Re: [PATCH] StoreID stage 2\3 fully functional StoreID interface.

From: Amos Jeffries <squid3_at_treenet.co.nz>
Date: Thu, 13 Dec 2012 14:54:53 +1300

On 13.12.2012 13:40, Eliezer Croitoru wrote:
> On 12/13/2012 1:55 AM, Amos Jeffries wrote:
>> On 13.12.2012 12:34, Eliezer Croitoru wrote:
>>> I worked on the old code and was reading introduction to c++ to
>>> understand better things in the code and it really gave me a lot.
>>> Now I got some problem while using concurrency with store_id
>>> helper.
>>>
>>> If I use concurrency and even on "1" I get assertion problem in the
>>> cache log as:
>>> 2012/12/13 01:26:56 kid2| assertion failed:
>>> ../../12500/src/HttpHeader.cc:1271: "name && value"
>>> 2012/12/13 01:26:56 kid1| assertion failed:
>>> ../../12500/src/HttpHeader.cc:1271: "name && value"
>>>
>>> Which basically is this code:
>>>
>>>
>>> http://bazaar.launchpad.net/~squid/squid/trunk/view/head:/src/HttpHeader.cc#L1289
>>>
>>>
>>> For this specific branch I am using is 12500 so I will later update
>>> it.
>>> Was there anything that fixed or could have fixed this kind of
>>> issue?
>>
>> What is the helper line being returned?
> Same as always but in the new format:
> request VS response id with full answer:
> 1 http://i1.ytimg.com/vi/H-0OuRVCAK8/default.jpg 192.168.10.100/- -
> GET myip=192.168.10.201 myport=3128
> 1 OK store-id=http://ytimg.squid.internal/vi/H-0OuRVCAK8/default.jpg
>
> What do you think about?

putExt() is used internally by the Notes objects being converted to
NotePair for logging use. name or value being NULL is a problem and
indicates that something is being added as a Note which is not a kv-pair
format.

Are you able to get me a backtrace from the assert please?

Amos
Received on Thu Dec 13 2012 - 01:54:57 MST

This archive was generated by hypermail 2.2.0 : Thu Dec 13 2012 - 12:00:11 MST