Similar Questions

• Access control for iSCSI

  Hi all

  I have two EqualLogic PS6500E. On volumes of, created under the access control list , I applied and restriction of IP address to allow only the beach of 192.168.1. *.

  Since a few days under connections , I noticed traffic/connections from other IP addresses (not high-end 192.168.1. *) actually configured on the servers of access to this volume.

  Any ideas how is this possible?

  Thanks in advance

  Sébastien

  Sorry, I wanted to add.  Run the Remote Setup Wizard and select the option "Configure MPIO".  That will list all subnets available on this server.  Exclude everything except the iSCSI subnet.

  Kind regards

• What is my best option?

  I have 3 physical blades that have 2 NIC in them and connect to our SAN via iSCSI, each blade aura 3-5 virtual computers on it. Order for vMotion and HA to work what is the best way to configure my management ports, VMkernels (iscsi, vmotion) as not to kill my network when I need to use vMotion? How can I split everything that upward through the NETWORK interface card? My vMotion iSCSI must be on a subnet on its own network card with all the traffic server and management makes on the other network adapter? Any help would be greatly appreciated.

  Thank you

  John

  Well, you won't pretend to CPU\Memory... but I would suggest that you remove the CPU limit, so that every care receives full 2.66 mhz frequency.  To improve throughput of ISCSI, you need to configure jumbo frames.

  On the vmnic...

  	Assets	Standby
  Management	VMNIC0	VMNIC1
  VMotion
  iSCSI	VMNIC1	VMNIC0
  vmnetwork

  That's what I'm doing... using VMXNET3 would really help with performance

• Need help with the data storage store, local array and network connections

  Need help with my ESXi 4.1 installation

  My hardware:

  I built a server with an Asus P6T whitebox, i7 920, 12 Gig RAM, NIC, Intel Pro1000 PT Quad, 3ware 9650SE-12ML with 8 1.5 TB SATA green in a raid 6 array gives me about 8 + TB with a spare drive all housed within a NORCO RPC-4220 4U Rackmount Server chassis.  I also have a 500 GB SATA drive which will hold the ESXi and virtual machines.

  The network includes a firewall, Netgear Prosafe FVS336G, GS724Tv of Netgear ProSafe 24 port Gigabit Managed Switch on a dhcp cable modem internet service provider.

  I also have 2 old NetGear SC101T NAS disks (4to) I want to connect to the system how some - at a later date have... data on them and want to transfer to the new storage array. I always looking into the question of whether they will work with ESXi 4.1, or I might have to only access it through Windows XP.

  My Situation:

  I have already installed ESXi 4.1 and vsphere client with no problems and it is connected to a dhcp cable internet service.  I've set up host via a dynamic DNS service name give me a static hostname on the internet.  I installed three machines to virtual OS successfully at the moment and now want to first start by creating a multimedia storage server which will use some of this new 8 TB array, then separate data storage for use with a web server small overhead storage and a backup.  It is a domestic installation.

  Help with the data store and network:

  I was doing some reading, because I'm new to this, and it looks like I'll probably want to set up my table via ESXi as a nfs disk format.  Now, the data store is usually in another physical box from what I understand, but I put my readers and ESXi all in the same box.  I'm not sure that the best way to put in place with grouped network cards, but I want to make this work.

  I understand that in ESXi 4.1 using iSCSi LUN must be less than 2 TB, but nfs - I should be able to add a bigger partition then 2 TB (for my multimedia) in nfs, right? or should I still add it separately as a separate 2 TB drives and then extend them to get the biggest space.

  Any suggestions or direct resources showing examples on how to actually add some parts of the table as data warehouses separate nfs.  I know that to go to the configuration tab, and then select Add to storage, and then select nfs. I have not my picture, but it's here that I don't know what to do because ESXi 4.1 system already has an address, should I put the same thing to the new data store array also (will it work?), and what should I use for the name of the folder and the store of data... just do something to the top.  I thought to later install Openfiler (for a multimedia storage using this table server) as a virtual machine, use the table with esxi so that I can access the same storage space with widows and linux-based systems.

