Re: Release date for Privoxy 3.0.20 beta

On 2013-01-01, at 8:55 AM, Fabian Keil wrote:

> Lee <[email protected]> wrote:
> 
>> On 12/31/12, Fabian Keil <[email protected]> wrote:
>>> Lee <[email protected]> wrote:
>>> 
>>>> On 12/29/12, Fabian Keil <[email protected]> wrote:
>>>>> I believe there are enough changes in CVS to warrant a new release.
>>>>> 
>>>>> As several of the changes are HTTP-related and affect pretty
>>>>> much every request, I think it should be a beta.
>>>>> 
>>>>> I propose we release Privoxy 3.0.20 beta in the middle of
>>>>> January and 3.0.21 stable a couple of weeks later.
> 
> New proposal: about 2 weeks after the Windows-specific issues are fixed.

Sounds okay - and prompts me to actually consider building an OS X installer for alpha testing!

Ian

> 
>>>> restart privoxy, restart firefox, clear the cache, go here
>>>>  http://www.microsoft.com/en-us/download/default.aspx
>>>> then click on the link for the directx download
>>>>  http://www.microsoft.com/en-us/download/details.aspx?id=35
>>>> 
>>>> The 2nd page never displays
>>>> 
>>>> I'll send the log + my config in a unicast
>>> 
>>> Thanks.
>>> 
>>> Please try the attached patch. If it doesn't help,
>> 
>> browser immediately comes back with an error message about the proxy
>> closing the connection
> 
> I managed to get the logic wrong in a last-minute change. It should be:
> 
> +         if (!data_is_available(fd, 0))
> +         {
> +            break;
> +         }
> 
> It's unexpected (to me) that the incorrect version had any effect
> at all, though, so there probably is another bug somewhere and
> the corrected version might not work either.
> 
>>> please temporarily
>>> reduce drain_and_close_socket() to the close_socket() call at the end.
>> 
>> That works, but privoxy complains about 'The client side of the
>> connection on socket NNN got closed without sending a complete request
>> line.' when clicking on the link to the 2nd page.
> 
> Please send me a a Privoxy log with:
> 
> debug 1
> debug 2
> debug 4
> debug 8
> debug 16
> debug 32
> debug 128
> debug 256
> debug 512
> debug 1024
> debug 4096
> debug 8192
> debug 32768
> debug 65536
> 
> Btw, can you reproduce the problem by clicking around in the
> CGI interface, or only with microsoft.com?
> 
>> I changed data_is_available:
>>   /* LR was: return ((n == 1) && (1 == recv(fd, buf, 1, MSG_PEEK))); */
>>   status =  recv(fd, buf, 1, MSG_PEEK);
>>   log_error(LOG_LEVEL_INFO, "data_is_available: socket %d select=%d
>> recv=%d", fd, n, status);   /* LR */
>>   return ((n == 1) && (1 == status));
>> }
>> 
>> 
>> 2012-12-31 16:31:25.155 00000d20 Request:
>> http://i.microsoft.com/global/ImageStore/PublishingImages/FY13/Asset/features/WinEval_V2_1127_920x330_EN_US.jpg
>> 2012-12-31 16:31:25.233 00000ebc Info: data_is_available: socket 556
>> select=1 recv=1
>> 2012-12-31 16:31:25.233 00000ebc Info: data_is_available: socket 556
>> select=1 recv=1
>> 2012-12-31 16:31:25.233 00000ebc Request:
>> http://i.microsoft.com/global/ImageStore/PublishingImages/FY13/Asset/features/VS_1127_920x330_EN_US.jpg
>> 2012-12-31 16:31:29.304 00000dc8 Info: data_is_available: socket 1696
>> select=1 recv=0
>> 2012-12-31 16:31:29.304 00000dc8 Error: The client side of the
>> connection on socket 1696 got closed without sending a complete
>> request line.
> 
> That looks reasonable.
> 
>> 2012-12-31 16:31:29.304 00000dc8 Info: data_is_available: socket 1696
>> select=1 recv=0
>> 2012-12-31 16:31:30.084 00000d20 Info: data_is_available: socket 192
>> select=1 recv=1
>> 2012-12-31 16:31:30.084 00000d20 Info: data_is_available: socket 192
>> select=1 recv=1
>> 2012-12-31 16:31:30.084 00000d20 Request:
>> http://www.microsoft.com/en-us/download/details.aspx?id=35
>> 2012-12-31 16:31:30.552 00000d20 Info: data_is_available: socket 192
>> select=1 recv=1
