Henrik Nordstrom has voted approve.
Status is now: Semi-approved
Comment:
For details, see:
http://squid-cache.org/bundlebuggy//request/%3C20080314234236.GA3185%40motherbox.xtech.com.ar%3E
Henrik Nordstrom has voted approve.
Status is now: Semi-approved
For details, see:
http://squid-cache.org/bundlebuggy//request/%3C20080309114825.BBA36E9FF1%40treenet.co.nz%3E
Henrik Nordstrom has voted abstain.
Status is now: Semi-approved
Comment:
Not familiar with the fine details about templates to say if this is
good or bad.. but if it doesn't break other platforms I am fine with it.
For details, see:
http://squid-cache.org/bundlebuggy//request
On Fri, 2008-03-14 at 12:34 +0900, Adrian Chadd wrote:
Has anyone actually committed to the bzr tree? I haven't seen any commit
messages.
Commit works fine, but commit messages is not yet operational. Waiting
for Robert to set up a mailer cron job..
Regards
Henrik
On Fri, 2008-03-14 at 17:28 +1300, Amos Jeffries wrote:
On a side note. Henrik, are you able to get the maintainers .update
scripts pulling the changesets from bzr?
I think we will do something else for bzr. bzr is already tree oriented
and the bzr browser provides changeset patches out of
Have finally been making some significant progress on handing of large
response headers in Squid-3. After some consideration the approach taken
is to simply seek over the response headers in the data stream clone the
response from the store instead of parsing it again. This is similar to
what
On Mon, 2008-03-10 at 14:12 +1300, Amos Jeffries wrote:
As it stands we can call one or the other an up/down and leave things as-is.
I'm minded to call 2.6 a 'down' of 3.0 and 3.0/2.7/2.6 downs of 3.1. I
have not looked closely at the update script to see if thats right though.
As I said I
On Tue, 2008-03-11 at 22:43 +0100, Guido Serassio wrote:
This file comes from the original work of Romeo Anghelache.
After some search, I have found the original one from Apache 1.3:
http://svn.apache.org/viewvc/httpd/httpd/branches/1.3.x/src/os/win32/readdir.c?view=markup
If I remember
On Thu, 2008-03-13 at 20:48 +0100, Guido Serassio wrote:
Too much times something like this is happened:
- Update from CVS of my work dir
- Fix of build problems
- Commit of fixes
- Finished the little time that I have available for development,
usually during weekend
- Hope to do some
On Mon, 2008-03-10 at 23:19 +1300, Amos Jeffries wrote:
Q: who shall do the announce and to where?
I know of squid-dev, squid-users, noc, anywhere else?
squid-users@ and squid-announce@
Regards
Henrik
On Sun, 2008-03-09 at 21:28 +1300, Amos Jeffries wrote:
1) I don't have write access to the FTP area to copy the files in.
I think its /server/ftp/pub/squid-3/STABLE/ anyway.
Fixed.
(The STABLE1 signature files and diff appear to be missing too)
I have now copied the STABLE1
On Sat, 2008-03-01 at 10:57 +0100, Guido Serassio wrote:
This is very critical on the side of the DOS/Unix
text format: Visual Studio doesn't work with Unix text files.
Usually I commit the files on this directory only from Windows machines.
Thats easy to deal with, in fact most likely not
On Fri, 2008-02-29 at 01:13 +1300, Amos Jeffries wrote:
Hm, yes. Henrik, is there a report somewhere that lgroups the merged
changesets the same way the maintainer merge.html lists the un-merged
and non-merged ones?
There is a merge.html in 2.7 showing what's unique for 2.7 and not merged to
On Sun, 2008-03-09 at 11:23 -0600, Alex Rousskov wrote:
IMO, virtually all Squid sources should include (directly or indirectly)
squid.h as the first include file. In my experience, this design is very
useful for portability and other hacks.
We have two such files
include/config.h project
Henrik Nordstrom has voted approve.
Status is now: Approved
Comment:
Good start.
For details, see:
http://squid-cache.org/bundlebuggy//request/%3C1198974291.6327.19.camel%40lifeless-64%3E
On Tue, 2008-03-04 at 20:56 +0100, Guido Serassio wrote:
On proprietary Unix a binary package for Python 2.4 or better is not
always simple to find:
python.org distributes binaries for Windows (32-bit, 64-bit and ia64)
and MacOS (x86 ppc).
