Discussion:
[privoxy-devel] Fw: Release date for Privoxy 3.0.23 stable
Fabian Keil
2015-01-16 10:11:32 UTC
Permalink
Apparently this mail was silently dropped ...

Fabian

Begin forwarded message:

Date: Thu, 15 Jan 2015 12:21:39 +0100
From: Fabian Keil <***@fabiankeil.de>
To: Privoxy developers <ijbswa-***@lists.sourceforge.net>
Subject: Release date for Privoxy 3.0.23 stable
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
A security issue (DoS) has recently been reported to me in private
and a couple of additional bugs (not all of them security issues)
were found while fuzzing with afl [0].
As usual, please contact me directly if you wish to test the
patches before they are committed.
I'm going to give a fuzzing-related talk at FOSDEM [1] were I'll
probably mention at least some of the issues, thus I'd like the
latest Privoxy release to be fixed by then (2015-01-31).
Any objections?
Fabian
[0] http://lcamtuf.coredump.cx/afl/
[1] https://fosdem.org/2015/schedule/event/fuzzing_freebsd/
Ian Silvester
2015-01-16 12:16:45 UTC
Permalink
No objections here.

----- Reply message -----
From: "Fabian Keil" <***@fabiankeil.de>
To: "Privoxy developers" <ijbswa-***@lists.sourceforge.net>
Subject: [privoxy-devel] Fw: Release date for Privoxy 3.0.23 stable
Date: Fri, Jan 16, 2015 5:11 AM

Apparently this mail was silently dropped ...

Fabian

Begin forwarded message:

Date: Thu, 15 Jan 2015 12:21:39 +0100
From: Fabian Keil <***@fabiankeil.de>
To: Privoxy developers <ijbswa-***@lists.sourceforge.net>
Subject: Release date for Privoxy 3.0.23 stable
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
A security issue (DoS) has recently been reported to me in private
and a couple of additional bugs (not all of them security issues)
were found while fuzzing with afl [0].
As usual, please contact me directly if you wish to test the
patches before they are committed.
I'm going to give a fuzzing-related talk at FOSDEM [1] were I'll
probably mention at least some of the issues, thus I'd like the
latest Privoxy release to be fixed by then (2015-01-31).
Any objections?
Fabian
[0] http://lcamtuf.coredump.cx/afl/
[1] https://fosdem.org/2015/schedule/event/fuzzing_freebsd/
Lee
2015-01-23 14:08:43 UTC
Permalink
I don't have much going on this weekend, so yes, a new release this
weekend sounds good.
Lee
Post by Fabian Keil
Apparently this mail was silently dropped ...
Fabian
Date: Thu, 15 Jan 2015 12:21:39 +0100
Subject: Release date for Privoxy 3.0.23 stable
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
A security issue (DoS) has recently been reported to me in private
and a couple of additional bugs (not all of them security issues)
were found while fuzzing with afl [0].
As usual, please contact me directly if you wish to test the
patches before they are committed.
I'm going to give a fuzzing-related talk at FOSDEM [1] were I'll
probably mention at least some of the issues, thus I'd like the
latest Privoxy release to be fixed by then (2015-01-31).
Any objections?
Fabian
[0] http://lcamtuf.coredump.cx/afl/
[1] https://fosdem.org/2015/schedule/event/fuzzing_freebsd/
Fabian Keil
2015-01-24 17:33:33 UTC
Permalink
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.

I uploaded a potential release tarball to:
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc

Please let me know if there are any issues.

Fabian
Lee
2015-01-24 18:39:52 UTC
Permalink
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
Looking good so far:
2015-01-24 13:35:07: Gathering regression tests from 4 action file(s)
delivered by Privoxy 3.0.23.
<.. snip ..>
2015-01-24 13:35:59: Executed 638 regression tests. Skipped 77. 638
successes, 0 failures.

Lee
Ian Silvester
2015-01-25 14:20:55 UTC
Permalink
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS? I ask because, having cvs updated 'current', I get
the following errors when I build from there:

autoheader: `configure.ac' or `configure.in' is required
autoconf: no input file
configure: error: cannot find sources (jcc.c) in . or ..

