Sun 7110 + Esx 4.1

Hi ,
we're setting up a vmware ha cluster to serve some virtual machine from one sun 7110 4,2tb,but iscsi and nfs performances are really really slow.We can't go in production with such slow system.We have tried both nfs and iscsi obtaining througputs inside linux vm starting from 14mb/s arriving to 100 mb/s .While a simple 6k$ sas external storage arrive to 220 mb/s of write throughput inside vm.
Why Oracle sells sun 7110s for virtualization,as written in www.techdirt.com/iti/resources/midsize_vmware_intel.pdf (written by oracle)??

Hi Scott,
Welcome to VMware Community,I saw the error message posted in the previous thread.I doubted it is issue with build version.Can you share your build version of esx host.
Have a try with re-scan adaptor.
VMware KB: Performing a rescan of the storage on an ESX/ESXi host
Based on the error message i found few KB below
VMware KB: Booting the ESX host fails with this message in the console: Restoring S/W iSCSI volumes
VMware KB: VMware ESX 4.1 Patch ESX410-201104406-BG: Updates mptsas, mptspi device drivers

Similar Messages

Using Sun 7110 in a Microsoft AD network

I have a SUN 7110, I would like to use it as a file server in my Microsoft AD network. I have an old file server running on Windows 2003 server. I would like to move all files over to the 7110. If I choose to use CIFS is there a way for the 7110 to talk to AD. How would I create the shares and use AD authentication for the file.
I'm new to the 7110 so this might be an easy question

Thanks for sharing your idea here, it's really useful
Mesothelioma

Recreate pool problem on a 7110 Sun Storage

Hi,
we have a Sun 7110 that runs mostly well, except when the iscsi volume fills up, the volume stop, and we have to drop it and recreate it. 3 weeks ago, to counter that, I switched on dedup and compression. No idea if it is a good or bad idea, but because we got lots of i/o hanging on the volume :
- we tried to drop again the volume via the web interface.
- after destroying it, impossible to recreate it, so we switch to the shell to see what's going on. In fact all space was filled up into the "zombie" share (pool-0/???/zombie). No idea what it is, because we never got snapshots. And the cpu was still in the roof like it was still compressing, deduplicating or what...
- we decided to connect to destroy the whole pool (with difficulty), and the CPU stop working as crazy
Now impossible to recreate it via the web interface.
- i can create the pool with zpool and create volume with zfs but they don't show up in the interface
- if we destroy again and try to create it in the web interface, we have of course no pool in the "Available pool", so I try to create it BUT i can't:
Model     Sun Storage 7110
Data Devices     -
Log Devices     -
Cache Devices     -
as is there weren't any disks!
I checked : the disks are ok from the OS point of view (/var/ak/logs/debug.sys doesn't complain for example, except for a spare disk that I labeled back)
Any idea how to "reset" the storage pool?
Do I must go through the "factory reset" (I'm not even sure it can help).
Or should I forgot Sun/oracle appliance kit, and install freenas instead? :-)
Any help will be appreciate (or similar concerns :-) )
Nicolas Zin

I answer to myself:
factory reset solved the problem.
Dawn box!

Sun VDI 3.0 patch 2 - linux guest display is laggy/unusable

Hello.
We have setup Sun VDI 3.0 to test out ubuntu desktop images. We have it functionaly setup but the lag/redraw is unusable. For example opening, closing and moving windows around has a horrible lag and takes a second or two to redraw. Just doing an "ls" from the command line is noticeably slow. I couldn't figure out any good way to measure the visual slowness I am seeing so I hope my plain english description makes sense.
The environment is setup as follows
Solaris 10 update 7 on x64 hardware.
Sun VDI 3.0 patch 2 (patch 1 behaved the same)
virtualbox 2.0.10 (version 2.0.8 with patch 1 behaved the same)
Sun 7110 as storage device
Desktop Images tried on the desktop
RedHat 5.3 (32 bit) and Ubuntu 8.04.3 (32 bit)
We installed the Sun VDI software into the all on one host demo mode. Our host is configured with 24GB of memory (24gb of swap as well per requirements). The virtualbox installs were done on the 2.0.8 version of virtualbox and then imported into the vdi solution. We ran between 2 or 3 vm's at one time so the box still had plenty of ram left since we only allocated 512MB to each vm (tried uppping that to 3GB with no perf improvements).
The Sun Ray units we tried is a stand alone 270 and 1 2FS unit. Surely i'm not the only one attempting linux desktops. I loaded up an windows XP desktop for testing purposes and it functioned great. Barely noticeable performance difference. Any ideas on what I'm missing?
Thanks. Deet.

