On Monday 18 January 2016 at 11:32:02, behrad eslami wrote:
> User traffic diagram is like this:
>
> _receive squid <-router1-- |
> | <---> internetuser
> ---send>
I just checked it. It'll work at the moment. But only because the
dependencies (and the dependency version) doesn't changed from 3.4.8 to
3.5. So there's is no guarantee that it will work with further releases.
On the other hand: Installing unstable software is not the way the state
system
Thanks for you reply
Im askinng about one way traffic. Some user sends traffic ,route from another
ISP and only received packet route from squid
On Sunday, January 17, 2016 12:21 PM, Antony Stone
wrote:
On Sunday 17 January 2016 at 08:55:56, behrad
Hai,
> I just checked it. It'll work at the moment. But only because the
> dependencies (and the dependency version) doesn't changed from 3.4.8 to 3.5.
> So there's is no guarantee that it will work > with further releases.
Yes and if depencies change, you can do the same for these
On Monday 18 January 2016 at 10:56:27, behrad eslami wrote:
> Thanks for you reply
> Im askinng about one way traffic. Some user sends traffic ,route from
> another ISP and only received packet route from squid
Sorry, this is still not clear to me.
Do the requests from the client to the server
You saved hours of debugging for me. Thank You!
It works with the patch applied.
--
Gergely EGERVARY
This is a bug in IPFilter 5. We're waiting to hear back from the
IPFilter maintainer before committing.
Try this patch (and read the PR for more info):
---
Hi list,
Since I upgraded the OS to CentOS 7.2 and Squid to 3.5.12 (RPM from
www1.ngtech.co.il the local squidclient gets no answer if pointed to
localhost:
# squidclient -v -p 3128 mgr:info
Request:
GET cache_object://localhost/info HTTP/1.0
Host: localhost
User-Agent: squidclient/3.5.12
Accept:
User traffic diagram is like this:
_receive squid <-router1-- |
| <---> internetuser
---send> router2
On Monday, January 18, 2016 1:41 PM, Antony Stone
On 16/01/2016 13:16, Egerváry Gergely wrote:
> Hi,
>
> I'm running on:
> - NetBSD 7.0_STABLE (checked out today)
> - Squid 3.5.12 from NetBSD pkgsrc 2015Q4
> - IP Filter: v5.1.2 (536)
>
> Configured with "--enable-ipf-transparent":
>
> $ ./configure --sysconfdir=/usr/pkg/etc/squid
>
On Monday 18 January 2016 at 17:55:51, romain noyer wrote:
> Is there a way to get all the messages a squid server can create and send
> to a syslog?
See the "syslog" method of:
http://www.squid-cache.org/Doc/config/access_log/
http://www.squid-cache.org/Doc/config/cache_store_log/
> The goal
On Monday 18 January 2016 at 18:22:24, Lucía Guevgeozian wrote:
> acl good_facebook urlpath_regex groups
> acl banned_sites url_regex "/etc/squid/config/banned_sites"
>
> inside banned_sites I have the word facebook
>
> http_access allow good_facebook
> http_access deny banned_sites
Okay, so
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
And more:
Facebook (like more others) uses Akamai CDN as background delivery service.
So, facebook.* domain is a little part of whole big fat Facebook :)
18.01.16 23:29, Antony Stone пишет:
> On Monday 18 January 2016 at 18:22:24, Lucía
On Monday 18 January 2016 at 18:31:40, Yuri Voinov wrote:
> Facebook (like more others) uses Akamai CDN as background delivery service.
>
> So, facebook.* domain is a little part of whole big fat Facebook :)
True, but that should still match *request* URLs (once the HTTP/S problem is
sorted
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
18.01.16 23:38, Antony Stone пишет:
> On Monday 18 January 2016 at 18:31:40, Yuri Voinov wrote:
>
>> Facebook (like more others) uses Akamai CDN as background delivery
service.
>>
>> So, facebook.* domain is a little part of whole big fat
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
18.01.16 23:56, Lucía Guevgeozian пишет:
> Thank you very much for your responses.
>
> I understand from http://www.squid-cache.org/Doc/config/http_access/ that
> http_access will not work with https in version of squid older than 3.3.
>
> Do
Hello,
I think I have a very basic question about acl, but I can't figure out why
this simple config is not working:
In my squid.conf file I have 2 acl
acl good_facebook urlpath_regex groups
acl banned_sites url_regex "/etc/squid/config/banned_sites"
inside banned_sites I have the word
Thank you very much for your responses.
I understand from http://www.squid-cache.org/Doc/config/http_access/ that
http_access will not work with https in version of squid older than 3.3.
Do you know if an alternative config exists without upgrading?
Regards,
Lucia
2016-01-18 14:38 GMT-03:00
Ok, thanks again for the quick reply, I'm upgrading :)
Regards,
Lucia
2016-01-18 14:58 GMT-03:00 Yuri Voinov :
>
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA256
>
>
>
> 18.01.16 23:56, Lucía Guevgeozian пишет:
> > Thank you very much for your responses.
> >
> > I
I didn't test this, but i think it works better:
*http_access deny banned_sites !good_facebook*
is it works?
