NFS over LAGG / lacp poor performance

[Gerrit Kühn]

On Fri, 25 Apr 2014 15:55:29 +0100 "Steven Hartland"
<killing at multiplay.co.uk> wrote about Re: NFS over LAGG / lacp poor
performance:

SH> Run top then enable system processes (shift+S) and threads (shift+H)
SH> then disable idle process display (z) if your seeing significant CPU
SH> usage to the igb interrupt handlers try reducing the number.

Hm, I'm a bit unsure what I see here:

---
last pid: 71026;  load averages:  0.36,  0.37, 0.27
up 18+03:14:21  17:48:02
295 processes: 5 running, 259 sleeping, 31 waiting
CPU:  0.0% user,  0.0% nice,  0.0% system,  0.0% interrupt,  100% idle
Mem: 3596K Active, 59M Inact, 11G Wired, 900K Cache, 1236M Buf, 1019M Free
Swap: 16G Total, 14M Used, 16G Free

  PID USERNAME   PRI NICE   SIZE    RES STATE   C   TIME   WCPU COMMAND
    0 root       -16    0     0K  2672K -       2 272:01  0.00% kernel{zio_write_issue_}
    0 root       -16    0     0K  2672K -       0 271:55  0.00% kernel{zio_write_issue_}
    0 root       -16    0     0K  2672K -       3 271:54  0.00% kernel{zio_write_issue_}
    0 root       -16    0     0K  2672K -       1 271:52  0.00% kernel{zio_write_issue_}
 1455 root        20    0  9912K   472K rpcsvc  2 173:53  0.00% nfsd{nfsd: master}
 1455 root        20    0  9912K   472K rpcsvc  0 166:37  0.00% nfsd{nfsd: service}
 1455 root        20    0  9912K   472K rpcsvc  0 164:25  0.00% nfsd{nfsd: service}
 1455 root        20    0  9912K   472K rpcsvc  3 163:06  0.00% nfsd{nfsd: service}
   12 root       -92    -     0K   496K WAIT    1 155:36  0.00% intr{irq262: igb1:que}
[...]
---


The irq-igb1 clibs to top every now and then (always less than 1% cpu
usage though), but there is never much cpu usage shown for any task.
However, the overall load is significant. What might the system be doing?


cu
  Gerrit