Notice: building the privoxy software.
rm -f a.out actions.o cgi.o cgiedit.o cgisimple.o deanimate.o encode.o
errlog.o filters.o gateway.o jbsockets.o jcc.o list.o loadcfg.o
loaders.o miscutil.o parsers.o ssplit.o urlmatch.o pcrs.o privoxy
default.action \
config.base config.tmp \
`find . \( -name TAGS -o -name tags \) -and -not -path
"./.git/refs/tags"` \
`find . -name "*.orig" -a -not -name rc.privoxy.orig`
make: *** No rule to make target `actions.c', needed by `actions.o'. Stop.


Is this an issue or should I disregard the CVS project for now and use
only your source tarball?

Ian
Post by Fabian Keil
Fabian
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Lee
2015-01-25 14:55:12 UTC
Permalink
Post by Ian Silvester
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS?
I would suggest not using CVS for anything but interim/test releases.

If you build from the source tarball anyone can duplicate your work.
If you build from CVS... who knows what's in your CVS tree.

Lee
Post by Ian Silvester
I ask because, having cvs updated 'current', I get
autoheader: `configure.ac' or `configure.in' is required
autoconf: no input file
configure: error: cannot find sources (jcc.c) in . or ..
Notice: building the privoxy software.
rm -f a.out actions.o cgi.o cgiedit.o cgisimple.o deanimate.o encode.o
errlog.o filters.o gateway.o jbsockets.o jcc.o list.o loadcfg.o
loaders.o miscutil.o parsers.o ssplit.o urlmatch.o pcrs.o privoxy
default.action \
config.base config.tmp \
`find . \( -name TAGS -o -name tags \) -and -not -path
"./.git/refs/tags"` \
`find . -name "*.orig" -a -not -name rc.privoxy.orig`
make: *** No rule to make target `actions.c', needed by `actions.o'. Stop.
Is this an issue or should I disregard the CVS project for now and use
only your source tarball?
Ian
Post by Fabian Keil
Fabian
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Ian Silvester
2015-01-25 15:20:14 UTC
Permalink
Post by Lee
Post by Ian Silvester
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS?
I would suggest not using CVS for anything but interim/test releases.
I use a clean checkout for releases using the tag since that is the
jumping-off point for all distributed files, source or otherwise, but
until the repository's tagged I can't do that ;o)
Post by Lee
If you build from the source tarball anyone can duplicate your work.
If you build from CVS... who knows what's in your CVS tree.
I do, so it's up to me to make sure it's clean I guess :o/ I'll try a
build from the tarball later on (family commitments for a few hours now)
and see if I get a different result.

Fabian if you've a comment I'd be interested; if memory serves there
were some changes to the build files for this release? Or was that for
3.0.22?


Cheers,

Ian
Post by Lee
Lee
Post by Ian Silvester
I ask because, having cvs updated 'current', I get
autoheader: `configure.ac' or `configure.in' is required
autoconf: no input file
configure: error: cannot find sources (jcc.c) in . or ..
Notice: building the privoxy software.
rm -f a.out actions.o cgi.o cgiedit.o cgisimple.o deanimate.o encode.o
errlog.o filters.o gateway.o jbsockets.o jcc.o list.o loadcfg.o
loaders.o miscutil.o parsers.o ssplit.o urlmatch.o pcrs.o privoxy
default.action \
config.base config.tmp \
`find . \( -name TAGS -o -name tags \) -and -not -path
"./.git/refs/tags"` \
`find . -name "*.orig" -a -not -name rc.privoxy.orig`
make: *** No rule to make target `actions.c', needed by `actions.o'. Stop.
Is this an issue or should I disregard the CVS project for now and use
only your source tarball?
Ian
Post by Fabian Keil
Fabian
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Fabian Keil
2015-01-25 17:12:06 UTC
Permalink
Post by Ian Silvester
Post by Lee
Post by Ian Silvester
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS?
It mainly suggests that Sourceforge did not let me upload it to the
project page ...
Post by Ian Silvester
Post by Lee
I would suggest not using CVS for anything but interim/test releases.
I use a clean checkout for releases using the tag since that is the
jumping-off point for all distributed files, source or otherwise, but
until the repository's tagged I can't do that ;o)
It's tagged now, so you could try again.

