Re: [squid-users] Squid work consideration

2011-11-05 Thread Benjamin

 Hi Amos,


While we are looking for cache_dir consideration , does a single big 
directory is better or multiple directory with good amount of size is 
better while we are looking for performance ?


And i tried to find from internet that which is better for cache_dir 
file system ext3 / ext4 / reseizerfs ( in terms of heavy loaded systems ) ?



Whenever i m trying to find how many squid processes are running i got,

pgrep squid
8311
8313

it means squid always has 2 squid processes. Can we increase it for high 
performance.?



Thanks,
Benjo



On Wed, 26 Oct 2011 22:02:47 +0530, benjamin fernandis wrote:

Hi All,

I have some basic queries which i would like to clear from your
suggestions and knowledge sharing.

1) How squid identity web object size ?

My understanding : Squid identifies web object size by http headers
for that web object

M i right ?



Headers are optional and only determine future size expectations. For 
optimal calculation and placement before the transfer starts.


Data actually transfered during the response processing is used for 
all objects. Including those without size headers, which have to be 
assumed to be very large/infinite until they are complete.




2) Suppose i define maximum_object_size on disk is 4 GB. then as per
squid consideration , squid will store web objects which are = 4
GB.so how to validate that this object  which is = 4GB is stored on
disk by squid for further cache gain?


Disk is optional. A 16GB RAM cache might store 4 of them without any 
disk.


There are several ways to locate an object:
 * use tools such as redbot.org to see if the object is actually 
storable.
 * use a HEAD (or GET) request, and see if the response was a HIT or 
MISS.
 * locate it in the cachemgr objects report. (maybe a very long, 
slow mgr response).
 * scan the swap.state journals to see if its had SWAPOUT done but not 
RELEASE.
 * use the purge tool to scan the disk cache object by object and see 
if it is there.


Amos





Re: [squid-users] Squid work consideration

2011-11-05 Thread Amos Jeffries

On 6/11/2011 12:50 a.m., Benjamin wrote:

 Hi Amos,


While we are looking for cache_dir consideration , does a single big 
directory is better or multiple directory with good amount of size is 
better while we are looking for performance ?




Multiple physical disks with cache_dir on each is best. That offers the 
most parallal reads/writes. Unless you are adding COSS or Rock storage 
cache_dir its not worth placing multiple cache_dor on one HDD. Total 
size of any given UFS cache_dir seems not to matter as much as 
parallelism on the disk I/O.


And i tried to find from internet that which is better for cache_dir 
file system ext3 / ext4 / reseizerfs ( in terms of heavy loaded 
systems ) ?




Disabling the journalling actions and atime updating is best for all of 
the above. I have not seen any recent speed benchmarks comparing them. 
If you base speed on the more generic benchmarks be aware Squid disk 
access has a unusually high portion of writes.





Whenever i m trying to find how many squid processes are running i got,

pgrep squid
8311
8313

it means squid always has 2 squid processes. Can we increase it for 
high performance.?


One of those will be the master process. One will be the actual worker 
process.


You can run as many pairs of mster+worker as you like on a box. Though 
we dont recommend running more than one worker per CPU.
See http://wiki.squid-cache.org/MultipleInstances for the relevant 
config details for 3.1 and older. If you have squid-3.2 there is a link 
there to the SMP support page.



Amos



[squid-users] How can I configure squid3 proxy to allow rtmp streaming traffic?

2011-11-05 Thread Andre Ferreira
How can I configure squid3 proxy to allow rtmp streaming traffic?
Video sites like Youtube work, but most rtmp sites timeout with server
not found error.

For example in 
http://www.tvi.iol.pt/mediacenter.html?mul_id=13507798load=1pagina=1pos=2
the ad shows ok, then the real streaming should start, but doesn't.
I already added 1935 to the Safe_Ports acl.

Thank you


Re: [squid-users] How can I configure squid3 proxy to allow rtmp streaming traffic?

2011-11-05 Thread Amos Jeffries

On 6/11/2011 2:36 a.m., Andre Ferreira wrote:

How can I configure squid3 proxy to allow rtmp streaming traffic?
Video sites like Youtube work, but most rtmp sites timeout with server
not found error.

For example in 
http://www.tvi.iol.pt/mediacenter.html?mul_id=13507798load=1pagina=1pos=2
the ad shows ok, then the real streaming should start, but doesn't.
I already added 1935 to the Safe_Ports acl.

Thank you


RTMP is a wholly different protcol to HTTP.  Squid only accepts HTTP and 
ICY protocols as input at present.


*if* your media agent is configured to use an HTTP proxy and supports 
the CONNECT method to transfer the stream. You may be able to get it to 
work by adding the RTMP port(s) to both Safe_ports and SSL_ports.


Amos


Re: [squid-users] Reverse Proxy + Exchange OWA issue

2011-11-05 Thread Rick Chisholm
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Thanks Amos - I will try the configuration change first and look into
the new version.

On 11-11-04 8:44 PM, Amos Jeffries wrote:
 On 5/11/2011 5:03 a.m., Rick Chisholm wrote:
 I'm not intimidated by that option, but I prefer to use ports
 (which is still compiling from source FWIW) to make update
 management easier.  Also this is a mission critical production
 box and I can't just blindly throw new versions at it without
 finding out what the issue is and why a certain solution (3.2) is
 the answer.
 
 I would guess the problem is OWA being broken in its use of
 HTTP/1.1. In particular the Expect: 100-continue feature to enable
 better persistent connections. You can try toggling the
 ignore_expect_100 configuration option (OFF is probably needed).
 
 2.7 (and 3.1) only have partial support for that feature. With
 that ignore directive to cause a quick reject or a client timeout
 to happen (5 minutes a coincidence? ). 3.2 is needed to make
 chunking and continue work through Squid (avoiding the timeouts). *
 in particular 3.2.0.8 is the most stable and production usable of
 the 3.2 series.
 
 Amos
 
 


- -- 
Rick Chisholm
sysadmin
Parallel42
e. rchish...@parallel42.ca
m. 519-325-8630
w. www.parallel42.ca
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.8 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk613kgACgkQeE4GKUWrEKtjsgCeK7R0//5Pm96/zjlZdl1qhQXv
Lf8AoI2E8Ev0cZhCiuAYjTy6uX/4tcfx
=6HtK
-END PGP SIGNATURE-