he Entries: 470
IPcache Requests: 216284
IPcache Hits: 150664
IPcache Negative Hits: 31
IPcache Misses: 35560
IPcache Redirected: 17430
^^
--
Evgeny Kotsuba.
On Fri, 4 Nov 2005 14:29:56 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Fri, 4 Nov 2005, Evgeny Kotsuba wrote:
In case of detach/daemon mode admin can't see , but still can hear
beeps.
In detached daemon mode there is no console output at all. Only log
file o
On Thu, 3 Nov 2005 23:17:12 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Thu, 3 Nov 2005, Evgeny Kotsuba wrote:
* UI patches (!) including usage of sound at errors/warnings and
visual
effects at errror/warning ( as sample -
http://www.laser.ru/evgen/soft/
t to the end of defrozing and next I can offer
* C <->C++ patches
* UI patches (!) including usage of sound at errors/warnings and
visual effects at errror/warning ( as sample -
http://www.laser.ru/evgen/soft/Squid2/squid2screan.gif )
SY,
Evgeny Kotsuba
On Tue, 1 Nov 2005 15:05:51 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
Summary of the opinions regarding a Squid-2.6 release
brr...
I can't understand, is it in "release" state now, or "just
defrosted" ?
SY,
Evgeny Kotsuba
On Sun, 23 Oct 2005 20:45:24 +0800
Adrian Chadd <[EMAIL PROTECTED]> wrote:
On Sun, Oct 23, 2005, Evgeny Kotsuba wrote:
On Sun, 23 Oct 2005 11:49:01 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
>On Sat, 22 Oct 2005, Henrik Nordstrom wrote:
>
>>List of t
supposed as first thing to do before migation to C++ and 3.0
i.e.
Change things like
int class;
to
int Class;
etc.
SY,
Evgeny Kotsuba
On Tue, 18 Oct 2005 01:20:27 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Mon, 17 Oct 2005, Evgeny Kotsuba wrote:
my %hits = ('TCP_HIT' => 1,
'TCP_REFRESH_HIT' => 1,
TCP_REFRESH_HIT is thing that had bytes to be downloaded from
its = ('TCP_HIT' => 1,
'TCP_REFRESH_HIT' => 1,
TCP_REFRESH_HIT is thing that had bytes to be downloaded from the
origin server due to not misses, but good or wrong will of the
origin server
SY,
Evgeny Kotsuba
On Mon, 17 Oct 2005 00:16:32 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Mon, 17 Oct 2005, Evgeny Kotsuba wrote:
--8<-
1129493582.720 2000 127.0.0.1 TCP_REFRESH_HIT/200 455 GET
http://images.rambler.ru/m3/w/l-i-Flagge
On Sun, 16 Oct 2005 13:04:28 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Sat, 15 Oct 2005, Evgeny Kotsuba wrote:
TCP_REFRESH_HIT indicates the cached copy was either stale, or
reload into
IMS was used forcing just a cache revalidation on "Pragma:
no-cache&qu
On Sat, 15 Oct 2005 05:10:19 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Fri, 14 Oct 2005, Evgeny Kotsuba wrote:
Well. If "client explicitly forbids the cache to be used to server
this
request." then why we have TCP_REFRESH_HIT for this request ?
Rig
On Fri, 14 Oct 2005 11:22:53 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Thu, 13 Oct 2005, Evgeny Kotsuba wrote:
And were is logic ?
requesting client said that Squid MUST NOT use a cached copy
Which is why there was not a cache hit. The client request forbits
this
On Wed, 12 Oct 2005 18:59:31 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Tue, 11 Oct 2005, Evgeny Kotsuba wrote:
Once again.
This url is _REDIRECTED_ to dot.gif at my local apache
___^__
Why quid ask anything about it ? As well why squ
On Wed, 12 Oct 2005 00:48:47 +0400
Andrey Shorin <[EMAIL PROTECTED]> wrote:
Hello Evgeny,
Tuesday, October 11, 2005, 23:54:06, Evgeny Kotsuba wrote:
Forced reload of the cache due to "Pragma: no-cache"
request header.
Once again.
