NetBSD-Users archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: Network very very slow... was iSCSI and jumbo frames
RVP a écrit :
> On Wed, 3 Feb 2021, BERTRAND Joël wrote:
>
>> raid0 and dk0 (raid1) are local disks, dk5 is my qNAP NAS.
>
> Of course. You mentioned it was iSCSI.
>
>> 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)
>
> This is OK.
>
>> 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)
>>
>> 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 retry with `iflag=direct oflag=creat,direct' added?
Of course.
legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=100 bs=1m
iflag=direct oflag=creat,direct
100+0 records in
100+0 records out
104857600 bytes transferred in 0.315 secs (332881269 bytes/sec)
legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=1000 bs=1m
iflag=direct oflag=creat,direct
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 53.515 secs (19594057 bytes/sec)
legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=10000 bs=1m
iflag=direct
oflag=creat,direct
legendre# dd if=/dev/zero of=/opt/bacula/test.dd count=10000 bs=1m
iflag=direct
oflag=creat,direct
10000+0 records in
10000+0 records out
10485760000 bytes transferred in 1040.933 secs (10073424 bytes/sec)
During last dd, I have done a top in qNAP terminal :
Mem: 953140K used, 71244K free, 0K shrd, 398780K buff, 216644K cached
Load average: 0.70, 0.67, 0.76 (State: S=sleeping R=running, W=waiting)
PID USER STATUS RSS PPID %CPU %MEM COMMAND
726 admin SW 0 2 0.8 0.0 md1_raid6
6882 admin SW< 0 2 0.5 0.0 QNAP_fileio_2
15949 admin SW 0 2 0.5 0.0 iscsi_trx
595 admin S 9768 1 0.4 0.9 hal_daemon
28745 admin Z N 0 9648 0.4 0.0 disk_manage.cgi
28746 admin Z N 0 9648 0.3 0.0 disk_manage.cgi
15677 admin S 17M 1 0.2 1.7 python
10912 admin S 3600 1 0.2 0.3 rfsd_qmonitor
12520 admin SW 0 2 0.2 0.0 kworker/u8:2
28768 admin R 5924 9648 0.1 0.5 chartReq.cgi
27344 admin R 1644 7189 0.1 0.1 top
38 admin SW 0 2 0.1 0.0 kswapd0
15948 admin SW 0 2 0.1 0.0 iscsi_ttx
27804 admin SW 0 2 0.1 0.0 kworker/u8:3
3796 admin S 28M 1 0.0 2.8 python
5824 admin S 17M 1 0.0 1.7 python
7813 admin S 17M 1 0.0 1.7 ncdb
7865 admin S 13M 1 0.0 1.3 ncd
19292 admin S 10M 1 0.0 1.0 bcclient
5670 admin S 10M 5425 0.0 1.0 gwd
8836 admin S < 10160 9949 0.0 0.9 apache_proxy
5668 admin S 10116 5425 0.0 0.9 npm
5425 admin S 9720 1 0.0 0.9 nmd
5673 admin S 8884 5425 0.0 0.8 slaacrrd
5674 admin S 8704 5425 0.0 0.8 bdmd
5676 admin S 8376 5425 0.0 0.8 dnsmasqd
5671 admin S 7776 5425 0.0 0.7 ddnsd
5432 admin S 7744 5425 0.0 0.7 qserviced
9699 admin S 7344 1 0.0 0.7 php-fpm-proxy
5431 admin S 7164 5425 0.0 0.6 ncaas
5672 admin S 6768 5425 0.0 0.6 ressd
10385 admin S 6740 1 0.0 0.6 apache_proxys
5354 admin S 6584 1 0.0 0.6 ifd
5681 admin S 6528 5672 0.0 0.6 ressd
17713 admin S 6052 1 0.0 0.5 dhclient
17423 admin S 6000 1 0.0 0.5 dhclient
15705 admin S 5948 10385 0.0 0.5 apache_proxys
5675 admin S 5864 5425 0.0 0.5 interfaces_moni
9949 admin S < 5608 1 0.0 0.5 apache_proxy
7098 admin S 5596 1 0.0 0.5 qpkgd
7016 admin S 5096 10380 0.0 0.4 sshd
13739 admin S 5032 1 0.0 0.4 qsnapman-smart
10380 admin S 4968 1 0.0 0.4 sshd
14151 admin S 4852 1 0.0 0.4 qsnapman-recyc
13027 admin S 4748 1 0.0 0.4 qsnapman-alive
18328 admin S 4652 1 0.0 0.4 upnpcd
12265 admin S 4508 1 0.0 0.4 qsnapman
11980 admin S 4452 1 0.0 0.4 qLogEngined
12236 admin S 4276 1 0.0 0.4 qstorman
9700 admin S 3948 9699 0.0 0.3 php-fpm-proxy
9701 admin S 3948 9699 0.0 0.3 php-fpm-proxy
I have found a 4GB DDR3L. This afternoon, I will try to replace NAS memory.
On NetBSD side, I obtain with top -It
load averages: 1.08, 0.96, 1.01; up 3+01:19:47 12:09:10
389 threads: 33 idle, 344 sleeping, 4 zombie, 8 on CPU
CPU states:0.0% user, 0.0% nice, 2.0% system, 0.2% interrupt, 97.7% idle
Memory: 1527M Act,2729M Inact,16M Wired,95M Exec, 3047M File, 8579M Free
Swap: 16G Total, 16G Free
PID LID USERNAME PRI STATE TIME WCPU CPU NAME COMMAND
0 192 root 124 syncer/7 76:45 0.93% 0.93% ioflush [system]
1141 16 root 85 nfsd/6 10:14 0.05% 0.05% slave nfsd
2134 1 bertrand 43 CPU/2 0:00 0.00% 0.00% - top
0 44 root 0 CPU/5 0:00 0.00% 0.00% idle/5 [system]
0 38 root 0 CPU/4 0:00 0.00% 0.00% idle/4 [system]
0 32 root 0 CPU/3 0:00 0.00% 0.00% idle/3 [system]
0 20 root 0 CPU/1 0:00 0.00% 0.00% idle/1 [system]
0 2 root 0 CPU/0 0:00 0.00% 0.00% idle/0 [system]
0 56 root 0 CPU/7 0:00 0.00% 0.00% idle/7 [system]
0 50 root 0 CPU/6 0:00 0.00% 0.00% idle/6 [system]
With top (without option) :
load averages: 1.11, 0.98, 1.01; up 3+01:20:46 12:10:09
106 processes: 104 sleeping, 2 on CPU
...
PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND
26833 root 85 0 18M 2460K pager_/2 0:07 2.25% 2.25% dd
...
JKB
Home |
Main Index |
Thread Index |
Old Index