Dirk.
Thanks for the input. I made the changes you mentioned and got an order of magnitude improvement! I was guilty of being a Solaris admin looking for a Solaris problem:)
Deet.

Sun Studio 12 and ESX question

Hi,
Can anyone tell me if I install Sun Studio 12 in a Solaris 10 x86 VM running on ESX 3.5 if my users will have any issues compiling with the AMD extensions? From what i gather at this point the the CPU features are not hidden from the VM so having an Intel based server is obviously out of the question and I must have an AMD unit, but i still have the question of whether or not the user will be able to compile with the AMD flags because they are running in a VM vs. a real host. Can anyone tell me if they have an environment like what i'm describing and what issues they might be having? Thanks in advance.

From what I know about this kind of virtualization, you get [almost all] what CPU can give. If you are on AMD and boot Solaris kernel in 64-bit mode, then yes, you can compile with x64 extensions. But then again, it's better to ask on VMWare forums - it's not up to Solaris (or Sun Studio compiler) to provide instruction set, it's up to ESX.

VMware ESX + Sun Cluster - is it supported?

Do you know if Sun Cluster installed on top of VMware ESX can be supported by Sun/Oracle?
Do you know what requirements newly installed cluster has to meet in order to satisfy Sun's support contract?
Regards,
Marek Barczyk
Edited by: Marek_Barczyk on Jun 29, 2010 7:13 AM

The binaries (executables) in an Oracle home are "linked" (link edited?) to the OS libraries on each server where the software is installed.
Unless the OS is IDENTICAL on each of the IDENTICAL(HW) servers -- that would share the Oracle home--, you could be in trouble.
The only supported configuration (I know of) where the Oracle binaries are shared between servers is 9i RAC. On 10g RAC the binaries are installed on each server.
Other wise I'd say it's NOT recommended, besides you don't save anything (execpt a cooupl of Gigs disk space).
:p

MMC with Computer Management Utility & Sun Storage 7110 in CIFS ?

Is it possible and supported to manage the filer SUN Storage 7110 configured in CIFS with the "Computer Management utility (MMC)" from Windows?
I've tried without success :-(...
Thanks in advance for the support.
Regards,
Joaquim Neves

I believe your source for users/groups should ideally come from LDAP or Active Directory. The "Users" interface is primarily for assigning administration capabilities for the 7110 itself.

Where to find the fishworks appliance for SUN Storage 7110

Hi,
we've got an old Storage 7110 we would like to re-use, but some disks have been interverted and that may be causing the problem: there's just no system booting up, hence no Storage interface available. All we got in the iLO working.
Is the appliance system supposed to be installed in one of the disks in the front bays ? How to recover or to install the appliance, which seems to be "fishworks" ? I coudln't find any full release of Fishworks anywhere, only some patches are [made available|https://wikis.oracle.com/display/FishWorks/Fishworks].
Thanks
-pph

989293 wrote:
yep, thanks, there isn't even the GRUB menu left. It tries to boot a weired Windows 2003 which ends up with a blue screen ok, that is definitely wrong.
But is the original operating system supposed to be installed on one of the disks or somewhere else, maybe inside on a flash disk ?The 7110 uses two of the 16 disks that are front-accessible for the OS. The 7120 has two extra disks on the back.
By the way, what firmware should I use to upgrade the machine ? I found the the latest update for X4240 servers (SW 3.2.1) but none for Storage 7110. Is it ok if I use the X4240 firmware to update ? It proposes to update both, ILOM and BIOS (v68) but the bios doesn't seem to get updated (still getting xxxsomethingxxx64).I am not 100% sure but I think you should not upgrade the iLOM and BIOS on your own and they are included in the software updates that are installed via the GUI. So I think without contacting support it might be impossible to revive this system
Bjoern

Sun Unified Storage 7110 and iSCSI Initiators

Is it possible to secure the number of iSCSI LUN's a host will see?
For example I have 6 LUN's and 4 Solaris guests (initiators) and I only want certain guests to see certain LUN's, not all of them?
Thanks

Hello,
You have to declare an initiator group with all the hosts you want to map LUNS to.
In the lun-protocol tab, select the initiator group that you want to see the lun. Only this initiator group will be able to see the lun.
Hope it helps
Regard
Pablo

@SUN: Sun VirtualBox 3.1 + VM "Teleportation" ... for VDI 3.1 too?

I just checked my Facebook account and there I got the announcement from Sun that VirtualBox 3.1 was released today and that it has an exciting new feature: VM "Teleportation". So running VM's can be live migrated without interruption between VirtualBox servers. Awesome!
BUT:
Will we get a special VDI version too? If anything then "Teleportation" is a must-have feature for Sun VDI. Will there be a patch that adds this new VirtualBox version to VDI ...?