ActivePython should also work. Available for
On Mon, 2008-03-03 at 18:45 +1300, Amos Jeffries wrote:
On the wiki you still have:
* Migrate in progress development branches
hno: I wote no on this. It's up to respective sub-project to merge over if
they like.
Is there going to be a merge script/method available to pull
silly test of bzr send, ignore.
# Bazaar merge directive format 2 (Bazaar 0.90)
# revision_id: [EMAIL PROTECTED]
# 3ujtv9ey18c1yiiw
# target_branch: file:///data/bzr/squid3/trunk/
# testament_sha1: 7178857878c79f284b96e3357c4509f1cc2e2e39
# timestamp: 2008-03-02 19:47:01 +0100
#
On Thu, 2008-02-28 at 12:41 +0900, Adrian Chadd wrote:
I'd suggest just disabling COSS in -3 until someone ports over the changes
or replaces it with something better.
This was even decided on for the 3.0 release, but never done as
Squid-3.0 didn't for from HEAD until very late in the release
On Thu, 2008-02-28 at 01:04 +1300, Amos Jeffries wrote:
Just a few issues remaining before 3.0.STABLE-2.
* patch 11467 back-porting.
cd English
ls ERR_* ../list
* Close more bugs?
http://www.squid-cache.org/bugs/show_bug.cgi?id=2127
On Thu, 2008-02-28 at 01:04 +1300, Amos Jeffries wrote:
Just a few issues remaining before 3.0.STABLE-2.
* patch 11467 back-porting.
cd English
ls ERR_* ../list
* Close more bugs?
http://www.squid-cache.org/bugs/show_bug.cgi?id=2127
Is closed and verified by the reporter.
fixed
On sön, 2008-01-20 at 18:58 +0100, Guido Serassio wrote:
Hi,
Im resending again this message, because nobody answered my first post.
I cannot fix this by myself because I don't have the needed write
permission and I don't know the location of the script for the daily
build of the
On tis, 2008-01-22 at 00:04 +, ian j hart wrote:
ref: http://www.mail-archive.com/squid-users%40squid-cache.org/msg52048.html
Can someone please point me to the code which expands the URL.
Only a sinlge %s is supported. Nothing else.
from errorpage.c/cc:
char *quoted_url =
On tor, 2008-01-17 at 11:34 +0900, Adrian Chadd wrote:
I think so. And caches won't revalidate it for a day.
Henrik?
Are we using mod_expires on the web site perhaps?
Regards
Henrik
To check for memory leaks please compile with valgrind support
(configure option). Then after triggering the leak while running on
valgrind, go to the memory usage page in cachemgr (squidclient mgr:mem).
Valgrind summary is reported in cachemgr, leaks on stderr.
Regards
Henrik
On lör,
On lör, 2008-01-05 at 21:25 +, chtsanti wrote:
When a socket is closed then it is possible that remains AsyncCalls to be
executed which will try to perform operation on the closed socket.
This patch does not close the socket when comm_close called but instead
schedules the socket closing
On ons, 2008-01-02 at 16:52 +1100, Mark Nottingham wrote:
Slightly different;
gcc -DHAVE_CONFIG_H -DDEFAULT_CONFIG_FILE=\/opt/squid/squid.conf\ -
I. -I. -I../include -I. -I. -I../include -I../include
-Wall -g -O2 -c cf_gen.c
In file included from squid.h:421,
from
On ons, 2008-01-02 at 20:51 +0100, Guido Serassio wrote:
There is some problem here:
I'm no more able to build NCSA on Windows without OpenSSL.
or even with OpenSSL..
The attached patch fix the problem, but I'm not sure if' it's correct.
It's partially correct.
The correct patch is backing
On ons, 2008-01-02 at 21:26 +0100, Henrik Nordstrom wrote:
On ons, 2008-01-02 at 20:51 +0100, Guido Serassio wrote:
There is some problem here:
I'm no more able to build NCSA on Windows without OpenSSL.
or even with OpenSSL..
The attached patch fix the problem, but I'm not sure
On mån, 2007-12-24 at 15:01 +0100, Thomas-Martin Seck wrote:
I just checked the 20071224 tarball on FreeBSD 6.2. It trips over when
trying to link in MD5 stuff.