If you previously checked out HEAD (currently the same as v_3_0_23) you
should already have gotten the latest configure.in, though, and it's not
obvious to me why your checkout did not contain the file at all.
Post by Ian Silvester
Post by Lee
If you build from the source tarball anyone can duplicate your work.
If you build from CVS... who knows what's in your CVS tree.
I do, so it's up to me to make sure it's clean I guess :o/ I'll try a
build from the tarball later on (family commitments for a few hours now)
and see if I get a different result.
Fabian if you've a comment I'd be interested; if memory serves there
were some changes to the build files for this release? Or was that for
3.0.22?
I'm not sure what changes you are referring to. I consider configure.in
a build file, but it has to be changed for every release to bump the version.

Fabian
Ian Silvester
2015-01-25 21:24:42 UTC
Permalink
Okay am building fine now that the repository is tagged - will kill my
HEAD checkout and recreated it from scratch.

Unfortunately now regression testing is failing and I am mystified as to
why:

<snip>$ ./privoxy-regression-test.pl
2015-01-25 16:09:12: Asking Privoxy for the number of action files
available ...
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Oh noes. Running curl failed 1 times in a row. Last
error: 'curl: (6) Couldn't resolve host 'p.p''. Fatal error. Exiting.

Privoxy 3.0.23 is installed and running, and browsing to http://p.p/
returns the status page as per usual. Using curl to some other address
at the command prompt works and is being routed through Privoxy as one
would expect (the number of requests increments on
http://config.privoxy.org/show-status) but for some reason Privoxy is
not intercepting p.p when it is curled from the commandline. My hosts
file has no entries for p.p by the way.

Any suggestions gratefully received,

Ian
Post by Fabian Keil
Post by Ian Silvester
Post by Lee
Post by Ian Silvester
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS?
It mainly suggests that Sourceforge did not let me upload it to the
project page ...
Post by Ian Silvester
Post by Lee
I would suggest not using CVS for anything but interim/test releases.
I use a clean checkout for releases using the tag since that is the
jumping-off point for all distributed files, source or otherwise, but
until the repository's tagged I can't do that ;o)
It's tagged now, so you could try again.
If you previously checked out HEAD (currently the same as v_3_0_23) you
should already have gotten the latest configure.in, though, and it's not
obvious to me why your checkout did not contain the file at all.
Post by Ian Silvester
Post by Lee
If you build from the source tarball anyone can duplicate your work.
If you build from CVS... who knows what's in your CVS tree.
I do, so it's up to me to make sure it's clean I guess :o/ I'll try a
build from the tarball later on (family commitments for a few hours now)
and see if I get a different result.
Fabian if you've a comment I'd be interested; if memory serves there
were some changes to the build files for this release? Or was that for
3.0.22?
I'm not sure what changes you are referring to. I consider configure.in
a build file, but it has to be changed for every release to bump the version.
Fabian
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Ian Silvester
2015-01-26 02:06:03 UTC
Permalink
OS X builds for all three targets now uploaded to Sourceforge.

If anyone has any suggestions as to why I'm unable to run the regression
tests (see my previous mail) I'm keen to hear them.

Cheers,

Ian
Post by Ian Silvester
Okay am building fine now that the repository is tagged - will kill my
HEAD checkout and recreated it from scratch.
Unfortunately now regression testing is failing and I am mystified as
<snip>$ ./privoxy-regression-test.pl
2015-01-25 16:09:12: Asking Privoxy for the number of action files
available ...
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Oh noes. Running curl failed 1 times in a row.
Last error: 'curl: (6) Couldn't resolve host 'p.p''. Fatal error. Exiting.
Privoxy 3.0.23 is installed and running, and browsing to http://p.p/
returns the status page as per usual. Using curl to some other address
at the command prompt works and is being routed through Privoxy as one
would expect (the number of requests increments on
http://config.privoxy.org/show-status) but for some reason Privoxy is
not intercepting p.p when it is curled from the commandline. My hosts
file has no entries for p.p by the way.
Any suggestions gratefully received,
Ian
Post by Fabian Keil
Post by Ian Silvester
Post by Lee
Post by Ian Silvester
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS?
It mainly suggests that Sourceforge did not let me upload it to the
project page ...
Post by Ian Silvester
Post by Lee
I would suggest not using CVS for anything but interim/test releases.
I use a clean checkout for releases using the tag since that is the
jumping-off point for all distributed files, source or otherwise, but
until the repository's tagged I can't do that ;o)
It's tagged now, so you could try again.
If you previously checked out HEAD (currently the same as v_3_0_23) you
should already have gotten the latest configure.in, though, and it's not
obvious to me why your checkout did not contain the file at all.
Post by Ian Silvester
Post by Lee
If you build from the source tarball anyone can duplicate your work.
If you build from CVS... who knows what's in your CVS tree.
I do, so it's up to me to make sure it's clean I guess :o/ I'll try a
build from the tarball later on (family commitments for a few hours now)
and see if I get a different result.
Fabian if you've a comment I'd be interested; if memory serves there
were some changes to the build files for this release? Or was that for
3.0.22?
I'm not sure what changes you are referring to. I consider configure.in
a build file, but it has to be changed for every release to bump the version.
Fabian
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Fabian Keil
2015-01-26 11:09:29 UTC
Permalink
Post by Ian Silvester
OS X builds for all three targets now uploaded to Sourceforge.
Great. The website has been updated and I'm going to sent out
the announcement in a couple of minutes.

