
------------------------------------------------------

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Hilmar L. <hl...@ne...>]

On Apr 1, 2010, at 10:16 AM, Jon Auman wrote:

> A little php code inserted in the javascript could get the host  
> dynamically:

Just a note, you don't need PHP for this. JavaScript can give you the  
server name directly. The PHP code would be wrong, as it executes  
server-side, not browser-side.

	-hilmar

-- 
===========================================================
: Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
===========================================================

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[youjun g. <you...@ya...>]

Hi, Guys,

These javascrpt work on my server and I expect they will also work on
treebase.org:

function openPhylowidget(tree_id)
        {
             var realURL = "http://www.phylowidget.org/full/?tree='http://
"+location.host+"${url}"+"TB2:Tr"+tree_id+"'";
             window.open(realURL,'myplwidget')
        }

and then

<a href="javascript:void(0)" onClick="openPhylowidget(${tree.id}">

Code was submitted to sourceforge and need a re-deployment to see the
change.

Thank you for all of you, have a good good Friday.

Youjun

On Thu, Apr 1, 2010 at 10:46 AM, Hilmar Lapp <hl...@ne...> wrote:

>
> On Apr 1, 2010, at 10:16 AM, Jon Auman wrote:
>
> A little php code inserted in the javascript could get the host
> dynamically:
>
>
> Just a note, you don't need PHP for this. JavaScript can give you the
> server name directly. The PHP code would be wrong, as it executes
> server-side, not browser-side.
>
> -hilmar
>
> --
> ===========================================================
> : Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
> ===========================================================
>
>
>
>
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>
>

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Hilmar L. <hl...@ne...>]

Youjun -

doesn't this also mean that we are not using a locally modified  
version of PhyloWidget anymore?

	-hilmar

On Apr 1, 2010, at 5:13 PM, youjun guo wrote:

> Hi, Guys,
>
> These javascrpt work on my server and I expect they will also work  
> on treebase.org:
>
> function openPhylowidget(tree_id)
>         {
>              var realURL = "http://www.phylowidget.org/full/?tree='http:// 
> "+location.host+"${url}"+"TB2:Tr"+tree_id+"'";
>              window.open(realURL,'myplwidget')
>         }
>
> and then
>
> <a href="javascript:void(0)" onClick="openPhylowidget(${tree.id}">
>
> Code was submitted to sourceforge and need a re-deployment to see  
> the change.
>
> Thank you for all of you, have a good good Friday.
>
> Youjun
>
> On Thu, Apr 1, 2010 at 10:46 AM, Hilmar Lapp <hl...@ne...>  
> wrote:
>
> On Apr 1, 2010, at 10:16 AM, Jon Auman wrote:
>
>> A little php code inserted in the javascript could get the host  
>> dynamically:
>
> Just a note, you don't need PHP for this. JavaScript can give you  
> the server name directly. The PHP code would be wrong, as it  
> executes server-side, not browser-side.
>
> 	-hilmar
>
> -- 
> ===========================================================
> : Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
> ===========================================================
>
>
>
>
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel
>
>

-- 
===========================================================
: Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
===========================================================

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[William P. <wil...@ya...>]

On Apr 1, 2010, at 4:15 PM, Hilmar Lapp wrote:

> doesn't this also mean that we are not using a locally modified version of PhyloWidget anymore?

We are for the data submission side -- that one is modified to allow saving back to the database.  And so far that one is incompatible with FireFox-for-Mac.  

For the browse side, we use the latest version of PhyloWidget (whatever it is) on phylowidget.org. That one works on FireFox-for-Mac. 

I don't mind keeping a copy of the latest version of PhyloWidget on http://www.treebase.org behind Apache, so that we're not dependent on the proper functioning of phylowidget.org website.  Perhaps Jon can set it up there, and when it's working we'll just change the URL.  

bp

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[William P. <wil...@ya...>]

Thanks Youjun,

the javascript doesn't seem to be responding here:

http://130.132.27.212:8080/treebase-web/search/study/trees.html?id=264

bp



On Apr 1, 2010, at 4:13 PM, youjun guo wrote:

> Hi, Guys,
> 
> These javascrpt work on my server and I expect they will also work on treebase.org:
> 
> function openPhylowidget(tree_id)
>         {
>              var realURL = "http://www.phylowidget.org/full/?tree='http://"+location.host+"${url}"+"TB2:Tr"+tree_id+"'";
>              window.open(realURL,'myplwidget')
>         }
> 
> and then
> 
> <a href="javascript:void(0)" onClick="openPhylowidget(${tree.id}">
> 
> Code was submitted to sourceforge and need a re-deployment to see the change.
> 
> Thank you for all of you, have a good good Friday. 
> 
> Youjun
> 
> On Thu, Apr 1, 2010 at 10:46 AM, Hilmar Lapp <hl...@ne...> wrote:
> 
> On Apr 1, 2010, at 10:16 AM, Jon Auman wrote:
> 
>> A little php code inserted in the javascript could get the host dynamically:
> 
> Just a note, you don't need PHP for this. JavaScript can give you the server name directly. The PHP code would be wrong, as it executes server-side, not browser-side.
> 
> 	-hilmar
> 
> -- 
> ===========================================================
> : Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
> ===========================================================
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel
> 
> 
> ------------------------------------------------------------------------------
> Download Intel&#174; Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev_______________________________________________
> Treebase-devel mailing list
> Tre...@li...
> https://lists.sourceforge.net/lists/listinfo/treebase-devel

William H. Piel
Associate Director for Evolutionary Informatics
Peabody 401, Yale University
170 Whitney Ave.
New Haven CT 06511
(203) 436-4957
wil...@ya...

Re: [Treebase-devel] Can we do a new build? and proxy timeout?

[Hilmar L. <hl...@ne...>]

On Apr 1, 2010, at 5:18 PM, William Piel wrote:

> Thanks Youjun,
>
> the javascript doesn't seem to be responding here:
>
> http://130.132.27.212:8080/treebase-web/search/study/trees.html?id=264

There is a closing parenthesis missing in the openPhylowidget call:

> <a href="javascript:void(0)" onClick="openPhylowidget(${tree.id}">

so the JavaScript has a syntax error.

What I don't understand is why the script snippet needs to be repeated  
multiple times on the page?

	-hilmar

-- 
===========================================================
: Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
===========================================================