MySQL Lists are EOL. Please join:

List:Commits« Previous MessageNext Message »
From:Sergei Golubchik Date:May 14 2009 9:56am
Subject:Re: bzr commit into mysql-5.0-bugteam branch (satya.bn:2734)
Bug#37408
View as plain text  
Hi, satya!

please don't forget to cc: the list when replying.

On May 14, satya wrote:
> Hi Serg,
>
> I reproduced the bug. The following are my observations.
>
> Old Kernel                                                new kernel
> ==========                                       ==========
> swaps memory when compressed                      Same as the other but
> tables of size greater than                       the process kswapd0
> the ram size used.                                utilization is almost zero
> The process kswapd0 utiliztion                    (0.1 to 0.9%)
> is almost 30 -40% eating the cpu                  (mmaping)                          
>         
>              (mmaping)
> swap memory used - 160MB                          swap memory used - 100MB
>
> when normal tables are read,                      Same as the other
> there is almost no swapping.
> Very minimal <2MB of swapping
> occured. Might be due to other
> reason.
>
> Please let me know how to proceed.

okay, in that case, I suppose it's ok to add an option. but make it this
way:

    myisam-mmap-size=<number>

or even 

    mmap-memory-size=<number>

and it'll allow one to specify


   --mmap-memory-size=200M

for example, and *all* mmaps (not only compressed tables) will be
allow to use no more than 200M in total.

> Also I have a question regarding other
> BUG#10206 - Transaction requiring  Max_BinLog_Cache_size > 4GB always 
> rollsback
>
> As of now when transactions reach the max_binlog_cache_size they
> always fail saying "the multi transaction has the reached the
> max_binlog_cache_size, please increase.." The maximum it can be
> increased is 4GB on 32 bit.
>
> Is it that we don't want the transactions to fail when they reach the
> max_binlog_cache_size?  Please let me know how to proceed on this bug.

No, we want to be able to set max_binlog_cache_size to values larger
than 4GB.

Regards / Mit vielen Grüßen,
Sergei

-- 
   __  ___     ___ ____  __
  /  |/  /_ __/ __/ __ \/ /   Sergei Golubchik <serg@stripped>
 / /|_/ / // /\ \/ /_/ / /__  Principal Software Engineer/Server Architect
/_/  /_/\_, /___/\___\_\___/  Sun Microsystems GmbH, HRB München 161028
       <___/                  Sonnenallee 1, 85551 Kirchheim-Heimstetten
Geschäftsführer: Thomas Schroeder, Wolfgang Engels, Wolf Frenkel
Vorsitzender des Aufsichtsrates: Martin Häring
Thread
bzr commit into mysql-5.0-bugteam branch (satya.bn:2734) Bug#37408Satya B27 Apr
  • Re: bzr commit into mysql-5.0-bugteam branch (satya.bn:2734)Bug#37408Sergei Golubchik28 Apr
    • Re: bzr commit into mysql-5.0-bugteam branch (satya.bn:2734)Bug#37408Sergei Golubchik7 May
      • Re: bzr commit into mysql-5.0-bugteam branch (satya.bn:2734) Bug#37408satya7 May
        • Re: bzr commit into mysql-5.0-bugteam branch (satya.bn:2734)Bug#37408Sergei Golubchik7 May
          • Re: bzr commit into mysql-5.0-bugteam branch (satya.bn:2734) Bug#37408satya8 May
Re: bzr commit into mysql-5.0-bugteam branch (satya.bn:2734)Bug#37408Sergei Golubchik14 May