Fabian
Lee
2015-01-26 03:31:56 UTC
Permalink
Post by Ian Silvester
Okay am building fine now that the repository is tagged - will kill my
HEAD checkout and recreated it from scratch.
Unfortunately now regression testing is failing and I am mystified as to
<snip>$ ./privoxy-regression-test.pl
2015-01-25 16:09:12: Asking Privoxy for the number of action files
available ...
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Oh noes. Running curl failed 1 times in a row. Last
error: 'curl: (6) Couldn't resolve host 'p.p''. Fatal error. Exiting.
Privoxy 3.0.23 is installed and running, and browsing to http://p.p/
returns the status page as per usual. Using curl to some other address
at the command prompt works and is being routed through Privoxy as one
would expect (the number of requests increments on
http://config.privoxy.org/show-status) but for some reason Privoxy is
not intercepting p.p when it is curled from the commandline. My hosts
file has no entries for p.p by the way.
Any suggestions gratefully received,
You need to specify --privoxy-address maybe??
./privoxy-regression-test.pl --privoxy-address 127.0.0.1:8118

Lee
Post by Ian Silvester
Ian
Post by Fabian Keil
Post by Ian Silvester
Post by Lee
Post by Ian Silvester
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS?
It mainly suggests that Sourceforge did not let me upload it to the
project page ...
Post by Ian Silvester
Post by Lee
I would suggest not using CVS for anything but interim/test releases.
I use a clean checkout for releases using the tag since that is the
jumping-off point for all distributed files, source or otherwise, but
until the repository's tagged I can't do that ;o)
It's tagged now, so you could try again.
If you previously checked out HEAD (currently the same as v_3_0_23) you
should already have gotten the latest configure.in, though, and it's not
obvious to me why your checkout did not contain the file at all.
Post by Ian Silvester
Post by Lee
If you build from the source tarball anyone can duplicate your work.
If you build from CVS... who knows what's in your CVS tree.
I do, so it's up to me to make sure it's clean I guess :o/ I'll try a
build from the tarball later on (family commitments for a few hours now)
and see if I get a different result.
Fabian if you've a comment I'd be interested; if memory serves there
were some changes to the build files for this release? Or was that for
3.0.22?
I'm not sure what changes you are referring to. I consider configure.in
a build file, but it has to be changed for every release to bump the version.
Fabian
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Ian Silvester
2015-01-26 04:00:23 UTC
Permalink
Post by Lee
Post by Ian Silvester
Okay am building fine now that the repository is tagged - will kill my
HEAD checkout and recreated it from scratch.
Unfortunately now regression testing is failing and I am mystified as to
<snip>$ ./privoxy-regression-test.pl
2015-01-25 16:09:12: Asking Privoxy for the number of action files
available ...
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Oh noes. Running curl failed 1 times in a row. Last
error: 'curl: (6) Couldn't resolve host 'p.p''. Fatal error. Exiting.
Privoxy 3.0.23 is installed and running, and browsing to http://p.p/
returns the status page as per usual. Using curl to some other address
at the command prompt works and is being routed through Privoxy as one
would expect (the number of requests increments on
http://config.privoxy.org/show-status) but for some reason Privoxy is
not intercepting p.p when it is curled from the commandline. My hosts
file has no entries for p.p by the way.
Any suggestions gratefully received,
You need to specify --privoxy-address maybe??
./privoxy-regression-test.pl --privoxy-address 127.0.0.1:8118
Thanks Lee - that did it!