Can you try again with 2.HEAD? Did some changes to try address this on
the 27/12. (not yet merged over to 2.7)
Regards
Henrik
On sön, 2007-12-30 at 20:58 +0330, Hamid Choupani wrote:
At the end my source code is:
//buf is captured from ethereal software in last ntlm step:
char* buf =
TlRMTVNTUAADAAA
AGAAYAGIYABg
AegcABwBIAAA
ADQANAE8GAAY
AXACSAAA
helpers/negotiate_auth/squid_kerb_auth
squid_kerb_auth.c: In function ‘gethost_name’:
squid_kerb_auth.c:93: warning: implicit declaration of function ‘xgetaddrinfo’
squid_kerb_auth.c:95: warning: implicit declaration of function ‘xgai_strerror’
squid_kerb_auth.c:95: warning: format ‘%s’ expects
After the IPv6 merge cache.log is flooded with messages like this:
2007/12/22 00:01:13| storeLateRelease: released 0 objects
assert mtype != AF_INET6 at line 196
IPAddress invalid? with IsIPv4()=T, IsIPv6()=F
ADDRESS: 17f
2007/12/22 00:01:35| idnsRead: starting with FD 5
assert mtype !=
On ons, 2007-12-19 at 09:33 -0700, Alex Rousskov wrote:
What if Squid is not the only device doing CARP in a given environment?
It may be important to allow the administrator to configure Squid
hashing to match that of other devices so extending the hashing specs
may require an option to turn
On tor, 2007-12-20 at 21:43 +, Tony Dodd wrote:
hashing mechanism (out of interest, is there a reason p-name inherits
p-host if name= is not specified?).
Why shouldn't it? Most only have a sinlge proxy server per IP/hostname,
and find it easier to use that as reference instead of some
On tor, 2007-12-20 at 22:48 +1100, Robert Collins wrote:
All these meet the needs listed above. My strong preference is bzr; its
recently reached 1.0 and I'm extremely familiar with it due to having
spent some years on it. In a broad sense hg/monotone/bzr/git are very
similar, and darcs is
On tor, 2007-12-20 at 10:24 -0700, Alex Rousskov wrote:
On Tue, 2007-12-18 at 14:13 +0100, Henrik Nordstrom wrote:
On tis, 2007-12-18 at 23:32 +1300, Amos Jeffries wrote:
It really should not be for null.
The general syntax of cache_dir is
cache_dir type path option
path
Applied.
Regards
Henrik
On lör, 2007-12-15 at 12:56 +0200, Tsantilas Christos wrote:
Just a small patch to allow ICAP compile and run with the newly
added
IPV6 changes.
First tests looks good.
Regards,
Christos
signature.asc
Description: This is a digitally signed message part
Will be fixed in STABLE2.. (too late for STABLE1..)
Regards
Henrik
On lör, 2007-12-15 at 11:12 +0200, Tsantilas Christos wrote:
Hi all,
the http://www.squid-cache.org/Versions/v3/3.0/RELEASENOTES.html web
page still refers to Squid 3.0.RC1 pre-release not to Squid 3.0.STABLE1
release
On mån, 2007-12-10 at 17:26 +0900, Adrian Chadd wrote:
Why/where is the code assuming that offset 0 will be HDR_DATE and offset 1
will
be HDR_EXPIRES? The replacement code doesn't check to see whether anything is
at the position in the array that its inserting into..
It isn't assuming or
On tis, 2007-12-11 at 14:12 +0900, Adrian Chadd wrote:
Rightio. That makes sense. I was wondering if there were a reason why
the existing header, if any, wasn't simply replaced - and if there wasn't
a header, why it wasn't just appended.
Because it was easier to do it this way.
and should be
On sön, 2007-12-16 at 14:10 +1100, Reuben Farrelly wrote:
Still as of now failing to compile -HEAD for me:
clientInterpretRequestHeaders(ClientHttpRequest*)':
client_side_request.cc:733: error: 'no_addr' was not declared in this scope
client_side_request.cc:739: error: 'no_addr' was not
On fre, 2007-12-14 at 18:26 +1300, Amos Jeffries wrote:
He suggested a big(ger) delay than a few days to wait for bugs to
prevent 'inventing two patches for two branches'.
My opinion:
With SQUID_3_0 branched there is no reason to hold back stuff from
Squid-3 HEAD.