  I also know I have to find a way to better use my quad nic card... put in place of virtual switches, grouping, etc HELP?

  Any direction, assistance, similar facilities to sample, suggestions or resources that would help would be great. I did a lot of hunting, but still a little confused on how to best to put in place.

  You must think of VMDK files of large databases with records of random size guest go read some data (a DLL or an INI file), maybe write some data back, then go read other data. Some files are tiny, but certain DLLs are several megabytes. It's random i/o all and heavy on the search time. IO Opsys is small random operations that are often sequential (go read data, write data, go read other data,...) so that deadlines are critical to the overall performance. That's why people say OPS are / s of reference and forget the MBs flow. The only time where you bulk transfers are when you read media (ISO files).

  Well, now forget all this. Actually the disk activity will depend on the specific applications (database? mail server? machines compiler?), but the above is true for boots, and whenever applications are idle. You should see the profile to know.

  RAID 10 is faster (and often more reliable) than RAID 5 or RAID-6 except in certain specific cases. In General RAID 10 is ideal for many random writes, since the calculation of parity for RAID-5 and - 6 adds to the overall latency between command and response - latency is cumulative if a little slow here and a little slow it adds up to a lot of overall slow synchronous especially with e/s on a network. OTOH RAID-5 and -6 can produce faster readings due to the number of heads, so you can use it for virtual machines that transfer bulk. Test. You may find that you need several different types subdashboards for best results.

  You said 3ware, they have some good grades on their site, but don't believe it. With my 9650 that I found myself with only a couple of their recommendations-, I put the (simple) table for allocation size 256 k, nr_requests at 2 x the queue_depth and use the planner date limit. I had the habit for the Ext4 file system formatted with stride and stripe-width synced to the table and used the options large_files with fewer inodes (do not use the huge_files option unless you plan to have single VMDK files in the terabyte range). Use a cache of great reading in advance.

  Virtual machines use VMDK files in all cases except raw iSCSI LUN that they treat native disks. VMDK is easier to manage - you can make a backup by copying the file, you can move it to a PC and load it into another flavour of VMware, etc. There could be some features iSCSI to your San as a transparent migration but nothing for me. NFS has less chatter of Protocol if latency lower times to complete an operation. NFS is good to read and write a block of data, that's all it boils down to.

  UPS is good, but it won't help if something inside the machine explodes (UPS does nothing if the PC power supply goes down). If the RAID card has an option for a battery backup module, so it can contain some writings in memory and may end up the disk i/o after replacing the power supply. 3ware also limits the types of caching available if help is not installed, and you get just the right numbers with the module.

• the UCS blade iscsi initialization cannot install windows server

  Hi all

  I have a problem when I try to install windows server 2008 in the storage of nexenta using iSCSI via ucs b-series, but his works well when I install ESXi 5.5. I can't find disk under windows server installation, and I see all the LUNS in ESXi what im create before nexenta storage and work when I install it.

  This vnic iSCSI Configuration:
  ----------------------------
  vnic_id: 13
  LINK_STATE: to the top

  Initiator Cfg:
  initiator_state: ISCSI_INITIATOR_READY
  initiator_error_code: ISCSI_BOOT_NIC_NO_ERROR
  VLAN: 0
  DHCP status: false
  IQN: test: iqn.2014 - 11.com.:8
  IP address: 192.168.1.59
  Subnet mask: 255.255.255.0
  Gateway: 192.168.1.5

  Cfg of target:
  Target the Idx: 0
  State: ISCSI_TARGET_READY
  State of PREV: ISCSI_TARGET_DISABLED
  Target error: ISCSI_TARGET_NO_ERROR
  IQN: iqn.2010 - 08.org.illumos:02:windowsboot
  IP address: 192.168.1.49
  Port: 3260
  Starting logical unit number: 6
  Ping statistics: success (986.646ms)

  Info session:
  session_id: 0
  host_number: 0
  bus_number: 0
  target_id: 0

  can anyone help me to solve this problem?

