You are not logged in.
Hi all.
I have this problem (~55 secs) with first execution of pacman in a session (I think, at least since six months so far, but I'm not sure). I tried both with /var mounted in tmpfs, both with no special mountpoint, into the same / partition in ext4, but nothing changes. I have no connection problems, and time loss is all in local work.
First execution in a session:
[root@Archbox javier]# time pacman -Syu
:: Sincronizzazione dei database in corso...
core è aggiornato
extra è aggiornato
community è aggiornato
:: Aggiornamento del sistema in corso...
Non c'è nulla da fare
real 0m55.001s
user 0m0.227s
sys 0m0.727sSecond
[root@Archbox javier]# time pacman -Syu
:: Sincronizzazione dei database in corso...
core è aggiornato
extra è aggiornato
community è aggiornato
:: Aggiornamento del sistema in corso...
Non c'è nulla da fare
real 0m1.680s
user 0m0.157s
sys 0m0.083s95% of time is spent in
debug: loading package cache for repository 'extra'
debug: database path for tree extra set to /var/lib/pacman/sync/extra/
debug: loading package cache for repository 'community'
debug: database path for tree community set to /var/lib/pacman/sync/community/My hd is 6 years old, and sometimes I have also firefox freezes reading from it. Tell me what kind of test could I do to report to you hd status.
P.S.
I know formatting /var in reiserfs could markably increment performances in a partition full of small files, but first of all I want to know if this is the expected behaviour of pacman with /var in ext4. I tried pacman-optimize too, but it works just one time. Successive upgrades are slow again.
Thanks in advance.
Last edited by 4javier (2011-04-08 10:20:56)
Offline
This is because pacman has to read many small files from the hard-drive. Wait for pacman-3.5 which will speed this up substantially.
Offline
Until then, try using pacman-cage. It has considerably sped up pacman performance for me.
Offline
Thanks guys.
So, do you think this timings are normal?
@bzt
I prefer to keep a core-app like pacman as smooth as possibile.
I'll not mark this thread as solved, waiting for pacman 3.5
Last edited by 4javier (2011-02-03 14:00:47)
Offline
Thanks guys.
So, do you think this timings are normal?
For now, this *is* normal. This is not a bug, this is just how the things are.
This is about to change http://mailman.archlinux.org/pipermail/ … 18136.html
Offline
Thanks to you too.
I'll wait patiently for 3.5 ![]()
Offline
I forgot to update this thread two weeks ago. Thanks guys, last pacman version works as you mentioned.
time pacman -Syu
:: Sincronizzazione dei database in corso...
core è aggiornato
extra è aggiornato
community è aggiornato
testing è aggiornato
:: Aggiornamento del sistema in corso...
non ci sono aggiornamenti disponibili
real 0m8.180s
user 0m0.177s
sys 0m0.070sNow I can mark this thread as solved. Bye ![]()
Offline