Usually bugfixes is quite
On fre, 2007-12-14 at 14:36 -0700, Alex Rousskov wrote:
My only regret is that I did not do the astyle check yet. If you can
check what the latest astyle does to Squid3, please do that. I think
having common automated format before the big commits would minimize
formatting conflicts.
Has
On lör, 2007-12-15 at 12:00 +1300, Amos Jeffries wrote:
Can everyone please hold off commits to squid3-HEAD for 36-48 hours. I
don't want it to erase any work you have done. Thanks.
Aren't you applying the changes using patch?
Applying changes by copy is dangerous. There is up to a day delay
On lör, 2007-12-15 at 14:54 +1300, Amos Jeffries wrote:
Yes I applied the code by 'patch -p1'.
Good. Then there is no risk of accidently backing out others edits.
I will need some help about adding the website /Versions/v3/3.1/ pages
and the release-3.1.sgml notes. What need to
On tor, 2007-12-13 at 12:38 +0900, Adrian Chadd wrote:
Hm, I've done a search for 2.6 and 2.HEAD bugs in bugzilla - there's 13
of them. Some could be merged in before 2.7 is released. I can't see
anything thats a blocker which isn't current behaviour in 2.6.
Good.
Regards
Henrik
On fre, 2007-12-14 at 14:57 +1300, Amos Jeffries wrote:
Now that 3.0 is branched. (is it complete?)
Yes. 3.0.STABLE1 is even published.
I'll be starting to import the new files for 3.1 in about 7-8 hours.
Leaving the core code change until any post-release shakedown period has
finished.
Any
Should we do the same in Squid-2?
Regards
Henrik
On tor, 2007-12-13 at 14:54 -0700, Duane Wessels wrote:
wessels 2007/12/13 14:54:57 MST
Modified files:(Branch: SQUID_3_0)
.README
Log:
Remove references to myself and NLANR, add pointer to
On sön, 2007-12-09 at 00:38 +0900, Adrian Chadd wrote:
Does anyone have anything else they'd like to add to squid-2.HEAD ?
I think its about time to release Squid-2.7 and shuffle that development
train along a bit.
My queue is empty, but before 2.7 gets branched we need to go over the
bugzilla
On sön, 2007-12-02 at 14:48 +0100, Rafael Martinez (Squid development)
wrote:
Please, asking for your advice before commiting a non trivial change...
is this the desired effect ? (Leaving IP indexing on SNMP agent...)
I am fine with dropping the IP based index in Squid-3 entirely if it
On tis, 2007-11-27 at 23:27 +1300, Amos Jeffries wrote:
Sigh. We need to clean up these MD5 issues before they propagate further.
Not sure how much issues Squid-2 is having with it at the moment.. I am
fine with leaving it as it is until people complain.
Has the renaming fix in Squid-3 shown
On sön, 2007-11-25 at 18:48 +0100, Rafael Martinez Torres wrote:
Yes , I knew. For the time beeing, I'm only trying to reach the point on
Squid
3 for the same funcionality on Squid 2, even with the wrong IP indexing
schema
Squid-2 has both kinds of indexes for peers, and the IP based is
On sön, 2007-11-25 at 22:13 +0900, Adrian Chadd wrote:
Here's my proposal for refresh pattern changes to squid-2.HEAD; perhaps
Squid-3 too down the track.
+refresh_pattern cgi-bin 0 0% 0
+refresh_pattern \? 0 0% 0
refresh_pattern ^ftp:
On mån, 2007-11-26 at 08:55 +0900, Adrian Chadd wrote:
On Mon, Nov 26, 2007, Henrik Nordstrom wrote:
As discussed earlier in private max-age should be the normal one.. and
additionally it doesn't apply to ftp or gopher.. (neither have query
urls), and a little comments to why things
On fre, 2007-11-23 at 23:55 +1300, Amos Jeffries wrote:
SNMPv2-SMI::enterprises.3495.1.5.1.1.1.192.168.0.12 = STRING:
rio.treenetnz.com
Error: OID not increasing:
This is inherent to our broken way of indexing the peers. Not unique to
ipv6 but also seen for ipv4 peers even before the ipv6
On fre, 2007-11-23 at 09:55 +0900, Adrian Chadd wrote:
I've been fiddling with Squid-2.HEAD and google maps, and I've noticed that
weird stuff happens.
