Re: [galaxy-dev] download dataset from the proxy-ed Galaxy failed with broken link

2012-10-29 Thread Derrick Lin
If I downloaded with curl -i, the page it returned was:

pwbcad@gamma01:tmp$ curl -i
http://galaxy-qa.garvan.unsw.edu.au/datasets/dc097fa4e7ad53f6/display?to_ext=fastqsanger
HTTP/1.1 302 Found
Date: Tue, 30 Oct 2012 04:25:21 GMT
Server: PasteWSGIServer/0.5 Python/2.6.6
content-type: text/html
location: /root?dataset_id=dc097fa4e7ad53f6filename=None
Set-Cookie:
galaxysession=c6ca0ddb55be603a4107d5e4793b9ce28e93ce1e646977c209d1af46a242147b57953de37776fcc2;
expires=Mon, 28-Jan-2013 15:25:21 GMT; httponly; Max-Age=7776000; Path=/;
Version=1
Vary: Accept-Encoding
Transfer-Encoding: chunked

html
  headtitleFound/title/head
  body
h1Found/h1
pThe resource was found at a
href=/root?dataset_id=dc097fa4e7ad53f6amp;filename=None/root?dataset_id=dc097fa4e7ad53f6amp;filename=None/a;
you should be redirected automatically.

!--  --/p
hr noshade
div align=rightWSGI Server/div
  /body
/html

Don't know why the apache returned 302 only when access via wget or curl.

Derrick

On Sat, Oct 27, 2012 at 7:44 AM, Derrick Lin klin...@gmail.com wrote:

 Hi Nate,

 I checked and what downloaded is a webpage, I opened it on browser, and
 screenshot attached.

 Basically it displays error saying File not found. But dataset download
 works fine via browser.

 Cheers,
 Derrick

 On Fri, Oct 26, 2012 at 11:44 PM, Nate Coraor n...@bx.psu.edu wrote:

 On Oct 24, 2012, at 11:15 PM, Derrick Lin wrote:

  Hi guys,
 
  On our Galaxy is proxy-ed via Apache according to the wiki guide, the
 normal download dataset via browser is working fine.
 
  However, when I tried download from command line via wget or curl -O (I
 copied the dataset URL by right clicking the floppy disk icon).
 
  And here is the wget output:
 
  pwbcad@gamma01:tmp$ wget
 http://galaxy.demo.edu.au/galaxy/datasets/ccc5aff28d2c8616/display?to_ext=fastq
  --2012-10-25 10:57:42--
 http://galaxy.demo.edu.au/galaxy/datasets/ccc5aff28d2c8616/display?to_ext=fastq
  Resolving galaxy.demo.edu.au... 1xx.x4.1xx.xx5
  Connecting to galaxy.demo.edu.au|1xx.x4.1xx.xx5|:80... connected.
  HTTP request sent, awaiting response... 302 Found
  Location: /galaxy/root?dataset_id=ccc5aff28d2c8616filename=None
 [following]
  --2012-10-25 10:57:42--
 http://galaxy.demo.edu.au/galaxy/root?dataset_id=ccc5aff28d2c8616filename=None
  Connecting to galaxy.demo.edu.au|129.94.136.205|:80... connected.
  HTTP request sent, awaiting response... 200 OK
  Length: unspecified [text/html]
  Saving to: root?dataset_id=ccc5aff28d2c8616filename=None.1
 
  [ =

 ] 28,476  --.-K/s   in 0.001s
 
  2012-10-25 10:57:42 (31.4 MB/s) -
 root?dataset_id=ccc5aff28d2c8616filename=None.1 saved [28476]
 
  It seems like the Apache returned another URL (returned 302, probably
 due to proxy) and the new URL is broken.

 Hi Derrick,

 What are the contents of the file that wget saved?  This should give you
 a clue as to what's going on.

 --nate

 
  Thanks in advance.
 
  Cheers,
  Derrick
  ___
  Please keep all replies on the list by using reply all
  in your mail client.  To manage your subscriptions to this
  and other Galaxy lists, please use the interface at:
 
   http://lists.bx.psu.edu/



___
Please keep all replies on the list by using reply all
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/

Re: [galaxy-dev] download dataset from the proxy-ed Galaxy failed with broken link

2012-10-29 Thread Derrick Lin
Hi Ross,

Good point and it makes sense in term of security. But I just tried it, and
it didn't help.

