gump-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Leo Simons <m...@leosimons.com>
Subject Re: Brutus may have more downtime
Date Fri, 25 Mar 2005 16:42:44 GMT
On 25-03-2005 17:25, "Adam R. B. Jack" <ajack@apache.org> wrote:
>> Ehm, no, I don't think so. And we actually gained some memory. I'm hoping
>> that gump will be completely moving onto several vmware instances and some
>> solaris zones. (FYI, the solaris machine has stuff like 32GB of RAM. It's
> an
>> animal). That will free brutus up for other stuff.
> 
> That works for me. [I don't think Gump will "like" being on a VM (VMWare
> don't push their stuff for use cases that like to use 100% resources) but I
> think it is one Gump ought take for the team. It can move slower, and/or we
> can optimise it further.]

Well, I've been looking at gump perform, and most of the time, it seems like
IO wait is really the problem. We're really burning up those disks. The good
bit here is that I believe IO isn't that much slower on VMWare ESX.

> Most of re-install Gump ought be trivial, but I do worry about things like
> Dims' mono build, and such. Hopefully w/ a backup we (at least) have a crib
> sheet.

Right now I'm backing up /, /usr and /home to a firewire disk we have
attached to brutus. The only stuff I'm not backing up (provided I get the
commands right) is the actual gump workspaces.

> Thanks for all your efforts Leo, especially the recent extra infra work
> you've been putting in.

I'll relay that to the 10+ people that have been losing sleep over our poor
little mail server! :-D

Cheers,

Leo



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@gump.apache.org
For additional commands, e-mail: general-help@gump.apache.org


Mime
View raw message