  Thanks in advance - Fabre

  ENIC / fnic driver must be installed. they are not part of any Windows operating system.

  they can be found in

  Home downloads
  Products
  Servers - unified Computing
  Series B UCS Blade Server Software
   
  Unified Computing System (UCS) Drivers - 2.2 (3)

  ISO image of drivers based on the UCS 
  UCS-bxxx - drivers. 2.2.3.iso

  Depending on the version of the UCS.

• E1000 virtual network adapter on Win 2003 R2 causing CMTF on ESXi 5.1 (PF Exception 14)

  Hello

  We migrated a virtual machine running 2003 R2 64 bit of Hyper-V on ESXi 5.1 (build 1117900) using VMware Converter for migration V2V, which set up an Intel E1000 as NETWORK card (with driver inside of Microsoft virtual machine)

  The migration is smooth and the machine will work perfectly for over a month, but will be little stress conditions.

  
  

  As soon as the machine started the stress a little, with some network transfers, we started having some GROUP steps on the node ESX:

  

  After some reading in the forums where this issue seemed closest vmxnet3 for e1000 (in fact many suggesting they go for e1000...), we made the following according to certain conclusions autour:

  (1) upgraded to the latest version 5.1 ESXi (1157734, this was done without any specific indication, this could solve the problem)

  (2) drivers installed instead of those Microsoft Intel (Intel Pro Set)

  (3) turn off TCP unloading in Intel drivers.

  I will let you know if node will work more stable, but in the meantime I ask for suggestions, as appropriate.

  Unfortunately, no ESX kernel image was there so there is no need to open a call to VMware at the moment...

  I received the official response from VMware, 13387941310

  Solution: switch to vmxnet3 (after installation) or disable RSS in driver

  A patch is due Q1 2014 (ESX 5.1 U2)

• Failed to connect ISCSI Linux

  Hi all

  I have EQL 6210, box 3

  6210-1S

  2-6210-X

  3-6210-E

  Several volumes are mounted with windows and linux blade servers.

  a specific volume is mounted with linux based on physical server and SAP DB is running on it.

  I get messages in the MISTLETOE Group Manager "ISCSI connection to the target of the copycat was closed, ISCSI, Reset connection failure received on the connection: for this SAP DB volume.» Snapshot is attached.

  Please help identify and resolve the problem.

  Hello

  These resets are from the Linux server.  This can be caused by a bad configuration.

  Please see this PDF on RHEL/OEL configuration with Dell PS berries.

  http://en.community.Dell.com/TechCenter/Extras/m/white_papers/20442422?utm_source=release%20announcement&utm_medium=email&utm_campaign=document%20release

  Some clients have seen a reduction in these areas by increasing delays in the etc/iscsi/iscsid.conf file.

  # To specify the time-out to abort before answer
  # operation failed and try a reset of the logical unit change the line.
  # The value is in seconds, and the default value is 15 seconds.
  node.session.err_timeo.abort_timeout = 30

  # To specify the time-out period of a logical unit response
  # before the operation failed and trying recovery session
  # Change the line.
  # The value is in seconds, and the default value is 30 seconds.
  node.session.err_timeo.lu_reset_timeout = 60

  # To specify the timeout waiting for a response from target
  # before the operation failed and trying recovery session
  # Change the line.
  # The value is in seconds, and the default value is 30 seconds.
  node.session.err_timeo.tgt_reset_timeout = 60

  You can see that just increase one.

  Note: Editing this file does NOT spread downwards for all existing volumes.  You must find all the volumes after each change in the etc/iscsi/iscsid.conf file.

  You have opened a case of pension?    Latency on the network can also cause these resets.

  How EQL firmware revision are you currently using?

  Kind regards

  Don

• Migration of the iscsi initiator software to hardware initiator on 2012 Hyper-v cluster.

  Hello

  We have 10 blades x M620 with Broadcom 10gig 57810 M1000e CMC card dedicated for iSCSI Equallogic PS 6110 San.

