Strange memcached behavior on stress test

[Stephen Woodbridge]

Hi Paul,

No this was 32bit linux with 6 GB of memory and the HUGEMEM kernel patch 
and a 1 GB memcache.

Did you find out why this happened?
Did it go away and stay away after a restart?
How long do you typically run your memcached without a restart?

This is a little worrisome, because our production boxes typically run 
300-500 days without a reboot.

-Steve

Paul T wrote:
> 
> Is it happening on 64bit linux by any chance?
> 
> We got the similiar trouble once - after running
> memcached instance for 3 weeks in a row. 
> 
> Rgds.Paul.
> 
> --- Stephen Woodbridge <woodbri at swoodbridge.com>
> wrote:
> 
>> Hi all,
>>
>> I asked a question in my email about the stress test
>> script, but I think 
>> it might have gotten lost in the other information I
>> posted. So I'll ask 
>> again.
>>
>> Running the stress test with multiple clients
>> eventually starts to 
>> generate a lot of failures. After pushing a large
>> number of GB through 
>> the cache (like 64 GB+, don't remember the exact
>> number) we started 
>> getting a lot for failures on the test clients, like
>> one was failing 30% 
>> and another failing 60% of its sets. Restarting the
>> server cleared the 
>> problem until we pushed a lot of data through the
>> cache again.
>>
>> So what is going on? Is this the cache fragmentation
>> problem? Is there a 
>> fix for this problem? Other thoughts?
>>
>> Thanks,
>>    -Steve
>>
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com 
>