On Tue, 11 Oct 2005 00:51:07 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Tue, 11 Oct 2005, Evgeny Kotsuba wrote:
1128979352.320470 127.0.0.1 TCP_REFRESH_HIT/200 341 GET
http://u1386.48.spylog.com/cnt?p=1 evg NONE/- text/html [Host:
u1386.48.spylog.com\r\nUser
On Mon, 10 Oct 2005 00:48:43 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Sat, 8 Oct 2005, Evgeny Kotsuba wrote:
1128764141.130 4280 127.0.0.1 TCP_CLIENT_REFRESH_MISS/200 743
GET http://kubok.yandex.ru/i/empty-top.jpg evg DIRECT/213.180.204.2
imag
t *mem = e->mem_obj;
assert(mem != NULL);
assert(len >= 0);
assert(e->store_status == STORE_PENDING);
I can't repeate the query because there is now such directory at
this ftp now.
Nevertheless,I am afraid of possible anti-squid issue
SY,
Evgeny Kotsuba
On Sat, 30 Apr 2005 16:00:40 +0200
Serassio Guido <[EMAIL PROTECTED]> wrote:
Hi Evgeny,
At 13.05 30/04/2005, Evgeny Kotsuba wrote:
I have removed similar things some time ago from the Windows port for
poor performance reasons.
I don't understand this. Really call to read/write/sev
rc = 3;
break;
case 3:
rc = 4;
break;
}
return rc;
}
take 1 / 14 * 1000 * 1000 sec on good old P2-350. Without
optimization
if suppose that every call to such function is to read/write 512
byte than the limit is about 7 Gb/sec ;-)
SY,
Evgeny Kotsuba
t,
macroses FD_XXX, fd_set structure and
#pragma pack(4)
typedef struct fd_set {
u_short fd_count; /* how many are SET? */
int fd_array[FD_SETSIZE]; /* an array of SOCKETs */
} fd_set;
#pragma pack()
and it's usage in comm_select.c(pp)
Nevertheless this it seems solved
SY,
Evgeny Kotsuba
On Fri, 29 Apr 2005 11:16:23 +0200
Serassio Guido <[EMAIL PROTECTED]> wrote:
Hi Evgeny,
At 09.52 29/04/2005, Evgeny Kotsuba wrote:
FD_READ_METHOD/FD_WRITE_METHOD macros should be more efficient.
Look here:
http://cvs.sourceforge.net/viewcvs.py/squid/squid/src/fd.c?rev=1.7.16.9&only_w
default:
rc = ::read(fh, buf, len);
}
if(rc>0) bytes_read += rc;
return rc;
}
SY,
Evgeny Kotsuba
E:
return _read(fd, buf, siz);
case FD_SOCKET:
return recv(fd, buf, siz, 0);
...
}
}
This is right what I do in my OS/2 port of squid
SY,
Evgeny Kotsuba
On Wed, 13 Apr 2005 01:24:01 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Wed, 13 Apr 2005, Evgeny Kotsuba wrote:
Well, but there is no any key what server and what is bad in
{HTTP/1.0 200
OK} in warning message
Example reply headers where this message is seen:
HTTP/1
On Tue, 12 Apr 2005 12:04:54 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Tue, 12 Apr 2005, Evgeny Kotsuba wrote:
Has this [strange] message in log any reasonable information ?
WARNING: unparseable HTTP header field {HTTP/1.0 200 OK}
squid 2.5.x
This is commonly see
Hi,
Has this [strange] message in log any reasonable information ?
WARNING: unparseable HTTP header field {HTTP/1.0 200 OK}
squid 2.5.x
SY,
Evgeny Kotsuba
On Mon, 11 Apr 2005 22:22:16 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Mon, 11 Apr 2005, Evgeny Kotsuba wrote:
Is helperHandleRead() called in situation when squid just started
but there
is no one client connection ?
helperHandleRead is always pending for read fr
On Mon, 11 Apr 2005 11:17:48 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Mon, 11 Apr 2005, Evgeny Kotsuba wrote:
I have a look at void helper.c and can't find point were srv->rfd
is
closed
helperHandleRead() when seeing the connection closed by
(redirectors+auths+ etc.)
SY,
Evgeny Kotsuba
ent, we changed the way swap file
* numbers are assigned and stored. The high 16 bits used
--8<--
I didn`t quite catch what those ifs do, but
\Squid\S2_5\squid-s2_5\src\fs\coss\store_dir_coss.c in
storeCossRebuildFromSwapLog() still has those.