Just simply load balancing? Or something more?I often encounter potential customers who'd be interested to buy Sun VDI but they shy away from such thoughts because they realise that the VMware ESX license costs they would have to spend money for will be killing them financially. And they don't see VirtualBox as a "good enough" alternative: It's a "toy", it can't do live migration (e.g. with VMware if they wanted that feature one day they could simply license it and "unlock" it), it's not a bare metal hypervisor ....
So now I could at least counter the point about VirtualBox not being able to do live migrations because now it can do that and there are no additional licenses to pay and no features to unlock. The feature is there from start. So that's good.
But the other issues remain: it's not "bare metal". It would have been nice to have some sort of true "VirtualBox Server" or "VirtualBox for VDI -- Bare Metal Edition". Sun had such a product in the pipelines: "Sun xVM Server" ... but this product was killed after EA #2. Silly move, IMHO. It would have been wiser to finish this product and then have Sun VDI support it as virtualisation backend as well.
Talking of which: Any chance that Sun VDI might one day support the various XEN-based products that exist? If I am not mistaken Sun's "xVM" hypervisor in Solaris/OpenSolaris is in principle just another XEN implementation too. And then there is "Oracle VM" .... so in light of the merger with Oracle I'd say it would make sense if Solaris xVM and Oracle VM/Xen were supported as well. I find it odd that XEN-based products which are widely used in certain circles (e.g. Universities) are being omitted but that there is support for Microsoft Hyper-V which is so brand new and still so lacking in the features department that it is hardly used by anyone except by some paid "early adopters" ...

7110 refuses to update and has huge logs

I have a 7110 which is running 2009.04.10.0.0,1-1.2
I have tried updating it to the latest release and got an error 255 saying that something generic went wrong.
The error reads "
An unanticipated system error occurred:
failed to update: job failed with status 255
This may be due to transient failure, or a software defect. If this problem persists, contact your service provider."
I also tried updating to 2009.04.10.2.1,1-1.15 and got the same error.
The audit logs files are huge with 7057801 entries. How do I clear these.
When I reboot the system is takes a very long time before the bui is working, the ssh times out and drops me to the normal
unix shell. I can see the ak process consuming over 25% of cpu for a very long time. I always log out of this emergency shell without
making any changes to the system.
Im uploading a support bundle /cores/ak.b03120ea-64a7-69cc-ad72-b0854fd154b8.tar.gz in case someone at Sun has a look at this.
I am trying to upgrade this to the latest version as it is part of a VDI 3.1.1 setup - which is half upgraded from vdi 3.1 but waiting on the storage :)
Many thanks for any help.
Regards
Edited by: mlasham on Mar 27, 2010 9:27 AM
Edited by: mlasham on Mar 27, 2010 9:31 AM

alexandriap1975 wrote:
I have a Macbook Air that refuses to Update.
Update from what?
Update to what?
Which MacBook Air do you have (Apple icon from the menu bar > About This Mac > More Info)?
How are you attempting to update (e.g., use the App Store)?
alexandriap1975 wrote:
I can't get my Messages App as well as my Mailbox to quit.
Does "Mailbox" mean Apple's Mail app?
You could click the Apple logo in the menu bar, then "Force Quit" and exit both apps.

DB Import is very slow on ESX

Hi All,
We are performing OS/DB migration from Sun/Oracle to Win/SQL server. Export took 9 hours when we perform import it is taking more than 25 hours and we are using ESX ( VMWARE), import system configuration is
OS : Windows 2003
DB : SQL Server
8 CPUS
32 GB RAM
1.5 TB HD ( for databases)
500GB for log files 1
500GB for log file 2
Importing is taking very slow and we are split export and import method. Please help me reason for the slow performance and let know any specific configuration we have to use for sap on vmware specific to SQL Server DB.
Regards,
Kanthi.

> - How big is your queued disk I/O? (use performance monitor to check)
This number indicates whether you have an I/O problem (using too many R3loads in parallel). If you want to trace down performance problems you should get familiar with the Windows tools, you may need that later during production also.
> - How many parallel processes are you running?
>
> 20
And you make sure using migration monitor that no packages of the same table are loaded simultaneously if you have split the talbles?
> Network team is complaining about some ESX issues are experiencing, is this effect to the performance of the DB-Import.
I can't say since I don't know if you load the data from a network drive or locally or if the system loading the data into the database is local or remote.
I suggest you open an OSS call, they may have a look on the system and may give you some idea. Just from a description without having actual I/O key numbers it's impossible to give any hint where to look at to speed things up.
Markus

