s-andrews / FastQC

A quality control analysis tool for high throughput sequencing data
GNU General Public License v3.0
425 stars 84 forks source link

Download problems for FastQC #119

Open s-andrews opened 1 year ago

s-andrews commented 1 year ago

Over the last week or so we've had multiple reports of people failing to download fastqc, these mostly seem to be from the states and are getting either extremely slow downloads, or downloads which terminate unexpectedly.

s-andrews commented 1 year ago

I had a look in our web server logs and there's no sign of any errors when downloading fastqc.

In the access logs I can see many requests for the fastqc zip file and all of these either get a status of 200, or get a 301 status, presumably because they tried http and were redirected to https, and then a subsequent request which got 200 status.

Nothing suggests any problems from our end.

s-andrews commented 1 year ago

Just tried a download from an EC2 instance in London:

[ec2-user@ip-172-31-26-222 ~]$ wget https://www.bioinformatics.babraham.ac.uk/projects/fastqc/fastqc_v0.12.1.zip
--2023-06-12 08:21:56--  https://www.bioinformatics.babraham.ac.uk/projects/fastqc/fastqc_v0.12.1.zip
Resolving www.bioinformatics.babraham.ac.uk (www.bioinformatics.babraham.ac.uk)... 149.155.133.4
Connecting to www.bioinformatics.babraham.ac.uk (www.bioinformatics.babraham.ac.uk)|149.155.133.4|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 11709692 (11M) [application/zip]
Saving to: ‘fastqc_v0.12.1.zip’

fastqc_v0.12.1.zip               100%[=========================================================>]  11.17M  7.49MB/s    in 1.5s

2023-06-12 08:21:58 (7.49 MB/s) - ‘fastqc_v0.12.1.zip’ saved [11709692/11709692]
s-andrews commented 1 year ago

Tried a download from California:

[ec2-user@ip-172-31-28-125 ~]$ wget https://www.bioinformatics.babraham.ac.uk/projects/fastqc/fastqc_v0.12.1.zip
--2023-06-12 08:25:26--  https://www.bioinformatics.babraham.ac.uk/projects/fastqc/fastqc_v0.12.1.zip
Resolving www.bioinformatics.babraham.ac.uk (www.bioinformatics.babraham.ac.uk)... 149.155.133.4
Connecting to www.bioinformatics.babraham.ac.uk (www.bioinformatics.babraham.ac.uk)|149.155.133.4|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 11709692 (11M) [application/zip]
Saving to: ‘fastqc_v0.12.1.zip’

fastqc_v0.12.1.zip               100%[=========================================================>]  11.17M  2.85MB/s    in 4.3s

2023-06-12 08:25:32 (2.59 MB/s) - ‘fastqc_v0.12.1.zip’ saved [11709692/11709692]
s-andrews commented 1 year ago

Tried from the Ohio datacentre which is the closest I can find to the site which last reported a problem:

[ec2-user@ip-172-31-21-220 ~]$ wget https://www.bioinformatics.babraham.ac.uk/projects/fastqc/fastqc_v0.12.1.zip
--2023-06-12 08:29:40--  https://www.bioinformatics.babraham.ac.uk/projects/fastqc/fastqc_v0.12.1.zip
Resolving www.bioinformatics.babraham.ac.uk (www.bioinformatics.babraham.ac.uk)... 149.155.133.4
Connecting to www.bioinformatics.babraham.ac.uk (www.bioinformatics.babraham.ac.uk)|149.155.133.4|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 11709692 (11M) [application/zip]
Saving to: ‘fastqc_v0.12.1.zip’

fastqc_v0.12.1.zip               100%[=========================================================>]  11.17M  1.22MB/s    in 7.3s

2023-06-12 08:29:48 (1.53 MB/s) - ‘fastqc_v0.12.1.zip’ saved [11709692/11709692]
s-andrews commented 1 year ago

From everything I can see our site is available and working and the routes to it from various places across the US (where the problems have mostly been reported) are also OK.

It could be a timing thing - maybe something is getting messed up at certain times of day? I'll ask our IT group to check their logs and see if they saw any traffic spikes.

s-andrews commented 1 year ago

OK, so we've reproduced this, albeit not with fastqc. I've been building training images on the EC2 California data centre and our big data downloads are stalling then hanging. No diagnostics from it, but it looks like exactly the same behaviour as was reported here.