Surprised me since the listen address is quoted in the source code and
the stated error was to do with Privoxy's config intercept address but
whatever, it did the trick!

So then, I have one failure:

$ ./privoxy-regression-test.pl --privoxy-address 127.0.0.1:8118
2015-01-25 22:48:57: Asking Privoxy for the number of action files
available ...
2015-01-25 22:48:57: Gathering regression tests from 4 action file(s)
delivered by Privoxy 3.0.23.
2015-01-25 22:48:57: Executing regression tests ...
2015-01-25 22:49:04: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
2015-01-25 22:49:04: Failure for test 515. Header 'Keep-Alive: 20' and
tag 'Pass Keep-Alive header'
2015-01-25 22:49:06: Executed 664 regression tests. Skipped 42. 663
successes, 1 failures.


Fabian - care to comment?

Ian
Post by Lee
Lee
Post by Ian Silvester
Ian
Post by Fabian Keil
Post by Ian Silvester
Post by Lee
Post by Ian Silvester
Post by Fabian Keil
I'd like to release Privoxy 3.0.23 stable around 2015-01-24
as a bug fix release.
Thanks for the feedback.
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz
http://www.privoxy.org/downloads/privoxy-3.0.23-stable-src.tar.gz.asc
Please let me know if there are any issues.
'Scuse ignorance but does your providing a source tarball suggest that I
can't build from CVS?
It mainly suggests that Sourceforge did not let me upload it to the
project page ...
Post by Ian Silvester
Post by Lee
I would suggest not using CVS for anything but interim/test releases.
I use a clean checkout for releases using the tag since that is the
jumping-off point for all distributed files, source or otherwise, but
until the repository's tagged I can't do that ;o)
It's tagged now, so you could try again.
If you previously checked out HEAD (currently the same as v_3_0_23) you
should already have gotten the latest configure.in, though, and it's not
obvious to me why your checkout did not contain the file at all.
Post by Ian Silvester
Post by Lee
If you build from the source tarball anyone can duplicate your work.
If you build from CVS... who knows what's in your CVS tree.
I do, so it's up to me to make sure it's clean I guess :o/ I'll try a
build from the tarball later on (family commitments for a few hours now)
and see if I get a different result.
Fabian if you've a comment I'd be interested; if memory serves there
were some changes to the build files for this release? Or was that for
3.0.22?
I'm not sure what changes you are referring to. I consider configure.in
a build file, but it has to be changed for every release to bump the version.
Fabian
------------------------------------------------------------------------------
New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
GigeNET is offering a free month of service with a new server in Ashburn.
Choose from 2 high performing configs, both with 100TB of bandwidth.
Higher redundancy.Lower latency.Increased capacity.Completely compliant.
http://p.sf.net/sfu/gigenet
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Fabian Keil
2015-01-26 11:06:19 UTC
Permalink
Post by Ian Silvester
Post by Lee
Post by Ian Silvester
Okay am building fine now that the repository is tagged - will kill my
HEAD checkout and recreated it from scratch.
Unfortunately now regression testing is failing and I am mystified as to
<snip>$ ./privoxy-regression-test.pl
2015-01-25 16:09:12: Asking Privoxy for the number of action files
available ...
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Oh noes. Running curl failed 1 times in a row. Last
error: 'curl: (6) Couldn't resolve host 'p.p''. Fatal error. Exiting.
Privoxy 3.0.23 is installed and running, and browsing to http://p.p/
returns the status page as per usual. Using curl to some other address
at the command prompt works and is being routed through Privoxy as one
would expect (the number of requests increments on
http://config.privoxy.org/show-status) but for some reason Privoxy is
not intercepting p.p when it is curled from the commandline. My hosts
file has no entries for p.p by the way.
Any suggestions gratefully received,
You need to specify --privoxy-address maybe??
./privoxy-regression-test.pl --privoxy-address 127.0.0.1:8118
Thanks Lee - that did it!
Surprised me since the listen address is quoted in the source code and
the stated error was to do with Privoxy's config intercept address but
whatever, it did the trick!
As an alternative you can set the http_proxy environment variable
which is also used by other http clients. If Privoxy's address isn't
specified at all, Privoxy has no chance to intercept the request
(unless a packet filter is redirecting it).
[...]
Post by Ian Silvester
2015-01-25 22:49:04: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
2015-01-25 22:49:04: Failure for test 515. Header 'Keep-Alive: 20' and
tag 'Pass Keep-Alive header'
[...]
Post by Ian Silvester
Fabian - care to comment?
It's probably a false positive. What's the keep-alive-timeout specified
in the config file?

