no until i test it cpl of month and i show it to the dev team they might do
better
--
View this message in context:
http://squid-web-proxy-cache.1019090.n4.nabble.com/Vary-object-loop-returns-tp4677716p4677725.html
Sent from the Squid - Users mailing list archive at Nabble.com.
__
On Thu, May 26, 2016 at 8:04 AM, Amos Jeffries wrote:
> On 24/05/2016 7:52 p.m., Tomas Mozes wrote:
> > Hello,
> > on two different squid servers I've observed a crash of pinger. First it
> > appeared on version 3.5.15 and later on version 3.5.17.
> >
> > Cache.log contains these lines:
> >
> > (
Hello. Whenever I tried to get squid to transparently cache https
content (mainly to speed up facebook browsing at my home), I get all
kinds of problems.
Is there a cookbook available for the recent squid versions?
Many thanks, -turgut
___
squid-u
Heh,
the issue occurs also with disabled collapsed_forwarding:
2016/05/31 16:58:12 kid1| varyEvaluateMatch: Oops. Not a Vary match on
second attempt,
'https://ru.wikipedia.org/w/load.php?debug=false&lang=ru&modules=jquery.accessKeyLabel%2Cclient%7Cmediawiki.RegExp%2Cnotify%2Cutil%7Cmediawiki.l
yup try accel the cache.log will have rely lots of those
my idea of fixing this or having better hit %
is by erasing anything start with %20 from the vary
'accept-encoding="gzip,%20deflate,%20sdch,%20br"'
those ,%20deflate,%20sdch,%20br should be removed and keep only
'accept-encoding="gzip
bluecoat doing same as my idea
i test my code against bluecoat server
--
View this message in context:
http://squid-web-proxy-cache.1019090.n4.nabble.com/Vary-object-loop-returns-tp4677716p4677730.html
Sent from the Squid - Users mailing list archive at Nabble.com.
Hmmm. Seems as dirty hack.
I suggest better to correct strings encoding routines, or simple strip
symbols starting with% from accept-encoding only.
31.05.2016 17:03, joe пишет:
yup try accel the cache.log will have rely lots of those
my idea of fixing this or having better hit %
is by
Bluecoat can do what they want :) They rich and fat CIA-tails. :)
31.05.2016 17:07, joe пишет:
bluecoat doing same as my idea
i test my code against bluecoat server
--
View this message in context:
http://squid-web-proxy-cache.1019090.n4.nabble.com/Vary-object-loop-returns-tp4677716p4677730
>I suggest better to correct strings encoding routines, or simple strip
>symbols starting with% from accept-encoding only.
yup exactly
my problem is how to represent stuff wen i post :(
that why they ignore or push me out with different answers
--
View this message in context:
http://squ
I swear this bug was only rarely triggered without collapsed_forwarding
enabled... but I think you'll have to wipe your cache to really test it...
--
Heiler Bemerguy - (91) 98151-4894
Assessor Técnico - CINBESA (91) 3184-1751
Em 31/05/2016 08:00, Yuri escreveu:
Heh,
the issue occurs also
This is too expensive action. Cold out cache will decrease hit-ratio for
weeks.
Also, I suggest force reloading objects must overwrite stale cached, no?
31.05.2016 19:32, Heiler Bemerguy пишет:
I swear this bug was only rarely triggered without
collapsed_forwarding enabled... but I think
Yeah, I know. This kind of stuff only appeared after enabling
collapsed_forwarding, I think. Now it's off again.. but I didn't wipe
the cache too, so I don't know.
2016/05/31 10:50:24 kid2| varyEvaluateMatch: Oops. Not a Vary match on
second attempt, 'http://js.statig.com.br/pub/adtags.j
load this link on both browserchrome and firefox
'http://js.statig.com.br/pub/adtags.js'
keep clicking reload on both browser and look at your cache.log
with or without collapsed_forwarding
Heiler Bemerguy wrote
> Yeah, I know. This kind of stuff only appeared after enabling
> collapsed_f
Without collapsed_forwarding seems ok:
https://i1.someimage.com/E8pM9tu.png
no warnings/errors in cache.log.
Seems Heiler right.
31.05.2016 19:33, joe пишет:
load this link on both browserchrome and firefox
'http://js.statig.com.br/pub/adtags.js'
keep clicking reload on both browser an
i dont have this issue right now with collapsed_forwarding on on my test
code
yes collaps and accel it trigger that bug more then collaps and intercept
without collaps on right no error in log but a miss no hit if u try on both
browsser
--
View this message in context:
http://squid-web-pro
Just did a test here.
Disabled every cache_dir and fully restarted squid 3.5.19
No more "vary object loop", but these continues:
2016/05/31 11:35:13 kid2| Bug: Missing MemObject::storeId value
2016/05/31 11:35:13 kid2| mem_hdr: 0x3177f40 nodes.start() 0x7370880
2016/05/31 11:35:13 kid2| mem_hd
Bad test. Problem is with swapped onto disk objects exactly. So, when we
disable cache_dirs, problem is gone.
31.05.2016 20:36, Heiler Bemerguy пишет:
Just did a test here.