Solaris won't boot after installation on ESX 3.5

Hello everyone,
After installing last Solaris 10 x86 (download last iso from sun site) on a virtual machine on ESX 3.5 does not start at boot because it shuts down without displaying any errors.
Someone has done it? How to set up your vm?
I have already read these documents from Google but have not solved the problem
- http://pubs.vmware.com/guestnotes/wwhelp/wwhimpl/common/html/wwhelp.htm?context=guestnotes&file=guestos_solaris10.html
- http://www.hashpee.com/2009/09/sun-cluster-32-on-esx-35.html
- http://wotho.ethz.ch/ESX_solaris/Install_Solaris_on_ESX.html
Thanks in advance
Edited by: alman64 on Oct 1, 2009 9:20 AM

I'm having the EXACT same problem and have found little to help...
I too would like to use ZFS for a larger disk pool, but at this point I am thinking about loading Redhat instead.

Help Requested - Creating Solaris 10 x64 Virtual Machine on an ESX Server

Hi All,
I have been trying to create a Solaris 10 64 bit Virtual Machine on ESX and VMWare server Virtualization Platform.
One of the forum entries said - The installer doesn't really care. At every boot, the processor type will be determined and a 32-bit or a 64-bit kernel will be run (unless you override it)
I hence downloaded the iso from http://www.sun.com/software/solaris/get.jsp and I installed it. When I run uname -a command on the terminal, this is what is displayed -
SunOS unknown 5.10 Generic_127128-11 i86pc i386 i86pc
This indicates that the kernel is 32 bit right?
Please help me create the 64 bit VM.
Thanks in Advance,
Raj.

Hi,
I am having lots of trouble doing a jumpstart
installation of Solaris 10
on a T1000 Server.
My setup
Using Debian Sarge as the Jumpstart Server ( I have
used this machine to install Sol 8 and 9 on Sun
V240/210. Running TFTP Server, RARPD,
BOOTPARAMD,NFS.
T1000 - No OS installed.
Attempted an Install by doing - boot net -v install
Below is the screen output.
Rebooting with command: boot net -v install
Boot device: /pci@7c0/pci@0/network@4 File and args:
-v install
100 Mbps FDX Link up
Requesting Internet Address for 0:14:4f:21:b9:c4
Requesting Internet Address for 0:14:4f:21:b9:c4
100 Mbps FDX Link up
Using RARP/BOOTPARAMS...
Internet address is: 192.168.2.35
hostname: sun-server
domainname: milky
Found 192.168.2.1 @ 0:20:e0:65:48:3d
root server: 192.168.2.1 (192.168.2.1)
root directory:
/jumpstart/2.10/Solaris_10/Tools/Boot
Found 192.168.5 @ 0:d:56:10:13:28Don't see any problem here.
boot: cannot open kernel/sparcv9/unix
Enter filename [kernel/sparcv9/unix]:
Do it a favour. Go Sarge box and goto /jumpstart/2.10/Solaris_10/Tools/Boot
and check do you have required kernel (kernel/sparcv9/unix)
also check your privilege for the file please.
I think it is possibly not mounting the nfs share. I
have already changed the NFS_client variable to 3
under etc/default/nfs in the Boot directory.
I would appreciate any help or suggestions.

VDI 3.1 cloning fail if one esx from a cluster is in maintenance mode

hi
i have a vsphere cluster with 2 esx included and ha and drm enabled
if one of the esx is in maintenance mode the cloning job failes but with a misleading error message
INFO: Datastore 'Sun Storage 1' is not available for cloning, it is disconnected or in maintenance mode
i see here 2 bugs
1) the wrong error message cause the datastore is neither disconnected nor in maintenance mode
2) cloning should still possible if one esx server is not available
br
Andre

Hi login,
It should be possible. You will need to route the output of the counter to a specific RTSI line and then use that line as the sample clock by the MIO board. The function you will want to use to route the counter output to the RTSI bus is the Route Signal.vi in LabVIEW or the Select_Signal() in NI-DAQ function calls. You will want to use constants such as ND_GPCTR[x]_OUTPUT where [x] is the counter number and ND_RTSI_[x] where [x] is the RTSI line number. Your source is the counter output and your signal is the RTSI line of choice.
Similarly, for the MIO card, you will want to route the RTSI line to the AI Scan Start to route to your scan clock. The constant for this should be ND_IN_SCAN_START and it should be the signal value while your RTSI s
hould be the source.
The point is you will need to route the signal for both cards which amounts to 2 calls (or 2 VIs). Anyway, hope that helps. Have a good day.
Ron
Applications Engineering
National Instruments

Sun 7110 + Esx 4.1

Similar Messages

Maybe you are looking for