Re: Various COSS minor problems

On Sun, Sep 24, 2006, Guido Serassio wrote:

> AioDone() now is called from StoreAufsDirDone() during squid shutdown
> and calls squidaio_shutdown(). We must be sure that AioDone() is
> executed once after both aufs and COSS shutdown.
> I think that we could add a reference counter incremented after every
> AioInit() call and decremented after any AioDone(). When the counter
> is 0, AioDone is really executed.

I'd just put it in the main shutdown code, after the filesystems have
been flushed and shut down. Its much less pain. Then all you need is a
check to make sure AioDone() only does its thing if AioInit() was called.

> >Hm! Well, you could make squid-z actually write out the correctly-sized
> >file if it isn't correctly-sized..
>
> I like this, it should be more portable.

Yup.

> >Thanks for finding more coss corner-case issues!
>
> This is the funny of the platform porting .... :-)
>
> Just found another little problem:
>
> When squid is executed, the following is printed to the terminal:
> rea:~# /usr/local/squid/sbin/squid -k rotate
> 2006/09/24 10:54:38| COSS: max disk fileno is 6291456
> 2006/09/24 10:54:38| COSS: number of stripes: 3072 of 1048576 bytes each
> 2006/09/24 10:54:38| COSS: number of memory-only stripes 10 of
> 1048576 bytes each
>
> This happens because there are some debug() called to early during
> the configuration parsing.
> This could be very annoying into cron scheduled jobs.

Thats my fault for introducing it. Just bump the log level up to debug(blah, 2)
and it should go away?

Adrian
Received on Mon Sep 25 2006 - 18:29:55 MDT