Fabian
Ian Silvester
2015-01-26 12:04:32 UTC
Permalink
Post by Fabian Keil
Post by Ian Silvester
Post by Lee
Post by Ian Silvester
Okay am building fine now that the repository is tagged - will kill my
HEAD checkout and recreated it from scratch.
Unfortunately now regression testing is failing and I am mystified as to
<snip>$ ./privoxy-regression-test.pl
2015-01-25 16:09:12: Asking Privoxy for the number of action files
available ...
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Fetch failure: 'curl: (6) Couldn't resolve host 'p.p''
2015-01-25 16:09:13: Oh noes. Running curl failed 1 times in a row. Last
error: 'curl: (6) Couldn't resolve host 'p.p''. Fatal error. Exiting.
Privoxy 3.0.23 is installed and running, and browsing to http://p.p/
returns the status page as per usual. Using curl to some other address
at the command prompt works and is being routed through Privoxy as one
would expect (the number of requests increments on
http://config.privoxy.org/show-status) but for some reason Privoxy is
not intercepting p.p when it is curled from the commandline. My hosts
file has no entries for p.p by the way.
Any suggestions gratefully received,
You need to specify --privoxy-address maybe??
./privoxy-regression-test.pl --privoxy-address 127.0.0.1:8118
Thanks Lee - that did it!
Surprised me since the listen address is quoted in the source code and
the stated error was to do with Privoxy's config intercept address but
whatever, it did the trick!
As an alternative you can set the http_proxy environment variable
which is also used by other http clients. If Privoxy's address isn't
specified at all, Privoxy has no chance to intercept the request
(unless a packet filter is redirecting it).
I'll do that.
Post by Fabian Keil
[...]
Post by Ian Silvester
2015-01-25 22:49:04: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
2015-01-25 22:49:04: Failure for test 515. Header 'Keep-Alive: 20' and
tag 'Pass Keep-Alive header'
[...]
Post by Ian Silvester
Fabian - care to comment?
It's probably a false positive. What's the keep-alive-timeout specified
in the config file?
The default for the distribution:

keep-alive-timeout 5


Ian
Post by Fabian Keil
Fabian
------------------------------------------------------------------------------
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Fabian Keil
2015-01-26 16:39:57 UTC
Permalink
Post by Ian Silvester
Post by Fabian Keil
[...]
Post by Ian Silvester
2015-01-25 22:49:04: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
2015-01-25 22:49:04: Failure for test 515. Header 'Keep-Alive: 20' and
tag 'Pass Keep-Alive header'
[...]
Post by Ian Silvester
Fabian - care to comment?
It's probably a false positive. What's the keep-alive-timeout specified
in the config file?
keep-alive-timeout 5
Are you using the latest regression-test.action?

Looking at your test output again, the number of tests available
(664 + 42 = 706) seems to suggest otherwise.

If you are missing the following commit, it would explain the error:

***@r500 ~/git/privoxy $git show 8531411
commit 8531411aea9e4b490c376ecff79b4c2af370b3cb
Author: fabiankeil <fabiankeil>
Date: Tue Nov 18 14:21:56 2014 +0000

Skip test level 13 unless the keep-alive-timeout is above 20

The tests are expected to fail with lower values, for example
the 5 seconds specified in the default config file.

Reported by Lee.

diff --git a/regression-tests.action b/regression-tests.action
index 33c2936..6cb9eb1 100644
--- a/regression-tests.action
+++ b/regression-tests.action
@@ -1,5 +1,5 @@
#############################################################################
-# $Id: regression-tests.action,v 1.56 2014/11/14 12:59:11 fabiankeil Exp $
+# $Id: regression-tests.action,v 1.57 2014/11/18 14:21:56 fabiankeil Exp $
#############################################################################
#
# This is a configuration file for Privoxy-Regression-Test
@@ -34,7 +34,7 @@ for-privoxy-version=3.0.22
# Level 9 needs = config line user-manual\s+(\.?\.?/|[A-Za-z]:)
# Level 12 needs = config line enable-edit-actions\s+1
# Level 13 needs = feature status FEATURE_CONNECTION_KEEP_ALIVE Yes
-# Level 13 needs = config line keep-alive-timeout\s+[1-9]+
+# Level 13 needs = config line keep-alive-timeout\s+(2[1-9]+|[3-9]\d+|1\d{2,})
# Level 14 needs = feature status FEATURE_CONNECTION_KEEP_ALIVE No
# Level 15 needs = config line connection-sharing\s+0
# Level 16 needs = feature status FEATURE_CONNECTION_KEEP_ALIVE Yes

