On Jan 25, 2008 2:58 AM, bijayant kumar [EMAIL PROTECTED] wrote:
Hi Arana,
Thanks for your reply. As you are suggesting in your
reply that incresing the filedescriptor can be
dangerous. Is there any other way to get rid of this
warning, because this warning makes browsing dead
by the
I would recommend you to run ./configure with
--with-maxfd=you_desired_limit and --enable-epoll
Watch for messages like this in configure output:
checking if epoll works... yes
Using epoll for the IO loop.
...
Maximum filedescriptors set to 131072
...
Having large number of FDs with select is
On Jan 9, 2008 10:35 AM, Shalvi Ziv [EMAIL PROTECTED] wrote:
Thanks. I'll look for the cache manager options.
About 'requests not handled by squid' - it means requests that got to
the squid, then the squid crashed unexpectedly and could not handle the
requests (hence, no entry in the
with these issues and my proposed fix.
Once there is a desition on these, I'll forward this to 2.7.
Regards,
On Jan 8, 2008 1:12 PM, Adrian Chadd [EMAIL PROTECTED] wrote:
On Tue, Jan 08, 2008, Gonzalo Arana wrote:
(I do)**3
Will test it tomorow.
Porting the changeset to 2.6 and 2.7
On Jan 9, 2008 11:30 AM, Shalvi Ziv [EMAIL PROTECTED] wrote:
OF COURSE it does not crashes but for SLA (support) purposes we need to
provide a report of unprocessed requests... just in case.
I see.
As I see now, when increasing debug level (e.g. section 33 level 5) I
can see when request is
On Jan 7, 2008 1:49 PM, Adrian Chadd [EMAIL PROTECTED] wrote:
I gave up and coded up a way to include configuration files in squid.conf.
you can now do something like this:
include /full/path/to/another/file/to/include
without quotes or anything like that. Heck, spaces in the filename may
Ziv,
On Jan 7, 2008 7:34 PM, Shalvi Ziv [EMAIL PROTECTED] wrote:
For support and availability reports purposes, I need to count the total
number of requests to Squid, including those that have not been handled
by the squid, and compare it with the number of handled requests.
Meaning, in case
On Dec 12, 2007 2:49 PM, Jason Gauthier [EMAIL PROTECTED] wrote:
All,
Is there really any difference on a Linux system between utilizing the TPROXY
method versus Netfilter method? And are there anything that outweighs one
from the other?
AFAIK, the differences are:
1) TPROXY allows you to
On 9/20/07, zulkarnain [EMAIL PROTECTED] wrote:
--- tech user [EMAIL PROTECTED] wrote:
Are you running 64-bit CPU,64-bit OS and 64-bit
Squid program?
Otherwise I don't think your squid can use full of
4G memory for cache_mem.
Yes! I'm running 64-bit CPU, 64-bit OS and 64-bit
squid.
On 9/21/07, Arthur Tumanyan [EMAIL PROTECTED] wrote:
Henrik Nordstrom-5 wrote:
Probably your redirectors are broken. What redirector are you using?
Regards
Henrik
I think so,because the redirector is just writing ,and all this working in
test mode only.
Hi,
Have a look at cache manager:
http://wiki.squid-cache.org/SquidFaq/CacheManager
HTH,
On 9/20/07, zulkarnain [EMAIL PROTECTED] wrote:
Hi all,
I've squid running with 4GB of cache_mem, seemed my
squid unable to use 4GB of cache_mem. I would like to
know is there any tools to analyze
in 32bit or 64bit? I am unfamiliar with
Fedora, so I can't provide much help in this one, sory.
HTH,
On 9/20/07, zulkarnain [EMAIL PROTECTED] wrote:
--- Gonzalo Arana [EMAIL PROTECTED] wrote:
Hi,
Have a look at cache manager:
http://wiki.squid-cache.org/SquidFaq/CacheManager
Here
Hi Preetish,
From your previous email, I suspect your squid CPU usage is that high
due to url_regex acls (perhaps the acl files contain too many regexes
to evaluate). Try commenting them out, or simply making those files
empty.
HTH,
--
Gonzalo A. Arana
See ./configure --help output, there was a specific flag for enabling
content encoding (--enable-content-encoding, I assume).
Hope this helps,
On 3/16/07, Wahyu [EMAIL PROTECTED] wrote:
Hi,
Any one have try squid3+content encode successfully? i got problem with
my installation. Patch and build
Hi,
I'd suggest:
1) memory_pools on (and memory_pools_limit 50MB should be enough).
2) make sure that you have enough FDs available (cache manager may be useful).