2016-01-18 16:35 GMT-02:00 Lucía Guevgeozian :
> Ok, thanks again for the quick reply, I'm upgrading :)
>
> Regards,
> Lucia
>
> 2016-01-18 14:58 GMT-03:00 Yuri
Hi, unfortunately I tried that already and in 3.0 version I can say it
didn't work.
cheers
2016-01-18 15:43 GMT-03:00 Jorgeley Junior :
> I didn't test this, but i think it works better:
> *http_access deny banned_sites !good_facebook*
> is it works?
>
> 2016-01-18 16:35
Hello all,
I really apologize in advance if this topic have already been discussed,
but I didn't found anything.
Is there a way to get all the messages a squid server can create and send
to a syslog? Or somewhere I can get a database or .xls file or whatever
containing the whole list of messages
Hi Squid experts,
Can you please tell me if below scenario is possible to be implemented in Squid?
<--> User 1 enters the proxy, browses some pages.
<--> User 2 tries to enter, and he receives a reject.
<--> User1 stops browsing pages.
<--> User2 tries to enter, but because the TTL is not
On 19/01/2016 5:19 a.m., Murat Balkan wrote:
> Hi Squid experts,
>
> Can you please tell me if below scenario is possible to be implemented in
> Squid?
>
> <--> User 1 enters the proxy, browses some pages.
> <--> User 2 tries to enter, and he receives a reject.
> <--> User1 stops browsing
On 19/01/2016 6:56 a.m., Lucía Guevgeozian wrote:
> Thank you very much for your responses.
>
> I understand from http://www.squid-cache.org/Doc/config/http_access/ that
> http_access will not work with https in version of squid older than 3.3.
Incorrect. http_access works with any HTTP message
On 16/01/2016 4:11 a.m., Murat Balkan wrote:
> Hi,
>
> Thanks for the response.
> What I want to achieve is to prevent 2 users enter with the same username
> simultaneously.
User usernames are unique. Same username == same user.
There cannot be a second user with same username.
However:
On 18/01/2016 10:13 a.m., Roman Gelfand wrote:
> I am not sure where I am going wrong here...
>
>
> ssl bump certificate
> openssl req -new -newkey rsa:2048 -sha256 -days 365 -nodes -x509 -keyout
> squidCA.pem -out squidCA.pem
>
> The der certificate was generated and deployed on client
Hi
Is it possible to implement delay pools such that
if file is less than 10M
then
allow 60Mb/s
else
allow 20Mb/s
fi
is that possible the aim is to allow a higher through put for smaller
files, but to limit bigger / longer connections
Alex
___
On 18/01/2016 1:58 p.m., Alex Samad wrote:
> Hi
>
> so I have this in place now
>
> This works well for delaying YAY
>
> #
> # Delay Pools
> # http://wiki.squid-cache.org/Features/DelayPools
> #
>
On 19 January 2016 at 16:59, Amos Jeffries wrote:
>
> Hmm. Are you using the exact same HTTP headers as WU tools on the other
> machines do to prefetch the URL into the cache ?
I have a script that checks the squid logs and then does a download of
the files through the
On 19/01/2016 7:11 p.m., Alex Samad wrote:
> Hi
>
> Think I answered my own on this
> refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
>
>
> Does the last refresh_pattern config win ?
>
No, this one does:
"windowsupdate.com/.*\.(cab|exe|ms[i|u|f]|[ap]sf|wm[v|a]|dat|zip)"
The problem is
On 19/01/2016 6:52 p.m., Alex Samad wrote:
> Hi
>
> Is it possible to implement delay pools such that
>
> if file is less than 10M
> then
> allow 60Mb/s
> else
> allow 20Mb/s
> fi
>
There is no "file" in HTTP. Only messages.
Some messages have payloads. Sometimes those payload sizes are
On 19 January 2016 at 16:59, Amos Jeffries wrote:
>> refresh_pattern -i
>> microsoft.com/.*\.(cab|exe|ms[i|u|f]|[ap]sf|wm[v|a]|dat|zip) 4320 80%
>> 129600 reload-into-ims
>> refresh_pattern -i
>> windowsupdate.com/.*\.(cab|exe|ms[i|u|f]|[ap]sf|wm[v|a]|dat|zip) 4320
>> 80%
On 18/01/2016 11:13 p.m., Henri Wahl wrote:
> Hi list,
> Since I upgraded the OS to CentOS 7.2 and Squid to 3.5.12 (RPM from
> www1.ngtech.co.il the local squidclient gets no answer if pointed to
> localhost:
>
> # squidclient -v -p 3128 mgr:info
> Request:
> GET cache_object://localhost/info
Hi
Think I answered my own on this
refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
Does the last refresh_pattern config win ?
On 19 January 2016 at 17:08, Alex Samad wrote:
> On 19 January 2016 at 16:59, Amos Jeffries wrote:
>>> refresh_pattern -i
On 2016-01-18 14:59, Antony Stone wrote:
Forwarding private reply back to the list...
-- Forwarded Message Starts --
Thanks for your answer.
Sorry for my poor english, I'll try to reword because I'm not looking
for a
log analyzer. In fact, I don't even need Squid itself
35 matches
Mail list logo