Skip to Main Content

Infrastructure Software

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

t1000 zones - response time very slow

807559Dec 16 2009 — edited Apr 21 2010
Hi,

I've setup a few zones on a T1000 and I'm finding that the response time is very poor. even though the load is very load and memory seems fine.
% time uptime
  5:14pm  up 2 day(s),  1:12,  5 users,  load average: 0.28, 0.24, 0.27
uptime  0.00s user 2.83s system 130% cpu 2.161 total
% time ls -n
total 48
drwxr-x---   2 60007    6005         241 Dec 15 10:32 hsperfdata_jboss
drwxr-xr-x   2 0        0            117 Dec 14 16:03 hsperfdata_root
drwxr-xr-x   2 60013    1            117 Dec 15 10:18 hsperfdata_XXX
ls -n  0.00s user 1.60s system 177% cpu 0.899 total
% 
Response time in the global zone seem fine
# time ls -n
total 22592
-rw-r--r--   1 0        3             53 Aug 12 06:54 csn.213
-rw-r--r--   1 0        0        11534903 Nov  5 22:05 dot.lupi_zones.filelist.15007.15006
drwxr-xr-x   2 60002    60002        177 Aug 12 06:54 hsperfdata_noaccess
drwxr-xr-x   2 0        0            117 Aug 12 06:55 hsperfdata_root
ls -n  0.00s user 0.13s system 144% cpu 0.090 total
# 
The only thing which I can seem which might be causing issues is that the zone is stored on an iscsi share off a netapp filer. However if I do something like an 'ls' in the same directory from the global zone, response time is fine.

All zones on this machine seem to have this problem. I have another t1000 box which doesn't have this problem.

Where should I be looking, what could be causing this?

running Solaris 10 11/06 s10s_u3wos_10 SPARC

Thanks
Comments
Locked Post
New comments cannot be posted to this locked post.
Post Details
Locked on May 19 2010
Added on Dec 16 2009
4 comments
215 views