Re: Sibling Relation

From: Kashan Sadiq <kashan@dont-contact.us>
Date: Mon, 28 Jun 1999 14:13:57 +0500

Hi Henrik

I m very thankful to u to provide me support and guidance parallel with my work. In fact I didn't learn from this assignment as much I learnt from ur guidance.

Now I prepared another proxy in the premises where proxy2 is placed. And made the same configuration as in proxy1. Both were on the identical LAN. And they worked fine. I surprised to see that the new(suppose proxy3) is continuously communicate with proxy2 as well as proxy1. A little bit TIMEOUT_DIRECTs are there but ignore able.

In my opinion this is some kind of Bandwidth problem or router configuration problem b/w the premises where proxy1 is placed and where proxy2 is placed. What u say?

"In fact I increase the timouts for icp_query, the browsing of users using proxy1 become slow. So is there anyway to eliminate the problem?"
 
Anyway now I can prove to my seniors that there is no problem with my configuration.

Again thanx for co-operation

Kashan

*********** REPLY SEPARATOR ***********

On 6/24/99, at 8:38 PM, Henrik Nordstrom wrote:

>Kashan Sadiq wrote:
>>
>> Hi Henrik,
>>
>> Sorry for bothering u again and again. I have tried the cache
>> digest too. But the same result.
>
>For one thing, in a digest peering there is no such thing as a
>TIMEOUT_DIRECT.
>
>> As u said that on ur squid page apply a patch from there. I saw
>> there for this purpose is "Squid-2.2.STABLE2: ICP timeout selection "
>
>That is another patch. What you are looking for are:
>
>Squid-2.2.DEVEL3: Improved ICP dead peer detection
> Some minor adjustments of ICP dead peer detection to make
> it behave well on startup and low load servers. Also lessens
> the amount of queries sent to dead peers.
>
>> So is it applicable to STABLE3. I think its not.
>
>All patches listes as patches to 2.2.STABLE3 is applicable to
>2.2.STABLE3, even if the patch was made at a earlier release.
>
>> If still there is possibility to overcome this problem please
>> guide me.
>
>Every problem an be overcome once you know what the cause to the problem
>is.
>
>--
>Henrik Nordstrom
>Spare time Squid hacker
Received on Mon Jun 28 1999 - 03:24:54 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:47:02 MST