3) make sure your squid use kqueue/epoll rather than poll/select.
4) Monitoring FD usage ipcache size is a good idea.
5) aufs should
maximum_object_size perhaps.
On 1/10/07, zulkarnain [EMAIL PROTECTED] wrote:
--- Paul Johnson [EMAIL PROTECTED] wrote:
Not quite sure if I can really provide much insight
on this, since it Just
Worked(tm) for me.
Just wondering how you do that? refresh_pattern?
Regards,
--
Gonzalo A.
Just shooting to the air, but perhaps disabling TCP timestamps would help.
I had a problem with www.wdc.com some time ago related with tcp timestamps.
HTH,
On 10/13/06, Kumoro Wibowo [EMAIL PROTECTED] wrote:
I have experienced the same problem also with squid-2.6.STABLE4 on FC5.
However, I
On 10/13/06, Jens Strohschnitter [EMAIL PROTECTED] wrote:
...
Just shooting to the air, but perhaps disabling TCP timestamps would help.
I had a problem with www.wdc.com some time ago related with tcp timestamps.
Where do I have to deactivate them ? On my firewall or in the squid.conf ?
On 10/11/06, Tom Warren [EMAIL PROTECTED] wrote:
Hi Gonzalo!
On 10/11/06, Gonzalo Arana [EMAIL PROTECTED] wrote:
Hi,
What kinds of errors are you getting? Timeouts?
It depends on which browser and which site, but with MSN we're getting
'connection reset' on Firefox and 'page cannot
Hi,
What kinds of errors are you getting? Timeouts?
Anything in your cache.log?
Try to get the access.log lines that represent the problem.
Here is a minimal troubleshooting guide:
1) Try to access them without squid, using any web browser (explorer, firefox).
2) Try to access using elinks
Did you run w3m in you squid box?
On 10/11/06, Jens Strohschnitter [EMAIL PROTECTED] wrote:
out users can't get an access to the url: http://www.bundessortenamt.de
through the squid-proxy (Version 2.5.STABLE5). In the logs I can't find
any error. When I try to access the site without
On 8/31/06, chima s [EMAIL PROTECTED] wrote:
Hi,
I am using FC2-Linux-2.6.10 Kernel in
dual P4 Xeon Processer
3 GB Physical RAM
2 * 12 GB disked Cache_dir with different SCSI harddisk.
OS File Systen - reiserfs
Squid- 2.6.STABLE2 with
cache_replacement_policy heap LFUDA
Stefano,
As far as I know, dstdomain acls are stored in a splay tree, so there
is no linear search on this.
Regards,
On 8/24/06, Stefano Fraccaro [EMAIL PROTECTED] wrote:
Hey all,
I've SquidNT 2.6 Stable3 with very large porn domains file: 600.000
sites blocked with dstdomain acl. When i try
(I did reply only to Marco, sory, shifting back to the list).
On 8/10/06, Marco Berizzi [EMAIL PROTECTED] wrote:
Gonzalo Arana wrote:
Hi Marco,
Ciao Gonzalo.
Just for the record, epoll works just fine on debian sarge (libc
2.3.2) with vanilla kernel 2.6.17.1
mhhh...
Sounds to me that you
On 8/10/06, Marco Berizzi [EMAIL PROTECTED] wrote:
Gonzalo Arana wrote:
[EMAIL PROTECTED]:/usr/src/linux-2.6.17.8# grep -ir CONFIG_EPOLL .config
CONFIG_EPOLL=y
This may sound silly, but is it the actual kernel running?
Yes it is the running kernel. However CONFIG_EPOLL=y
cannot be even
Hi,
check for ulimit -n, If it's 1024, modify squid init script rising
this value (check /proc/sys/fs/file-max as well).
Regards,
On 8/10/06, Stuart J. Newman [EMAIL PROTECTED] wrote:
I am running squid 2.5.STABLE3 from Redhat in Redhat Enterprise Linux 3. I have
followed the instructions
17:48 -0300 skrev Gonzalo Arana:
Hi,
I am a little puzzled by squid 2.6 behaviour: Authentication header
is not forwarded to origin servers when squid is in accelerator mode.
See the cache_peer directive.
Regards
Henrik
--
Gonzalo A. Arana
Hi,
I am a little puzzled by squid 2.6 behaviour: Authentication header
is not forwarded to origin servers when squid is in accelerator mode.
My squid.conf says something like this:
http_port 80 vhost
cache_peer host1 80 0 proxy-only origin-server
acl my_domain dstdomain .my.tld
28 matches
Mail list logo