I captured these two from the Paste log (not apache log):

xx.xx.1xx.xx - - [30/Oct/2012:15:42:17 +1100] GET
/galaxy_dev/datasets/10b866a1b258d1f1/display?to_ext=fastqsanger HTTP/1.1
302 - - Wget/1.12 (linux-gnu)
galaxy.web.framework DEBUG 2012-10-30 15:42:17,676 Error: this request
returned None from get_history():
http://pwbc.garvan.unsw.edu.au/galaxy_dev/root
xx.xx.1xx.xx - - [30/Oct/2012:15:42:17 +1100] GET
/galaxy_dev/root?dataset_id=10b866a1b258d1f1filename=None HTTP/1.1 200 -
- Wget/1.12 (linux-gnu)

Another observation is, I can wget my own dataset from the main Galaxy
straight away without providing any credential.

Maybe something wrong with my universe ini file.

Derrick

On Tue, Oct 30, 2012 at 3:32 PM, Ross ross.laza...@gmail.com wrote:

 That url requires a login from here - have you tried supplying it - eg
 curl -i -u user:password target

 On Tue, Oct 30, 2012 at 3:27 PM, Derrick Lin klin...@gmail.com wrote:
  If I downloaded with curl -i, the page it returned was:
 
  pwbcad@gamma01:tmp$ curl -i
 
 http://galaxy-qa.garvan.unsw.edu.au/datasets/dc097fa4e7ad53f6/display?to_ext=fastqsanger
  HTTP/1.1 302 Found
  Date: Tue, 30 Oct 2012 04:25:21 GMT
  Server: PasteWSGIServer/0.5 Python/2.6.6
  content-type: text/html
  location: /root?dataset_id=dc097fa4e7ad53f6filename=None
  Set-Cookie:
 
 galaxysession=c6ca0ddb55be603a4107d5e4793b9ce28e93ce1e646977c209d1af46a242147b57953de37776fcc2;
  expires=Mon, 28-Jan-2013 15:25:21 GMT; httponly; Max-Age=7776000; Path=/;
  Version=1
  Vary: Accept-Encoding
  Transfer-Encoding: chunked
 
  html
headtitleFound/title/head
body
  h1Found/h1
  pThe resource was found at a
 
 href=/root?dataset_id=dc097fa4e7ad53f6amp;filename=None/root?dataset_id=dc097fa4e7ad53f6amp;filename=None/a;
  you should be redirected automatically.
 
  !--  --/p
  hr noshade
  div align=rightWSGI Server/div
/body
  /html
 
  Don't know why the apache returned 302 only when access via wget or curl.
 
  Derrick
 
  On Sat, Oct 27, 2012 at 7:44 AM, Derrick Lin klin...@gmail.com wrote:
 
  Hi Nate,
 
  I checked and what downloaded is a webpage, I opened it on browser, and
  screenshot attached.
 
  Basically it displays error saying File not found. But dataset download
  works fine via browser.
 
  Cheers,
  Derrick
 
  On Fri, Oct 26, 2012 at 11:44 PM, Nate Coraor n...@bx.psu.edu wrote:
 
  On Oct 24, 2012, at 11:15 PM, Derrick Lin wrote:
 
   Hi guys,
  
   On our Galaxy is proxy-ed via Apache according to the wiki guide, the
   normal download dataset via browser is working fine.
  
   However, when I tried download from command line via wget or curl -O
 (I
   copied the dataset URL by right clicking the floppy disk icon).
  
   And here is the wget output:
  
   pwbcad@gamma01:tmp$ wget
  
 http://galaxy.demo.edu.au/galaxy/datasets/ccc5aff28d2c8616/display?to_ext=fastq
   --2012-10-25 10:57:42--
  
 http://galaxy.demo.edu.au/galaxy/datasets/ccc5aff28d2c8616/display?to_ext=fastq
   Resolving galaxy.demo.edu.au... 1xx.x4.1xx.xx5
   Connecting to galaxy.demo.edu.au|1xx.x4.1xx.xx5|:80... connected.
   HTTP request sent, awaiting response... 302 Found
   Location: /galaxy/root?dataset_id=ccc5aff28d2c8616filename=None
   [following]
   --2012-10-25 10:57:42--
  
 http://galaxy.demo.edu.au/galaxy/root?dataset_id=ccc5aff28d2c8616filename=None
   Connecting to galaxy.demo.edu.au|129.94.136.205|:80... connected.
   HTTP request sent, awaiting response... 200 OK
   Length: unspecified [text/html]
   Saving to: root?dataset_id=ccc5aff28d2c8616filename=None.1
  
   [ =
   ] 28,476  --.-K/s   in 0.001s
  
   2012-10-25 10:57:42 (31.4 MB/s) -
   root?dataset_id=ccc5aff28d2c8616filename=None.1 saved [28476]
  
   It seems like the Apache returned another URL (returned 302, probably
   due to proxy) and the new URL is broken.
 
  Hi Derrick,
 
  What are the contents of the file that wget saved?  This should give
 you
  a clue as to what's going on.
 
