| 1 |
If the elevated iowait from iostat is on the host you might be able to |
| 2 |
find something hogging you io bandwidth with iotop. Also look for D |
| 3 |
state procs with ps auxr. Are you on a software raid? |
| 4 |
|
| 5 |
If you are on linux soft raid you might check your disks for errors |
| 6 |
with smartmontools. Other than that the only thing I can think of is |
| 7 |
something like a performance regression in the ide/scsi/sata |
| 8 |
controller (on host or virtual) or mdadm on host. If the host system |
| 9 |
is bogged before starting vmware instances I would suspect the former |
| 10 |
(host controller or mdadm). |
| 11 |
|
| 12 |
On 3/11/10, Stefan G. Weichinger <lists@×××××.at> wrote: |
| 13 |
> Am 11.03.2010 16:54, schrieb Kyle Bader: |
| 14 |
>> If you use the cfq scheduler (linux default) you might try turning off |
| 15 |
>> low latency mode (introduced in 2.6.32): |
| 16 |
>> |
| 17 |
>> Echo 0 > /sys/class/block/<device name>/queue/iosched/low_latency |
| 18 |
>> |
| 19 |
>> http://kernelnewbies.org/Linux_2_6_32 |
| 20 |
> |
| 21 |
> That sounded good, but unfortunately it is not really doing the trick. |
| 22 |
> The VM still takes minutes to boot ... and this after I copied it back |
| 23 |
> to the RAID1-array which should in theory be faster than the |
| 24 |
> noraid-partition before. |
| 25 |
> |
| 26 |
> Thanks anyway, I will test that setting ... |
| 27 |
> |
| 28 |
> Stefan |
| 29 |
> |
| 30 |
> |
| 31 |
> |
| 32 |
|
| 33 |
-- |
| 34 |
Sent from my mobile device |
| 35 |
|
| 36 |
|
| 37 |
Kyle |
Archives