SY,
Evgeny Kotsuba
name '%s' not defined!\n", t);
self_destruct();
memFree(L, MEM_ACL_LIST);
continue;
}
===
aclParseUserMaxIP() has no any additional info on the reason for
self_destruct
error:
+ debug(28, 0) ("ERROR: aclParseUserMaxIP: Malformed ACL\n");
self_destruct();
---
SY,
Evgeny Kotsuba
On Mon, 14 Feb 2005 23:23:21 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Mon, 14 Feb 2005, Evgeny Kotsuba wrote:
On Mon, 14 Feb 2005 11:00:23 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
Hi Guido, and thanks for making many small FreeBSD cleanups in
Squid-2
On Mon, 14 Feb 2005 23:35:02 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Tue, 15 Feb 2005, Evgeny Kotsuba wrote:
Well, why it is parsed properly without return
It didn't. Earlier versions just did not notice it could not parse
this malformed header properly..
and why
On Mon, 14 Feb 2005 20:22:25 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Mon, 14 Feb 2005, Evgeny Kotsuba wrote:
Hi,
What is WARNING: suspicious CR characters in HTTP header ?
Say, what is wrong in
"Set-Cookie: sbtpoll1067196868\r=1067196868\r; path=/; expire
ner
with bug you will see "invalid responce" error essage.
On Mon, 14 Feb 2005 18:51:30 +0300
"Evgeny Kotsuba" <[EMAIL PROTECTED]> wrote:
Hi,
What is WARNING: suspicious CR characters in HTTP header ?
Say, what is wrong in
"Set-Cookie: sbtpoll1067196868\r=106719
spicious CR characters in
HTTP header near {%s}\n",
getStringPrefix(field_start, header_end));
//??EKreturn httpHeaderReset(hdr);
}
---8<---
SY,
Evgeny Kotsuba
?
SY,
Evgeny Kotsuba
On Sun, 6 Feb 2005 02:03:23 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Sun, 6 Feb 2005, Evgeny Kotsuba wrote:
One more potential rakes with sqiuid: call to signal() doesn't
check
for errors
Most UNIX software doesn't as this call rarely if ever errors, but
omething like:
#else
if (SIG_ERR == signal(sig, func))
debug(50, 0) ("WARNING: signal: sig=%d func=%p: %s\n", sig, func,
xstrerror());
#endif
SY,
Evgeny Kotsuba
On Wed, 26 Jan 2005 00:08:47 +0100 (CET)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Wed, 26 Jan 2005, Evgeny Kotsuba wrote:
What is the reason for last changes in strwordtok() (tools.c
/tools.cc) ?
To support the external_acl protocol in a more robust manner.
I.e is it now
* Simi
hesi.e.. native form of paths for OS/2 and
Windows. The worse thing is that it is with paths like
C:\Squid\auth\ncsa_auth.exe or C:\Squid\redirect.exe but not
mention anywere in config and/or docs
How often '\n' and '\r' are used in config ?
SY,
Evgeny Kotsuba
Hi,
cf.data.pre
---8<---
NAME: memory_pools_limit
COMMENT: (bytes)
TYPE: b_size_t
DEFAULT: 5 MB
If not set (default) or set to zero, Squid will keep all memory it
^
can. That is, there will be no limit on the total amount of m
if (!escape && (ch <= 0x1F || ch >= 0x7f) && ch != '\n' && ch
!= '\r' && ch != '\t') {
.
}
Is there any drawbacks for such change ?
SY,
Evgeny Kotsuba
Hi,
On Thu, 19 Aug 2004 11:28:48 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Wed, 18 Aug 2004, Evgeny Kotsuba wrote:
One my client askes me if it is possible to pass client io address
to auth
module.
Please describe why this is needed in the authentication module and
h
t possible for us to run two session of squid
server
on one machine.
Yes. Just make sure they do not share any common resources (port
numbers, cache directories, log files etc...)
But how two squids can distinguish signals via -k signal option ?
SY,
Evgeny Kotsuba
Hi,
One my client askes me if it is possible to pass client io address to
auth module.
Where I can find start point for squid 2.5 modification ?
PS: What about Squid 3.0 PRE 3 for OS/2 - I don't know anybody who
make it while there are as minimum 3 sources for ver 2.5
SY,
Evgeny Kotsuba
s
"\0x20\0x20\0x20NOT_IN_MEMORY SWAPOUT_DONE
PING_NONE\0x20\0x20\0x20\0x00" ?
Is it reasonable to store extra 6 spaces ?
