Re: Network very very slow... was iSCSI and jumbo frames

[Paul Ripke <stix%stix.id.au@localhost>]

On Wed, Feb 03, 2021 at 11:04:02AM +0100, BERTRAND Jo?l wrote:
> Paul Ripke a ?crit?:
> >> legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=10 bs=10m
> >> 10+0 records in
> >> 10+0 records out
> >> 104857600 bytes transferred in 0.537 secs (195265549 bytes/sec)
> >> legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=10 bs=100m
> >> 10+0 records in
> >> 10+0 records out
> >> 1048576000 bytes transferred in 53.396 secs (19637725 bytes/sec)
> >> legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=10 bs=1000m
> >> 10+0 records in
> >> 10+0 records out
> >> 10485760000 bytes transferred in 1026.927 secs (10210813 bytes/sec)
> >> legendre#
> > 
> > How much RAM does this system have?
> 
> 	NetBSD : 16GB
> 	NAS : 1GB but I think I have some 4 or 8GB DDR3L somewhere.
> 
> > I'm guessing you might be paging in
> > those last two operations. Instead, try:

Hmm. No, definitely shouldn't be paging.

> > dd if=/dev/zero of=/opt/bacula/test.dd count=100 bs=1m
> 
> legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=100 bs=1m
> 100+0 records in
> 100+0 records out
> 104857600 bytes transferred in 0.313 secs (335008306 bytes/sec)
> 
> > dd if=/dev/zero of=/opt/bacula/test.dd count=1000 bs=1m
> 
> legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=1000 bs=1m
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 53.680 secs (19533830 bytes/sec)
> 
> > dd if=/dev/zero of=/opt/bacula/test.dd count=10000 bs=1m
> legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=10000 bs=1m
> 10000+0 records in
> 10000+0 records out
> 10485760000 bytes transferred in 1046.321 secs (10021551 bytes/sec)

Can you add 'oflag=direct', to the dd, making sure that the destination
file exists? Looks like there's some client-side caching going on.

> On NAS side, during last dd, I note :
> - iops (on one disk) : 39/s
> - latency : 3 ms
> - cpu : less than 10%
> - free memory : 500 MB.
> 
> 	JKB

-- 
Paul Ripke
"Great minds discuss ideas, average minds discuss events, small minds
 discuss people."
-- Disputed: Often attributed to Eleanor Roosevelt. 1948.