performance of jailed processes

[Alexander Leidinger]

On Wed, 31 Mar 2004 10:00:14 +0200 (CEST)
Martin Blapp <mb at imp.ch> wrote:

> 
> Hi DES,
> 
> > Specifically, running a query against a remote MySQL server from
> > inside a jail takes an order of magnitude more time than from outside
> > the jail.  Tcpdump shows that the TCP packets carrying the result are
> > evenly spaced, so this is not a matter of the server timing out on a
> > DNS lookup or anything like that.
> 
> Could you try the following things ? Use libthr or libpthreads for mysql and
> see if the results are still the same. Do the same test with linux-threads
> (all our mysqls on 4.x with jails use linux-threads and we don't see any
> problems)

I just had a similar problem with MySQL on a system without jails
(-current as of Mar 19), but with one IP-alias. Feeding the initial
content of a DB from a 4.6 MB file needed too long (I aborted after two
minutes). Binding MySQL to the main IP only didn't changed this. I
recompiled with linux-threads and it finished importing the data after
20 seconds.

I can't test more on this machine, since it is scheduled to go into
production today or tomorrow, but I could add mysql to a test machine
(with a more recent -current) and use the same file there if needed.

Bye,
Alexander.

-- 
                    I'm available to get hired.

http://www.Leidinger.net                       Alexander @ Leidinger.net
  GPG fingerprint = C518 BC70 E67F 143F BE91  3365 79E2 9C60 B006 3FE7