Fabian
Ian Silvester
2015-01-26 17:37:39 UTC
Permalink
Post by Fabian Keil
Post by Ian Silvester
Post by Fabian Keil
[...]
Post by Ian Silvester
2015-01-25 22:49:04: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
2015-01-25 22:49:04: Failure for test 515. Header 'Keep-Alive: 20' and
tag 'Pass Keep-Alive header'
[...]
Post by Ian Silvester
Fabian - care to comment?
It's probably a false positive. What's the keep-alive-timeout specified
in the config file?
keep-alive-timeout 5
Are you using the latest regression-test.action?
Looking at your test output again, the number of tests available
(664 + 42 = 706) seems to suggest otherwise.
commit 8531411aea9e4b490c376ecff79b4c2af370b3cb
Author: fabiankeil <fabiankeil>
Date: Tue Nov 18 14:21:56 2014 +0000
Skip test level 13 unless the keep-alive-timeout is above 20
The tests are expected to fail with lower values, for example
the 5 seconds specified in the default config file.
Reported by Lee.
diff --git a/regression-tests.action b/regression-tests.action
index 33c2936..6cb9eb1 100644
--- a/regression-tests.action
+++ b/regression-tests.action
@@ -1,5 +1,5 @@
#############################################################################
-# $Id: regression-tests.action,v 1.56 2014/11/14 12:59:11 fabiankeil Exp $
+# $Id: regression-tests.action,v 1.57 2014/11/18 14:21:56 fabiankeil Exp $
#############################################################################
#
# This is a configuration file for Privoxy-Regression-Test
@@ -34,7 +34,7 @@ for-privoxy-version=3.0.22
# Level 9 needs = config line user-manual\s+(\.?\.?/|[A-Za-z]:)
# Level 12 needs = config line enable-edit-actions\s+1
# Level 13 needs = feature status FEATURE_CONNECTION_KEEP_ALIVE Yes
-# Level 13 needs = config line keep-alive-timeout\s+[1-9]+
+# Level 13 needs = config line keep-alive-timeout\s+(2[1-9]+|[3-9]\d+|1\d{2,})
# Level 14 needs = feature status FEATURE_CONNECTION_KEEP_ALIVE No
# Level 15 needs = config line connection-sharing\s+0
# Level 16 needs = feature status FEATURE_CONNECTION_KEEP_ALIVE Yes
Fabian
Bingo! Latest file in place, no errors at all now:

./privoxy-regression-test.pl --privoxy-address 127.0.0.1:8118
2015-01-26 12:36:34: Asking Privoxy for the number of action files
available ...
2015-01-26 12:36:34: Gathering regression tests from 4 action file(s)
delivered by Privoxy 3.0.23.
2015-01-26 12:36:34: Executing regression tests ...
2015-01-26 12:36:44: Executed 663 regression tests. Skipped 52. 663
successes, 0 failures.

Ian
Post by Fabian Keil
------------------------------------------------------------------------------
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Ijbswa-developers mailing list
https://lists.sourceforge.net/lists/listinfo/ijbswa-developers
Fabian Keil
2015-01-27 17:17:28 UTC
Permalink
Post by Fabian Keil
Post by Ian Silvester
Post by Fabian Keil
[...]
Post by Ian Silvester
2015-01-25 22:49:04: Ooops. Got: 'REMOVAL' while expecting: 'NO CHANGE'
2015-01-25 22:49:04: Failure for test 515. Header 'Keep-Alive: 20' and
tag 'Pass Keep-Alive header'
[...]
Post by Ian Silvester
Fabian - care to comment?
It's probably a false positive. What's the keep-alive-timeout specified
in the config file?
keep-alive-timeout 5
Are you using the latest regression-test.action?
Looking at your test output again, the number of tests available
(664 + 42 = 706) seems to suggest otherwise.
Great.

Fabian

Loading...