A couple of annoying F8t3 things
Rodd Clarkson
rodd at clarkson.id.au
Thu Oct 25 08:42:49 UTC 2007
On Thu, 2007-10-25 at 16:22 +0800, John Summerfield wrote:
> Caolan McNamara wrote:
> > On Thu, 2007-10-25 at 16:56 +1000, Rodd Clarkson wrote:
> >> I've noticed two annoying issues in f8t3.
> >>
> >> 1. Applications seems to take quite a while to 'appear' on the desktop.
> >> After clicking an application on the panel (gedit, gnome-terminal, etc)
> >> it takes about 10 secs for them to appear on the desktop.
> >>
> >> 2. Evolution (particularly compose) seems to 'pause' for 10 seconds or
> >> so every now and then and it becomes unresponsive. Makes typing emails
> >> a pain in the bum (and I've seen it three times in this short email.)
> >>
> >> I'm going to bz these, but we wondering if others have seen this too.
> >
> > I've sometimes seen stuff like that happen with slightly busted
> > networking or something, i.e. strace -f gedit from a terminal and see if
>
> If it's blocking significantly on something in /tmp/, that's serious.
> Those are unix domain sockets, and their traffic is local to the box.
>
> > it is blocking a long time on /tmp/.ICE-unix/XXXX and compare startup
> > time of apps before after after deleting /tmp/.ICE-unix/XXXX
>
> What I was wondering is what size* box is Rodd using, and might
> Evolution 's pauses be while it's saving the work file? Seamonkey does
> that on SL5 (Tikanga-clone), and loses keystrokes. It bugs me.
There's always a pause just after sending an email (with something down
in the status bar that is like saving file) When I'm in compose I don't
see anything showing in the status bar, but it might be the compose
window saving the message in case of a crash (so it can offer it up when
you restart).
>
> CPU, RAM, disk disk speed as measured by "hdparm -t /dev/sda"
Centrino 2.0GHz
1 Gig RAM
[root at localhost ~]# hdparm -t /dev/sda
/dev/sda:
Timing buffered disk reads: 118 MB in 3.02 seconds = 39.09 MB/sec
R.
--
"It's a fine line between denial and faith.
It's much better on my side"
More information about the test
mailing list