Slack & file transfers
 

Hi guys!

I just now found a somewhat odd behaviour
of my Linux installation at work... if I write large
files to the hard-drive (either off a nfs-mounted
directory or moving them on the local hard-drive)
the machine seems to lock up for several seconds.

Has anyone ever had such behaviour? it didn't do that
with win2k, but my boxes at home and the notebook
don't seem to have this odd phenomenon. I find it really
annoying, since it delays typing (and display), and the
mouse just freezes in spot while in X.

I'd have thought that a P IV 1.6 could handle that? ;)

Cheers,
Tink

try hdparm -u 1 /dev/<device_name> . man hdparm for more info :) . This is only valid for IDE devices.

Either your system is really low in memory and has to swap even for simple things like moving the mouse around (unlikely), or your chipset / hardware configuration with IDE is causing serious problems. Need more information on the machine you are running on and kernel you are using.

Hi there...

if you need any particular other info just tell me what :}

Cheers,
Tink

it's a Compaq Evo, P IV 1.6 GHz, 256 MB,

interrupts:
CPU0
0: 76032 XT-PIC timer
1: 4088 XT-PIC keyboard
2: 0 XT-PIC cascade
5: 11413 XT-PIC eth0
8: 1 XT-PIC rtc
11: 0 XT-PIC Intel ICH2
12: 29940 XT-PIC PS/2 Mouse
14: 266551 XT-PIC ide0
15: 4 XT-PIC ide1
NMI: 0
ERR: 0

ioports:
0000-001f : dma1
0020-003f : pic1
0040-005f : timer
0060-006f : keyboard
0070-007f : rtc
0080-008f : dma page reg
00a0-00bf : pic2
00c0-00df : dma2
00f0-00ff : fpu
0170-0177 : ide1
01f0-01f7 : ide0
02f8-02ff : serial(auto)
0376-0376 : ide1
03c0-03df : vga+
03f6-03f6 : ide0
03f8-03ff : serial(auto)
0cf8-0cff : PCI conf1
1000-1fff : PCI Bus #02
1000-103f : Intel Corp. 82820 (ICH2) Chipset Ethernet Controller
1000-103f : eepro100
2000-20ff : Intel Corp. 82820 820 (Camino 2) Chipset AC'97 Audio Controller
2000-20ff : Intel ICH2
2400-243f : Intel Corp. 82820 820 (Camino 2) Chipset AC'97 Audio Controller
2400-243f : Intel ICH2
2440-245f : Intel Corp. 82820 820 (Camino 2) Chipset USB (Hub A)
2460-247f : Intel Corp. 82820 820 (Camino 2) Chipset USB (Hub B)
2480-248f : Intel Corp. 82820 820 (Camino 2) Chipset IDE U100
2480-2487 : ide0
2488-248f : ide1

pci:

PCI devices found:
Bus 0, device 0, function 0:
Host bridge: Intel Corp. 82845 845 (Brookdale) Chipset Host Bridge (rev 3).
Prefetchable 32 bit memory at 0xf8000000 [0xfbffffff].
Bus 0, device 1, function 0:
PCI bridge: Intel Corp. 82845 845 (Brookdale) Chipset AGP Bridge (rev 3).
Master Capable. Latency=64. Min Gnt=14.
Bus 0, device 30, function 0:
PCI bridge: Intel Corp. 82820 820 (Camino 2) Chipset PCI (rev 18).
Master Capable. No bursts. Min Gnt=6.
Bus 0, device 31, function 0:
ISA bridge: Intel Corp. 82820 820 (Camino 2) Chipset ISA Bridge (ICH2) (rev 18).
Bus 0, device 31, function 1:
IDE interface: Intel Corp. 82820 820 (Camino 2) Chipset IDE U100 (rev 18).
I/O at 0x2480 [0x248f].
Bus 0, device 31, function 2:
USB Controller: Intel Corp. 82820 820 (Camino 2) Chipset USB (Hub A) (rev 18).
IRQ 11.
I/O at 0x2440 [0x245f].
Bus 0, device 31, function 4:
USB Controller: Intel Corp. 82820 820 (Camino 2) Chipset USB (Hub B) (rev 18).
IRQ 10.
I/O at 0x2460 [0x247f].
Bus 0, device 31, function 5:
Multimedia audio controller: Intel Corp. 82820 820 (Camino 2) Chipset AC'97 Audio Controller (rev 18).
IRQ 11.
I/O at 0x2000 [0x20ff].
I/O at 0x2400 [0x243f].
Bus 1, device 0, function 0:
VGA compatible controller: nVidia Corporation Vanta [NV6] (rev 21).
IRQ 10.
Master Capable. Latency=64. Min Gnt=5.Max Lat=1.
Non-prefetchable 32 bit memory at 0xf7000000 [0xf7ffffff].
Prefetchable 32 bit memory at 0xfc000000 [0xfdffffff].
Bus 2, device 8, function 0:
Ethernet controller: Intel Corp. 82820 (ICH2) Chipset Ethernet Controller (rev 3).
IRQ 5.
Master Capable. Latency=66. Min Gnt=8.Max Lat=56.
Non-prefetchable 32 bit memory at 0xfe200000 [0xfe200fff].
I/O at 0x1000 [0x103f].

hda:
MAXTOR 6L020J1

hdc (cdrom):
Compaq CRD-8484B

If you have a IDE disk why not enable DMA?
hdparm -d 1 /dev/hd? where ? is your drive letter.
THIS CAN POSSIBLY BE DANGEROUS. Make sure your disk can handle it before doing this. I did it once and couldn't boot my system beacause "Drive geometry is out of range" error, so make sure your disk can handle it. I beleive all newer disks should have no problems.
Run hdparm -t /dev/hd? before and after /hdparm -d 1 /dev/hd? to see what a speed increase you'll get!! I got a increase from 15mb/sec to 42mb/sec. Now thats sweet!
Hope that helps
-NSKL

This seems a little odd to me. Changing the hdparm will not affect timing of NFS reads and writes and the problem is reproducable even over a NFS share. Even if you disable DMA mode you still shouldn't notice a delay.

A couple of places to start looking:
Are there any error messages when you type dmesg? You might also try looking in /var/log/messages for anything unusual. Another thing to keep in mind is whether you're using a non-standard filesystem like reiserfs or ext3. These filesystems perform differently and the version you have may contain a bug. It may also have something to do with you're trying to read data from a bad sector on your hard disk in which case you would notice errors in /var/log/messages

Hi Griffin.

Nope, only one informational message about unknown bridge ressources,
and assuming transparent.

Looks fine to me, was the first place I was looking, too ;)

Hmmm ... using Reiser that comes with Slack (or rather 2.4.18). Using that on 4 machines, and only
the workstation at work (the supposedly fastest and best machine) shows this odd behaviour. My old PII-266 notebook on the same network runs smooth.

Possible ... :( I'll read up on hdchecks on Reiser.

Thanks a lot m8.


Cheers,
Tink