  --nate
 
  
   Thanks in advance.
  
   Cheers,
   Derrick
   ___
   Please keep all replies on the list by using reply all
   in your mail client.  To manage your subscriptions to this
   and other Galaxy lists, please use the interface at:
  
http://lists.bx.psu.edu/
 
 
 
 
  ___
  Please keep all replies on the list by using reply all
  in your mail client.  To manage your subscriptions to this
  and other Galaxy lists, please use the interface at:
 
http://lists.bx.psu.edu/



 --
 Ross Lazarus MBBS MPH;
 Head, Medical Bioinformatics, BakerIDI; Tel: +61 385321444
 http://scholar.google.com/citations?hl=enuser=UCUuEM4J

___
Please keep all replies on the list by using reply all
in your mail client.  To manage your subscriptions to this

Re: [galaxy-dev] download dataset from the proxy-ed Galaxy failed with broken link

2012-10-29 Thread Derrick Lin
On Tue, Oct 30, 2012 at 4:03 PM, Ross ross.laza...@gmail.com wrote:

 Doesn't make sense that you can wget directly from the paste process
 if you have security properly configured!


That's what you can do with the main Galaxy site. Not sure if that's
intentional though. You can try

wget --no-check-certificate
https://main.g2.bx.psu.edu/datasets/851fbe5b7ff576a6/display?to_ext=tabular

See if you can download my own dataset without providing any credential.


 Do you have Apache authenticating and passing headers through to the
 paste process - this section in universe_wsgi.ini

 # User authentication can be delegated to an upstream proxy server (usually
 # Apache).  The upstream proxy should set a REMOTE_USER header in the
 request.
 # Enabling remote user disables regular logins.  For more information, see:
 # http://wiki.g2.bx.psu.edu/Admin/Config/Apache%20Proxy
 use_remote_user = True

 If so, you should not be able to access anything via the paste process
 directly without adding authentication headers. Once that's fixed (you
 do NOT want anyone to be able to do what you can do - it bypasses all
 security!) the apache configuration will probably need tweaking. It's
 hard to advise - it's mostly voodoo IMHO - do you have an apache
 fluent sysadmin?


We don't use Apache for authentication. Thus use_remote_user = False. Will
read more about this remote user thing.

Cheers
Derrick
___
Please keep all replies on the list by using reply all
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/

Re: [galaxy-dev] download dataset from the proxy-ed Galaxy failed with broken link

2012-10-26 Thread Nate Coraor
On Oct 24, 2012, at 11:15 PM, Derrick Lin wrote:

 Hi guys,
 
 On our Galaxy is proxy-ed via Apache according to the wiki guide, the normal 
 download dataset via browser is working fine.
 
 However, when I tried download from command line via wget or curl -O (I 
 copied the dataset URL by right clicking the floppy disk icon).
 
 And here is the wget output:
 
 pwbcad@gamma01:tmp$ wget 
 http://galaxy.demo.edu.au/galaxy/datasets/ccc5aff28d2c8616/display?to_ext=fastq
 --2012-10-25 10:57:42--  
 http://galaxy.demo.edu.au/galaxy/datasets/ccc5aff28d2c8616/display?to_ext=fastq
 Resolving galaxy.demo.edu.au... 1xx.x4.1xx.xx5
 Connecting to galaxy.demo.edu.au|1xx.x4.1xx.xx5|:80... connected.
 HTTP request sent, awaiting response... 302 Found
 Location: /galaxy/root?dataset_id=ccc5aff28d2c8616filename=None [following]
 --2012-10-25 10:57:42--  
 http://galaxy.demo.edu.au/galaxy/root?dataset_id=ccc5aff28d2c8616filename=None
 Connecting to galaxy.demo.edu.au|129.94.136.205|:80... connected.
 HTTP request sent, awaiting response... 200 OK
 Length: unspecified [text/html]
 Saving to: root?dataset_id=ccc5aff28d2c8616filename=None.1
 
 [ = 
   
 ] 28,476  --.-K/s   in 0.001s
 
 2012-10-25 10:57:42 (31.4 MB/s) - 
 root?dataset_id=ccc5aff28d2c8616filename=None.1 saved [28476]
 
 It seems like the Apache returned another URL (returned 302, probably due to 
 proxy) and the new URL is broken.

Hi Derrick,

What are the contents of the file that wget saved?  This should give you a clue 
as to what's going on.

--nate

 
 Thanks in advance.
 
 Cheers,
 Derrick
 ___
 Please keep all replies on the list by using reply all
 in your mail client.  To manage your subscriptions to this
 and other Galaxy lists, please use the interface at:
 
  http://lists.bx.psu.edu/


___
Please keep all replies on the list by using reply all
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

  http://lists.bx.psu.edu/