Here's the log entries:
1195739570.596 SWAPOUT 00 00034F1B BEAF584D4381EE1DCAF38A162BAA0224 200
1195739570 1103259488 1227275570
On tor, 2007-11-22 at 15:51 +0900, Adrian Chadd wrote:
Could you make this stuff optional at all at runtime?
done
I'm just a bit wary about these sorts of temporary fixes going into the
codebase without fixing the actual root cause - we'll just end up creating
more work for ourselves in the
On ons, 2007-11-21 at 00:44 +0100, Rafael Martinez (Squid development)
wrote:
If snmp_core.cc revision 1.10.8.26
How can I come back into 1.10.8.24 ?
cvs update -j 1.10.8.26 -j 1.10.8.25 snmp_core.cc
cvs update -j 1.10.8.25 -j 1.10.8.24 snmp_core.cc
yes, or in one go if you
On tor, 2007-11-22 at 10:57 +1300, Amos Jeffries wrote:
Three cheers for Henrik! :-)
Any chance of a 3.1 port soon?
We need a bit of discussion on whats the best approach on how to solve
Bug #7 first.
I.e. should we do like I did for Squid-2, rewriting the entry, or should
we jump directly
On mån, 2007-11-19 at 01:04 +0900, Adrian Chadd wrote:
Argh! Just compile with --enable-ssl for now and I'll try to unravel what the
hell
has been done tomorrow.
Or with just --with-openssl if you want Squid to use MD5 from OpenSSL
but not enable SSL support.
Regards
Henrik
signature.asc
On mån, 2007-11-19 at 08:59 +0900, Adrian Chadd wrote:
Is there any particular reason why Squid just doesn't require openssl be
present during build for the MD5 support?
For one thing the OpenSSL license is not fully GPL compatible, so
distributors where OpenSSL is not part of the base
On tor, 2007-11-15 at 09:37 +0900, Adrian Chadd wrote:
G'day,
I'd like to propose a Squid modification - to cache dynamic content
thats playing good.
Yes, it's as simple as removing the cache lines from the default
suggested config and making sure your refresh_pattern rules is
reasonable.
On ons, 2007-11-14 at 18:20 -0700, Duane Wessels wrote:
While we're at it we could probably also get rid of the silly gopher
refresh_pattern line.
There is still gophers around, and it's a supported protocol. Without
that line gopher responses won't get cached as they do not have any
freshness
On tor, 2007-11-15 at 10:17 +0900, Adrian Chadd wrote:
On Wed, Nov 14, 2007, Duane Wessels wrote:
I'd like to see something default in the next Squid release, so we can
release it with a few interesting tag lines like Can cache google maps!
I can support removing '?' from the default
On tor, 2007-11-15 at 10:28 +0900, Adrian Chadd wrote:
Will what I've done above actually stop storing the data entirely, or will
it try revalidating it every request? Is there really a difference?
It will stop caching, at lest unless there is an ETag or Last-Modified.
Note: The RFC do not
On tor, 2007-11-15 at 11:49 +0900, Adrian Chadd wrote:
However, there is still the HTTP/1.0 MUST NOT cache requirement. Not
really an idea what that's about however.
If people -are- returning freshness info in a ? URL then its entirely
possible they've got clue, right? Or not?
Exactly.
On sön, 2007-11-11 at 09:17 -0800, Arthur Tumanyan wrote:
Hi.I have a question.
Please have a look at this part of configure.in script that I trying to
modify for my needs.
AM_CONDITIONAL(ENABLE_SHAGA, false)
AC_ARG_ENABLE(shaga,
[ --enable-shagaEnable ShagaEngine support],
On tis, 2007-11-06 at 13:28 -0700, Alex Rousskov wrote:
As you can see, a warning is printed but there are no consequences. That
is, Squid will still listen on the specified port although it probably
would not be able to do anything useful there without a valid SSL
context.
it's meant to
On tis, 2007-10-16 at 17:27 +1300, Amos Jeffries wrote:
The default for all accesses (HTTP, ICP, HTCP, SNMP) is deny unless
allowed.
precisely. Simply flagging a peer as htcp is not enough to turn it on. As
now documented.
