[ensembl-dev] prototype links to ensembl

[Andrew Tikhonov]

Dear Ensembl-dev,

	Any ideas on the below ? Please could someone help us ?

	Thanks a lot in advance.


Regards,
Andrew


On 20 Jun 2012, at 12:42, Nikolay Kolesnikov wrote:

> Hi Andrew,
> 
> Can you please forward this to Ensembl?
> 
> I captured the requests coming from Ensembl (ips 193.162.202.241, 193.162.202.242) and it seems like I'm not getting the partial request headers from them (Range: xxxxx) so the servlet assumes the request is for a complete download:
> 
> request 1 (HEAD):
> 
> HEAD /arrayexpress/files/E-GEOD-30206/E-GEOD-30206.BAM.SRR299029.bam HTTP/1.1
> User-Agent: genome.ucsc.edu/net.c
> Host: 93.97.26.5
> Accept: */*
> Via: 1.0 cachesrv1a.internal.sanger.ac.uk (squid/3.1.19)
> X-Forwarded-For: 172.17.22.7
> Cache-Control: max-age=259200
> Connection: keep-alive
> 
> response 1:
> 
> HTTP/1.1 200 OK
> Server: Apache-Coyote/1.1
> Content-Disposition: inline;filename="E-GEOD-30206.BAM.SRR299029.bam"
> Accept-Ranges: bytes
> ETag: E-GEOD-30206.BAM.SRR299029.bam_418450276_1338211716000
> Last-Modified: Mon, 28 May 2012 13:28:36 GMT
> Content-Range: bytes 0-418450275/418450276
> Content-Type: application/octet-stream
> Content-Length: 418450276
> Date: Wed, 20 Jun 2012 11:33:00 GMT
> 
> request 2 (GET):
> 
> GET /arrayexpress/files/E-GEOD-30206/E-GEOD-30206.BAM.SRR299029.bam HTTP/1.1
> User-Agent: genome.ucsc.edu/net.c
> Host: 93.97.26.5
> Accept: */*
> Via: 1.0 cachesrv1b.internal.sanger.ac.uk (squid/3.1.19)
> X-Forwarded-For: 172.17.22.7
> Cache-Control: max-age=259200
> Connection: keep-alive
> 
> response:
> 
> HTTP/1.1 200 OK
> Server: Apache-Coyote/1.1
> Content-Disposition: inline;filename="E-GEOD-30206.BAM.SRR299029.bam"
> Accept-Ranges: bytes
> ETag: E-GEOD-30206.BAM.SRR299029.bam_418450276_1338211716000
> Last-Modified: Mon, 28 May 2012 13:28:36 GMT
> Content-Range: bytes 0-418450275/418450276
> Content-Type: application/octet-stream
> Content-Length: 418450276
> Date: Wed, 20 Jun 2012 11:33:00 GMT
> 
> 
> On 20 Jun 2012, at 11:36, Andrew Tikhonov wrote:
> 
>> Ah, there's a separate servlet for downloading in AE.
>> 
>> Here's what Ensembl requires -- basically 206 instead of 200 if a range is requested. If you could add that and see how it goes, it'd be really awesome.
>> 
>> 
>>> After a quick check it appears that there is something in your server settings that is limiting this upload. In our error logs we have a series of the following messages: 
>>> 
>>> "Byte-range request was ignored by server. 
>>> Expected Partial Content 206. http://wwwdev.ebi.ac.uk/arrayexpress/files/E-MTAB-513/E-MTAB-513.BAM.ERR030856.bam;byterange=0-: 200 OK
>>> 
>>> Can't get data socket for http://wwwdev.ebi.ac.uk/arrayexpress/files/E-MTAB-513/E-MTAB-513.BAM.ERR030856.bam
>>> Byte-range request was ignored by server. 
>>> Expected Partial Content 206. http://wwwdev.ebi.ac.uk/arrayexpress/files/E-MTAB-513/E-MTAB-513.BAM.ERR030856.bam;byterange=0-: 200 OK"
>>> 
>> 
>> I asked for a proper engineer contact who can support us while we test things.
>> 
>> 
>> 
>> Regards,
>> Andrew
>> 
>> 
>> On 20 Jun 2012, at 11:28, Nikolay Kolesnikov wrote:
>> 
>>> The support for partial download request is written by me, not apache and I'd be happy to hear about the issues directly from Ensembl. Can you please connect us/
>>> 
>>> Nikolay.
>>> 
>>> On 20 Jun 2012, at 11:25, Andrew Tikhonov wrote:
>>> 
>>>> Hi all,
>>>> 
>>>> 	I'm in the middle of discussion with Ensembl and ES regarding the first item. It looks like the Apache Web server the ArrayExpress is running on doesn't correctly (according to spec) support partial content request, which Ensembl recognises as a failure. This has been fixed in one of the latest Apache releases. I sent a request to ES to check the version.
>>>> 
>>>> 	Regarding the second item -- Ensembl doesn't allow whole genome view. We discussed it with Johan. Only chromosome view is available. Such a hardcode will need to be replaced with something more sensible if the need arises.
>>>> 
>>>> 
>>>> Regards,
>>>> Andrew
>>>> 
>>>> 
>>>> On 19 Jun 2012, at 16:36, Nikolay Kolesnikov wrote:
>>>> 
>>>>> Hi all,
>>>>> 
>>>>> So I've added links to ENA and Ensembl here http://wwwdev.ebi.ac.uk/arrayexpress/experiments/E-MTAB-513/samples.html
>>>>> 
>>>>> I've encountered two problems:
>>>>> 
>>>>> 1. BAM file isn't recognised by Ensembl - I also checked the links Andrew sent me for the other experiments and there is an error there as well.
>>>>> 
>>>>> 2. Links shows some region of chromosome 1 which is hardcoded in the link. Is there a way to start with some whole genome overview?
>>>>> 
>>>>> Thanks,
>>>>> Nikolay.
>>>>> 
>>>> 
>>> 
>> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ensembl.org/pipermail/dev_ensembl.org/attachments/20120620/558a139e/attachment.html>