SY,
Evgeny Kotsuba
Hi,
Some time ago I get the subj from my CVS client.
So, is there some problems with it ?
SY,
Evgeny Kotsuba
On Fri, 16 Apr 2004 11:23:02 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Fri, 16 Apr 2004, Evgeny Kotsuba wrote:
Hi,
Squid 2.5 and Squid 3
file pinger.c/pinger.cc, function pingerSendEcho()
1)
I have
//MAX_PKT_SZ=294, sizeof(pkt)=294, sizeof(struct
icmhdr)=28,
of(struct timeval) + sizeof(char);
In 3.0 and 2.4
icmp_pktsize += sizeof(icmpEchoData) - MAX_PAYLOAD;
definitly that with different stucture alingment we will have differet
values for icmp_pktsize in both cases, so what is right ?
SY,
Evgeny Kotsuba
Hi,
What about perspective on using gzip encodind in squid ?
SY,
Evgeny Kotsuba
Hi,
squid2.5.
Is macro ALLOW_SOURCE_PING also dead macro as some others ?
SY,
Evgeny Kotsuba
ue_size < hlp->n_running)
return;
-if (hlp->stats.queue_size < hlp->n_running)
- return;
====
SY,
Evgeny Kotsuba
Hi,
What is macros WHEN_SQUID_IS_HTTP1_1 stand for in
squid-s2_5/src/HttpMsg.c ?
And where it is difined ?
SY,
Evgeny Kotsuba
C is a directory
separator
in HTTP URLs.
I suppose that IE is such broken client. Moreover, Mozilla is also
such client - it defenetly can read direct URL like
http://www.lasercomp.ru/technologies/..\img/lk.gif
But can't
http://www.lasercomp.ru/technologies/..%5cimg/lk.gif
SY,
Evgeny Kotsuba
}
+ *t = 0;
+}
if (stringHasWhitespace(urlpath)) {
debug(23, 2) ("urlParse: URI has whitespace: {%s}\n", url);
=
Evgeny Kotsuba
nt to hear such
arguments, may be due to using only gcc compilers
PS: By the way, https://sourceforge.net/projects/squid/ states, that #
Programming Language: C
SY,
Evgeny Kotsuba
On Thu, 28 Aug 2003 22:37:06 +0200
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Thursday 28 August 2003 17.37, Evgeny Kotsuba wrote:
My squid is not uzing signals for IPC, instead it uses named pipes
so now I add option to command line parameters to send to working
squid (section, level)
Hi,
My squid is not uzing signals for IPC, instead it uses named pipes so
now I add option to command line parameters to send to working squid
(section, level) as in debug_options in config.
May be this idea will be interesting to mainstream squid developers.
SY,
Evgeny Kotsuba
ds on "main changes" - say C ->
C++ changing and its pro et contra.
SY,
Evgeny Kotsuba
IBM's VAC3 I can't even compile things like
template
Range
Range::intersection (Range const &rhs)
{Range result (XMAX(start, rhs.start), XMIN(end, rhs.end));
return result;
}
Compiler wants
Range::intersection (Range const &rhs)
SY,
Evgeny Kotsuba
On Thu, 07 Aug 2003 12:34:07 +0200
Serassio Guido <[EMAIL PROTECTED]> wrote:
Hi,
At 12.18 07/08/2003, Evgeny Kotsuba wrote:
Hi,
What is the reason that tools.c -> strwordtok() eats all '\'
characters ?
With it it is impossible to put in config any path in non-unix
notati
On Thu, 7 Aug 2003 14:01:02 +0200 (CEST)
Henrik Nordstrom <[EMAIL PROTECTED]> wrote:
On Thu, 7 Aug 2003, Evgeny Kotsuba wrote:
What is the reason that tools.c -> strwordtok() eats all '\'
characters ?
Because strwordtok splits strings into words according to UNIX
"sh
t this place.
I can use backslashes for all other directories and pathes exept this
"auth_param basic program "
May be it is worth to use pure strtok in authBasicParse() ?
SY,
Evgeny Kotsuba
*/
But as I see this knowledge of quoting is a bit too rudimentary.
SY,
Evgeny Kotsuba
We are now looking to merge our patch into the main squid tree and
provide support for further squid release.
Very interesting,
I will wait for it for 2.5
SY,
EK
SY,
Evgeny Kotsuba
67 matches
Mail list logo
