Hi Philip,
Evan Powell, Nexenta's CEO were recently bloging about the so called
Delorean project - a Time Machine for Windows:
http://www.nexenta.com/corp/index.php?option=com_mojoItemid=153p=18
He says:
‘Delorean’ is intended to allow time travel for windows users by
using ZFS based
Could you try to set the IO scheduler to noop:
echo noop /sys/block/sdX/queue/scheduler
and see if that makes a difference?
Thomas Georgiou wrote:
I have 2 HP DL380G4s each connected directly to a FC JBOD (Powervault
224F). Each server has 2 QLA2200 HBAs so multipathing to the JBODs is
in
I would recommend to isolate Networking first. I.e. run iperf between
the boxes (Linux vs. OpenSolaris), single thread, 512 bytes payload -
this will somewhat emulate iSCSI traffic over single raw device. See
what you'll get...
Thomas Georgiou wrote:
Setting that under Linux, I get 28.6MB/s
Which NIC/driver you are using?
Thomas Georgiou wrote:
initiator(linux)-osol: 630Mbits/s
initiator-linux: 520Mbits/s
linux-initiator: 733Mbits/s
osol-initiator: 297Mbits/s
It appears that osol is having a little trouble transmitting information.
On Tue, Aug 4, 2009 at 6:16 PM,
Forget it.. you will not get good transmit results with bge driver as it
is not yet supporting LSO it seems:
http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6341158
Try Intel 1Gbps card, e1000g driver supports LSO
Thomas Georgiou wrote:
All the servers used have Broadcom
and what is mpstat 5 is saying while you running 512b traffic?
Also, do you see retransmits during the run?
i.e. netstat -s|grep Retra
Thomas Georgiou wrote:
I installed an Intel Gigabit nic and continue to get 280Mbits/s
through it (verified via iftop). However, if running iftop on
e1000g0,
Hi Roman,
take a look on what NexentaStor is providing:
auto-cdp - a commercial plugin which simplifies significantly ZFS/AVS
management of primary/secondary hosts
auto-sync - a free service which supports sophisticated ZFS send/recv
over SSH, NC. Or use send snapshots over RSYNC - which
All public plugins CDDL licensed, however they really depends on NMS and
component APIs. More public plugins coming soon such us: UPS and
ClamAV/ZFS - but also, this website can host projects which are related
to OpenSolaris / Nexenta, such as kernel gate:
BTW, there are two commercial plugins for NexentaStor as well which will
solve HA problem nicely in my opinion:
http://www.nexenta.com/corp/index.php?option=com_jreviewstask=listcategorysection=20cat=53dir=1Itemid=112
1. Simple-Failover plugin - active / passive manual failover
2. HA Cluster
Its available actually for NexentaStor/OpenSolaris as a separate plugin.
It supports remote API, FC and iSCSI targets:
http://www.nexenta.com/static/target-userguide-2.1.pdf
http://www.nexenta.com/corp/nexentastor-2x-optional-modules-storage-management/303-target-20
Vik wrote:
In a couple of
Guys,
just to let you know that Nexenta was working for quite some time on its
COMSTAR VTape driver. It is now beta ready and feature complete as far
as code is concerned:
http://www.nexentastor.org/projects/vtape
It requires b131+ to compile and run. Tested with NexentaStor and with
many
On 06/07/2010 06:12 PM, Andrew Jones wrote:
We've built a SAN based on NexentaStor 3.0, and have generally had a good
experience with the flexibility and performance of ZFS and the relatively new
COMSTAR kernel mode iSCSI story.
However, we are experiencing a frustrating and thus far
On 06/07/2010 10:04 PM, Andrew Jones wrote:
n...@hl-san:/$ show applance version
NMC version: 3.0.0-3
NMV version: 3.0.0-5
NMS version: 3.0.0-5
Operating System: Nexenta/OpenSolaris (version 3.0.0)
Its older version I would say... you need to upgrade your software.
Run the following NMC
___
storage-discuss mailing list
storage-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/storage-discuss
--
Erast
___
storage-discuss mailing list
storage-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo
is closed binary and there is no way to debug it further on my side.
Is this problem known/fixed yet?
Please advise. If you have new binary to verify - I'm willing to help.
Thank you.
--
Erast
___
storage-discuss mailing list
storage-discuss@opensolaris.org
On Fri, 2007-06-01 at 18:23 +0100, Pete Dennis wrote:
Hello Erast,
This could well be the same as a bug that was logged:
6553970 VMware Workstation 5.5.4/6.0beta choked when
booting snv_63 or s10u4_06
So you may want to talk VMware about this.
I have tried b65 on build 6.0.0
On Sat, 2007-06-02 at 08:39 +1000, James C. McPherson wrote:
Erast Benson wrote:
On Fri, 2007-06-01 at 18:23 +0100, Pete Dennis wrote:
Hello Erast,
This could well be the same as a bug that was logged:
6553970 VMware Workstation 5.5.4/6.0beta choked when
booting snv_63
Thanks Alan,
this been backported to Nexenta unstable repository, those who need the
fix now, simply apt-get it.
On Tue, 2008-01-22 at 17:07 -0800, Alan M Wright wrote:
Tim Cook wrote:
Allright, I have to say I'm a bit disappointed nobody thought
to point this out from the cifs team.
alternatively, NexentaStor provides ZFS-integrated indexer and
probabilistic search engine with nice GUI and CLI interfaces..
On Thu, 2008-02-28 at 13:00 -0600, Mark Martin wrote:
On Tue, Feb 19, 2008 at 2:30 AM, Ross [EMAIL PROTECTED] wrote:
Indexed attribute searching on top of ZFS
If you looking for extra intelligence for SNDR, search for solutions on
top of it, i.e. like NexentaStor AutoCDP plugin could be used to manage
ZFS pool mirroring:
http://www.nexenta.com/corp/index.php?option=com_contenttask=viewid=150Itemid=112
On Fri, 2008-08-08 at 11:20 -0700, Maurice Volaski
On Wed, 2008-09-03 at 16:47 +0100, saqmaster wrote:
Right now i'm looking at :-
- OpenFiler on generic x86 hardware and a bunch of sata disks
- FreeNAS, as above
Take a look on NexentaStor ZFS software appliance - this is similar to
OpenFiler and FreeNAS, but really moves things to the
On Mon, 2008-09-08 at 15:11 -0700, Sumit Gupta wrote:
On Sep 8, 2008, at 2:41 PM, Ben Rockwood wrote:
Peter Buckingham wrote:
Ben Rockwood wrote:
The COMSTAR iSCSI Target isn't here today, and looks to be some ways
off, thus its not really of concern. In a year I'd guess we'll
If you do ZFS snapshots of VMs, you need to quiesce VM states first.
This can not be done without serious integration with ESX servers. And
yes, it takes more time then just a few seconds of ZFS-end snapshot.
I suggest to take a look on commercial solution from Nexenta Systems,
which tightly
Great!
uploaded to Nexenta APT now, for NCP2 users just do:
# apt-get install ndmpcopy
On Sun, 2009-02-08 at 17:46 -0800, Ben Rockwood wrote:
Vilas Deshpande, I love you!!! Thank you! I've been needing this badly!
benr.
___
storage-discuss
24 matches
Mail list logo