>> 2012-12-31 16:31:30.552 00000d20 Info: data_is_available: socket 192
>> select=1 recv=1
>> 2012-12-31 16:31:30.568 00000d20 Request:
>> http://i.microsoft.com/en-us/download/style.cssx?k=~/shared/templates/components/mscomViews/grid/grid-css.aspx;~/shared/templates/components/mscomViews/controls/featureitem/featureItem-css.aspx;~/shared/templates/components/mscomViews/List/list-css.aspx&sc=/en-us/download/site.config&pc=/en-us/download/PageConfig/details.config.xml&v=1185941585
>> 2012-12-31 16:31:30.552 00000ea0 Info: data_is_available: socket 420
>> select=0 recv=1
>> 2012-12-31 16:31:30.568 00000ea0 Info: data_is_available: socket 420
>> select=1 recv=1
>> 2012-12-31 16:31:30.552 000006cc Info: data_is_available: socket 880
>> select=0 recv=1
>> 2012-12-31 16:31:30.568 000006cc Info: data_is_available: socket 880
>> select=1 recv=1
>> 2012-12-31 16:31:30.552 00000d98 Info: data_is_available: socket 476
>> select=0 recv=1
>> 2012-12-31 16:31:30.583 00000d98 Info: data_is_available: socket 476
>> select=1 recv=1
>> 2012-12-31 16:31:30.552 000006f8 Info: data_is_available: socket 532
>> select=0 recv=1
>> 2012-12-31 16:31:30.583 000006f8 Info: data_is_available: socket 532
>> select=1 recv=1
>> 2012-12-31 16:31:30.552 00000ddc Info: data_is_available: socket 808
>> select=0 recv=1
>> 2012-12-31 16:31:30.583 00000ddc Info: data_is_available: socket 808
>> select=1 recv=1
>> 2012-12-31 16:31:30.552 0000068c Info: data_is_available: socket 428
>> select=0 recv=1
>> 2012-12-31 16:31:30.552 00000c08 Info: data_is_available: socket 668
>> select=0 recv=1
>> 2012-12-31 16:31:30.552 00000140 Info: data_is_available: socket 620
>> select=0 recv=1
>> 
>>> My impression is that draining sockets when there is no data
>>> available causes problems:
>> 
>> I don't know about when there is no data available - in this case
>> there is.  I tried logging what got drained and it was the request for
>> the new page:
>> 2012-12-31 14:51:19.981 00000ff4 Request:
>> http://i.microsoft.com/global/ImageStore/PublishingImages/FY13/Asset/features/VS_1127_920x330_EN_US.jpg
>> 2012-12-31 14:51:30.339 00000130 Info: data_is_available: socket 192
>> select=0 recv=1
>> 2012-12-31 14:51:30.339 00000130 Info: data_is_available: socket 192
>> select=1 recv=1
>> 2012-12-31 14:51:30.339 00000130 Info: socket 192 drained 500 bytes:
>> GET http://www.microsoft.com/en-us/download/details.aspx?id=35
> [...]
>> 2012-12-31 14:51:30.370 00000130 Info: socket 192 drained 59 bytes:
>> 9:ss=1xxxxxxxxxxxx; WT_NVR=0=/:1=en-us:2=en-us/download
>> 
>> 2012-12-31 14:52:43.784 0000087c Info: data_is_available: socket 1960
>> select=1 recv=0
>> 2012-12-31 14:52:43.799 0000087c Error: The client side of the
>> connection on socket 1960 got closed without sending a complete
>> request line.
> 
> Note that there are two different threads working with a socket 192.
> Probably the socket number got reused in thread 0000087c after thread
> 00000130 closed its own connection on socket 192.
> 
> Fabian
> ------------------------------------------------------------------------------
> Master SQL Server Development, Administration, T-SQL, SSAS, SSIS, SSRS
> and more. Get SQL Server skills now (including 2012) with LearnDevNow -
> 200+ hours of step-by-step video tutorials by Microsoft MVPs and experts.
> SALE $99.99 this month only - learn more at:
> http://p.sf.net/sfu/learnmore_122512_______________________________________________
> Ijbswa-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/ijbswa-developers


------------------------------------------------------------------------------
Master SQL Server Development, Administration, T-SQL, SSAS, SSIS, SSRS
and more. Get SQL Server skills now (including 2012) with LearnDevNow -
200+ hours of step-by-step video tutorials by Microsoft MVPs and experts.
SALE $99.99 this month only - learn more at:
http://p.sf.net/sfu/learnmore_122512