Disabled every cache_dir and fully restarted squid 3.5.19
No more "vary object loop", but these continues:
2016/05/3
Yep I know it should/would vanish all those "Vary objects" messages..
but what are these other "Bug: Missing MemObject::storeId value" ?? I
thought it was related...
2016/05/31 11:43:27.594 kid2| clientProcessHit: URL mismatch,
'[unknown_URI]' !=
'http://www.agenciabelem.com.br/Resize/Creat
Hm. Seems related. The objects not the same in memory yet.
31.05.2016 20:45, Heiler Bemerguy пишет:
Yep I know it should/would vanish all those "Vary objects" messages..
but what are these other "Bug: Missing MemObject::storeId value" ?? I
thought it was related...
2016/05/31 11:43:27.594
I disabled ICP requests to it and these "Bug: Missing MemObject::storeId
value" messages stopped... any thoughts?
Known bug?
--
Best Regards,
Heiler Bemerguy
Network Manager - CINBESA
55 91 98151-4894/3184-1751
Em 31/05/2016 11:49, Yuri escreveu:
Hm. Seems related. The objects not the s
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
This proxy uses as peer for downstreams?
31.05.2016 21:09, Heiler Bemerguy пишет:
>
>
> I disabled ICP requests to it and these "Bug: Missing
MemObject::storeId value" messages stopped... any thoughts?
>
> Known bug?
>
> --
> Best Regards,
>
> He
This is the parent proxy of another one (10.1.10.101).
It was receiveing requests like these:
1464665138.326 0 10.1.10.101 UDP_MISS/000 72 ICP_QUERY
http://img.olx.com.br/thumbs/51/51763636766.jpg - HIER_NONE/- -
1464665138.331 0 10.1.10.101 UDP_MISS/000 72 ICP_QUERY
http://img.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
Aha,
for cached objects no matter HTTP or ICP query requested they.
31.05.2016 22:36, Heiler Bemerguy пишет:
>
>
> This is the parent proxy of another one (10.1.10.101).
>
> It was receiveing requests like these:
>
> 1464665138.326 0 10.1.10
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
We have Wiki.
http://wiki.squid-cache.org/ConfigExamples/Intercept/SslBumpExplicit
31.05.2016 16:00, turgut kalfaoğlu пишет:
>
> Hello. Whenever I tried to get squid to transparently cache https
content (mainly to speed up facebook browsing at m
Hi Nishant,
>>>* (The last two commands in as quick succession as possible - preferably on a
*>>>* single line separated by a semi-colon)
*>>* yes the problem still exists
*
>Could it be because squid is trying to resolve and connect to IPv6 address
>first?
>Try setting "dns_v4_first on
Hey Amos,
>What is causing the problem is that ICAP services need to be working
>*immediately* and do not wait for DNS results to come back. If they are
>not available immediately then the service is not contactable for that
>transaction.
>Adding /etc/hosts entries makes Squid load the name+IP de
Hi Nilesh,
Just add a –d to
# enable kerberos authentication
auth_param negotiate program /usr/lib/squid/negotiate_kerberos_auth -s
HTTP/hostname.domain@domain.org
like
# enable kerberos authentication
auth_param negotiate program /usr/lib/squid/negotiate_kerberos_auth -s
HTTP/hostname
Attaching full log of how it fails initially and then able to resolve dns
lookup in 3 minutes
collected by setting debug_options ALL, 3
>
2016/05/31 16:55:05.157| 42,2| IcmpPinger.cc(211) SendResult: return result
to squid. len=76
2016/05/31 16:55:05.157 kid1| 51,3| fd.cc(93) fd_close: fd_c
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
sslproxy_cipher
!DH-DSS-MISTY1-SHA:!DH-RSA-MISTY1-SHA:!SSLv2:+SSLv3:!AECDH:!ADH:!DES:HIGH:+3DES:!RC4:!MD5:!aNULL:!eNULL:!LOW:!EXP:!DSS:!PSK:!SEED:!SRP
20.05.2016 20:44, Walter H. пишет:
> sslproxy_cipher
> !SSLv2:+SSLv3:!AECDH:!ADH:!DES:HIGH:+3D
On 1/06/2016 4:36 a.m., Heiler Bemerguy wrote:
>
> This is the parent proxy of another one (10.1.10.101).
>
> It was receiveing requests like these:
>
> 1464665138.326 0 10.1.10.101 UDP_MISS/000 72 ICP_QUERY
> http://img.olx.com.br/thumbs/51/51763636766.jpg - HIER_NONE/- -
> 1464665138.
On 1/06/2016 1:34 a.m., Yuri wrote:
> This is too expensive action. Cold out cache will decrease hit-ratio for
> weeks.
>
You do not need to suffer from a simple test. Make a temporary empty
cache_dir and comment out the main one while testing.
That way you can revert to the larger main cache at
Dear squid-users,
at first:
thanks for adding me to your mailing list.
My question is about the cache_peer configuration directive.
In my config I have to use three cache_peer proxies:
To choose the cache_peer on the basis of the URL its responsible for I use
the cache_peer_access directive and
32 matches
Mail list logo
