[gentoo-user] wget -c strange behaviour

[gentoo-user] wget -c strange behaviour

[Iain Buchanan]

Hi all,

when using wget recently to get some "free" mp3's from Mike Baas'
website, I had to kill it halfway through to do some other downloading.

When I continued, I used `wget -c ...` but I got "strange" behaviour:

$ wget -c
"http://68.106.74.139/mp3/Mike_Baas/4/mp3/Mike_Baas_-_4_-_01_-_Why_Can't_I_Slow_The_World_Down.mp3" --10:38:16--  http://68.106.74.139/mp3/Mike_Baas/4/mp3/Mike_Baas_-_4_-_01_-_Why_Can't_I_Slow_The_World_Down.mp3            => `Mike_Baas_-_4_-_01_-_Why_Can't_I_Slow_The_World_Down.mp3' Connecting to 68.106.74.139:80... connected. HTTP request sent, awaiting response... 200 OK Length: 13,870,871 (13M) [audio/mpeg]

50% [=================>                   ] 13,870,871     3.67M/s
ETA 00:04

10:38:19 (3.69 MB/s) -
`Mike_Baas_-_4_-_01_-_Why_Can't_I_Slow_The_World_Down.mp3' saved
[13870871/13870871]

As you can see, wget didn't continue, instead it tried to download the
complete file again.  It must also be cached by squid, because I
certainly don't get 3.67 M/s on our shared 512k line :)

So the question is: why does wget try to download it from the beginning?
wget shouldn't do this even if the server can't resume, wget should just
die.

And why also does it not appear to finish, even though the mp3 sounds
complete?

many thanks,
-- 
Iain Buchanan <iaindb at netspace dot net dot au>

Loneliness is a terrible price to pay for independence.

-- 
gentoo-user@gentoo.org mailing list

[gentoo-user] Re: wget -c strange behaviour

[Simon Kellett]

Iain Buchanan <iaindb@netspace.net.au> writes:

> So the question is: why does wget try to download it from the
> beginning?  wget shouldn't do this even if the server can't resume,
> wget should just die.

man wget :-)

"Note that -c only works with FTP servers and with HTTP servers that
support the "Range" header."

I am guessing his does not !

-- 
Simon Kellett,        |   Gentoo Linux, Fvwm, Firefox 
Darmstadt, Germany    |              Xemacs, Vm, Gnus

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] Re: wget -c strange behaviour

[Iain Buchanan]

On Mon, 2006-03-27 at 22:33 +0200, Simon Kellett wrote:
> Iain Buchanan <iaindb@netspace.net.au> writes:
> 
> > So the question is: why does wget try to download it from the
> > beginning?  wget shouldn't do this even if the server can't resume,
> > wget should just die.
> 
> man wget :-)

ok:

  Beginning with Wget 1.7, if you use -c on a non-empty file, and it
  turns out that the server does not support continued downloading,
  Wget will refuse to start the download from scratch, which would
  effectively ruin existing contents.  If you really want the down-
  load to start from scratch, remove the file

and yet the file did start from scratch...

> "Note that -c only works with FTP servers and with HTTP servers that
> support the "Range" header."
> 
> I am guessing his does not !

actually, it does, as on another file:

$ wget -c
"http://68.106.74.139/mp3/Mike_Baas/4/mp3/Mike_Baas_-_4_-_03_-_Wave.mp3"
--09:56:35--
http://68.106.74.139/mp3/Mike_Baas/4/mp3/Mike_Baas_-_4_-_03_-_Wave.mp3
           => `Mike_Baas_-_4_-_03_-_Wave.mp3'
Connecting to 68.106.74.139:80... connected.
HTTP request sent, awaiting response... 206 Partial Content
Length: 13,985,672 (13M), 234,426 (229K) remaining [audio/mpeg]

100%[++++++++++++++++++++++++++++++++++++>] 13,985,672     8.97K/s
ETA 00:00

09:56:57 (11.18 KB/s) - `Mike_Baas_-_4_-_03_-_Wave.mp3' saved
[13985672/13985672]


But even if it didn't this still wouldn't explain why the download is
deleted, and then downloaded again.  -c shouldn't do that.

thanks,
-- 
Iain Buchanan <iaindb at netspace dot net dot au>

If you want to know what god thinks of money, just look at the people he gave
it to.
		-- Dorthy Parker

-- 
gentoo-user@gentoo.org mailing list