A requesting peer needs to be allowed by in
http_access
and
On lör, 2007-10-13 at 18:45 +1300, Amos Jeffries wrote:
Henrik Nordstrom wrote:
hno 2007/10/12 17:58:43 MDT
Modified files:
src/fs/aufs store_dir_aufs.c
Log:
Make aufs store rebuilding back off a little if I/O load too high
aufs could stress
On tor, 2007-10-11 at 20:17 -0600, Alex Rousskov wrote:
Hello,
To support loadable modules, we may need to alter Squid sources layout
or installation procedure. This email discusses the problem and four
possible solutions. Please review and suggest the best path forward.
Possibly, but
On tor, 2007-10-11 at 19:30 -0600, Alex Rousskov wrote:
1) --with-ltdl :: This options will tell Squid to use internal ltdl
library (created by libtoolize) to handle module loading, preloading,
and unloading. I have a wrapper class from Traffic Spicer that can
handle both ltld-based and
On fre, 2007-10-12 at 12:30 +1300, Amos Jeffries wrote:
The squid.conf docs for udp_incoming_address specify that its ICP-only
related.
However the Config.Addrs.udp_incoming is used by ICP, HTCP, and DNS.
Similarly, docs for udp_outgoing_address specify that its ICP-only related.
However
On tor, 2007-10-11 at 08:31 -0600, Alex Rousskov wrote:
I want to move your Log Daemon from
http://wiki.squid-cache.org/DeadProjects into the Features/ space, but I
suspect that the current wiki page does not describe the current state
accurately enough. What is the current status of that
On tis, 2007-10-09 at 11:35 +0400, Alexander V. Lukyanov wrote:
Not sure about this one.. this will make Squid block under load and
additionally screws up fs load metrics.
I was under impression that only maintenance thread will block. Maybe I was
wrong.
Squid is not threaded (other than
On mån, 2007-10-08 at 12:23 +1300, Amos Jeffries wrote:
Small preview:
http://squid.treenet.co.nz/features.dyn
The details are possibly inaccurate right now.
Corrections:
Squid-3 supports delay pools.
2.5 did not support external acl helpers, including the session helper..
2.5 did
On fre, 2007-10-05 at 13:40 -0600, Alex Rousskov wrote:
I just found http://wiki.squid-cache.org/Plans page that you have
created a month ago. Should this be merged with Squid2 or Squid3
roadmaps? Linked from http://wiki.squid-cache.org/RoadMap ? Or should we
have a catch all page for
On lör, 2007-10-06 at 09:23 -0600, Alex Rousskov wrote:
Not sure how to interpret the don't do that and great you are tidying
this up conflict :-). Should I start moving/merging things around? Is
there a better plan?
Don't stop.
Have given it a whole 5 seconds though on what we need to make
On lör, 2007-10-06 at 07:35 +1000, Robert Collins wrote:
Pros:
+ Many consider svn to be overall better than CVS.
Its branch model is terrible.
Well, it's not worse than CVS imho, but not noticeably better either.
The CVS model of branches/tags is reasonable to work with, but
On sön, 2007-10-07 at 03:47 +0200, Henrik Nordstrom wrote:
bzr and git is the only two left standing on my list of choices. Neither
is perfect, but both significantly better than most other.. with bzr
being the main candidate.
On a related note the Samba project is currently migrating
On ons, 2007-10-03 at 21:24 +, Amos Jeffries wrote:
} else if (strcmp(token, transparent) == 0) {
+ // TODO: make squid barf when this option is passed to https_port */
Two problems with that comment:
1. Doesn't belong in squid3-ipv6.
2. Not true. There is valid use cases
On tor, 2007-10-04 at 12:16 +1300, Amos Jeffries wrote:
oh, that configuration is turning those machines into open proxies for
anyone who wants to point their domain at 'em.
No it won't, accelerator mode means an implicit never_direct allow all
just to prevent this from happening, which means
On tor, 2007-10-04 at 07:46 +0800, Adrian Chadd wrote:
On Thu, Oct 04, 2007, Henrik Nordstrom wrote:
On tor, 2007-10-04 at 12:16 +1300, Amos Jeffries wrote:
oh, that configuration is turning those machines into open proxies for
anyone who wants to point their domain at 'em
On tor, 2007-10-04 at 12:46 +1300, Amos Jeffries wrote:
So this is on a per-connection based on the incoming port option flag?
Yes. The accel flag gets set on requests accepted on an accelerator mode
port, which among other things forces DIRECT_NO in peerSelectFoo()
unless overridden...
On mån, 2007-10-01 at 20:53 +0200, Guido Serassio wrote:
Just upgraded autoconf and automake and now my development
environment is broken: I cannot no more bootstrap squid . :-(
What happens?
