Mailinglist Archive: opensuse-bugs (19817 mails)

< Previous Next >
[Bug 390219] ext3 performance is extremely bad
  • From: bugzilla_noreply@xxxxxxxxxx
  • Date: Wed, 14 May 2008 16:29:07 -0600 (MDT)
  • Message-id: <20080514222907.8171024538D@xxxxxxxxxxxxxxxxxxxxxx>

User captain.magnus@xxxxxxxxxxxx added comment

Magnus Boman <captain.magnus@xxxxxxxxxxxx> changed:

What |Removed |Added
Info Provider|captain.magnus@xxxxxxxxxxxx |

--- Comment #2 from Magnus Boman <captain.magnus@xxxxxxxxxxxx> 2008-05-14
16:29:07 MST ---
All partitions are mounted with defaults, so assuming that ordering is not
default, no;

mboman@mblxsrv01:~> cat /etc/fstab
/dev/disk/by-id/scsi-SATA_SAMSUNG_HD501LJS0MUJDSP900849-part1 swap
swap defaults 0 0
/dev/disk/by-id/scsi-SATA_Maxtor_6Y120M0_Y3L2LLXE-part2 swap
swap defaults 0 0
/dev/disk/by-id/scsi-SATA_SAMSUNG_HD501LJS0MUJDSP900849-part7 /
ext3 acl,user_xattr 1 1
proc /proc proc defaults 0 0
sysfs /sys sysfs noauto 0 0
debugfs /sys/kernel/debug debugfs noauto 0 0
usbfs /proc/bus/usb usbfs noauto 0 0
devpts /dev/pts devpts mode=0620,gid=5 0 0
/dev/sda6 /data ext3 rw 0 0

Neither Beagle nor mandb needs to be running for this to be extremly slow. I
actually think Timo Hoenig opened a bug about mandb at some stage complaining
that it drags the machine down. Could be related to this as well.

Please suggest some benchmarking tools and I'll be happy to provide the data.
Not sure how exactly one measures response times in applications.

Looking on the web for more explanations I also found that talks
about some issues with the scheduler that didn't make it for our kernel, but
for all I know, it could have been backported already.

I did however turn synchronous OFF for Firefox as per comment 28 in and FF works a little bit
better after that.

Configure bugmail:
------- You are receiving this mail because: -------
You are on the CC list for the bug.

< Previous Next >