Regards
Henrik
signature.asc
Description: This is a digitally signed message part
On tis, 2007-10-02 at 11:20 +1300, Amos Jeffries wrote:
* are there any squid test units/programs to check my code?
So far I'm just running my STABLE13 patch without any problem.
If the patch is to be included in HEAD, I think it should need to
pass some tests.
hmm, good point. I'm
On sön, 2007-09-30 at 15:43 -0600, Alex Rousskov wrote:
Should _all_ size_t uses in BodyPipe be replaced with uint64_t then?
Anything that relates to the expected body content size to be
transferred over the BodyPipe, but not buffer or passwed chunk sizes.
Regards
Henrik
signature.asc
On sön, 2007-09-30 at 16:22 -0600, Alex Rousskov wrote:
BTW, why not make every size-related variable a signed or unsigned
int64_t? Is this a performance optimization issue?
To mainly to get a separation when something refers to in-memory data or
external data.
It's clearer in Squid-2 where
Both look strange to me. Shouldn't that be an ==
Regards
Henrik
On tor, 2007-09-27 at 11:28 +0200, Rafael Martinez (Squid development)
wrote:
Amos, a little typo:
Acording to original semantics of safe_inet_addr, I think the conditioal
expresion must be positive
SHOULD BE :
/*
On tor, 2007-09-27 at 12:08 +1200, Amos Jeffries wrote:
icap_class even says If there are multiple services per vectoring point,
they are processed in the specified order.
Yes.. config syntax already supports it, just never gets called..
Regards
Henrik
signature.asc
Description: This is a
On fre, 2007-09-28 at 09:01 +1200, Amos Jeffries wrote:
No, IPAddress can be assigned a string. It will try to do an inet_a/pton()
conversion. It returns true/false depending on whether the string was able
to be converted.
Then please use double ().
if ((mask = asc))
Regards
Henrik
On ons, 2007-09-26 at 11:53 +1200, Amos Jeffries wrote:
It occurs to me that a method superficially similar to that used for
cache_peer could look nice here.
- define icap_service with specific details, options +name on a single line
- define icap_access name acl ...
Isn't that what is
On ons, 2007-09-26 at 14:55 +1200, Amos Jeffries wrote:
no no. I meant all the *_access depends on acl etc we just finished
re-ordering the monolithic version for.
Ah, that. It's already there in cf_gen and will stay.
Regards
Henrik
signature.asc
Description: This is a digitally signed
On ons, 2007-09-26 at 03:54 +, Amos Jeffries wrote:
! DecodeMask(NULL, q-mask); // AYJ: is this just initialisation? if so
we don't need it.
! if (*l_mask !DecodeMask(l_mask, q-mask, iptype)) {
It's initialization to a /32 mask.
If the object starts out with a full mask then
On mån, 2007-09-24 at 09:49 -0600, Alex Rousskov wrote:
The implications range from the Module is a part of the GPL program
and, hence, must be GPLed! FSF point of view to there are no
implications except wasting time on discussing FSF FUD! point of view.
My personal view on that is that a
Hi,
I propose we change cf.data.pre to be split up in multiple files in the
CVS repository. Imho this would simplify maintenance noticeably.
There is already scripts for splitting and merging cf.data.pre in the
scripts directory. To see what I am talking about run
mkdir conf
On tis, 2007-09-25 at 23:38 +0800, Adrian Chadd wrote:
On Tue, Sep 25, 2007, Henrik Nordstrom wrote:
Hi,
I propose we change cf.data.pre to be split up in multiple files in the
CVS repository. Imho this would simplify maintenance noticeably.
One file per option, or?
Yes.
Regards
On ons, 2007-09-26 at 11:16 +1200, Amos Jeffries wrote:
That would require another directory just for the cf.data.pre content.
Yes, and a small script to pull it togeter for the build. (found in
scripts/)
I take it you've included the dependancy tracking in your merge script.
cf.data.pre
On mån, 2007-09-24 at 00:23 +0800, Adrian Chadd wrote:
* tune the udp buffer size rather than having it fixed - you could
in theory fire off 32k UDP packets if your receive side would
handle them fine. Thats ~200 logging entries each UDP packet.
You could, if you like fragmented IP
701 - 800 of 2137 matches
Mail list logo