  Currently software used on cluster hyper-v Windows 2012 connect SAN PS6110 iscsi initiator.

  150 VMS are in production on 15 CSVs.

  We want to change the initiator of the software to the hardware on all our existing blades without CSV output.

  Is it possible on hyper-v cluster 2012 to have 5 blades with initiator software and remaining 5 blades with hardware initiator with same CSV cluster?

  If coexistence of initiator software and material as possible through hosting then I can apply this change a bit on our 10 blades without cluster downtime.

  Hello

  What version of FW EQL do you use?

  Yes, it is possible, however a couple of updates warn first.

  1.) install the latest firmware / boot code and driver for Broadcom cards.

  2.) put the node into maint mode then you can disconnect the SW iSCSI provenance and sign in using the HW iSCSI adapter.

  3.) do not try to do it with an active node / online.

  The advantage of the HW iSCSI adapter reduces the CPU Overload, if you are maxing out the CPU not now, do not expect a large increase in performance after HW iSCSI card.

  Kind regards

• SERV DELL R220 - initializing iSCSI

  Hello

  I'm going to buy:

  SERV DELL R220, E3-1220v3 (3.1 GHz), 8 GB UDIMM, 2x1To 2.5 7.2 K 6Gbps SATA HD, iDRAC7 Express, PERC H310, 3YProSupport.

  DELL_SVR_R220C, 30230601C_R220

  Is it possible to start MS Win 2012 from the NAS via iSCSI? Is it supported by BIOS a built in network adapteror should I need to buy anything again (another network host adapter forex.)?

  Could move 2x1To hardrives SAR at the end?

  Thanks for any response in this regard.

  FR.

  fthis

  Exceptional! I did a little research, and it seems you are right. It supports boot from SAN. SO, I've put together a document with some instructions. I hope that find you them useful. Thanks for the heads up!

  iSCSI - BootfromSAN.docx

• VLANning questions about blades MXL

  Hello

  I have a new M100e with a SAN PS4110xs case, a few servers blade and two switches Force10 MXL (switches and SAN on fabric)

  I put it all in place and works fine, but using the default VLAN for all

  I want to connect the switches to a Juniper EX3300 via SFP ports + and send the subnet SAN out so that I can connect remotely to the IP address of the EqualLogic for VMM integration management

  However, every time I try to put SAN ports in a VLAN separate-, before even that I look into the chutes for the Juniper - I lose all connectivity on the San.

  I created a new VLAN (VLAN120), but as soon as I set the ports inside they cannot ping each other.

  int vlan120

  Tagged te0/1 (this is the port of SAN)

  Tagged te0/12 (it is a single server)

  Tagged te0/16 (that's the other server)

  as soon as I do that I lose all connectivity to the San and server cannot ping each other via the iSCSI network.

  I tried to use 'no label' instead of 'tag' with the same effect.

  The VLAN is active (non-stop)

  I am at a loss for what is the question.

  Any help gratefully received.

  Select this option.

  I understood the question - I got 2 trunk in the Juniper VLAN120 links - as soon as I disabled the second interface everything arises in life

  Now I need to try to find a way to implement a SHIFT between the two MXL in the stack and a Juniper ex3300 - I managed to get the connectivity that I need, but with only a single link - as it is a remote site I want redundancy :-)

• Try get job PowerConnect 8132 + PowerConnect 8024-k (Switch Blade) and PowerVault md3600i on 9000 MTU

  Try get job PowerConnect 8132 + PowerConnect 8024-k (Switch Blade) and PowerVault md3600i on 9000 MTU

  Everywhere I put 9000 MTU (Jumbo frames)

  My systems host is 2 x M520 with Broadcom 10 G (2 ports) mezz card

  On the 10G card that I am trying to get link iSCSI work with storage, but the work than the MTU 1500

  The M520 installed Linux CentOS 6.4 (where I've implemented the 9000 MTU)

  My scheme is:

  

  Please help :)

  Hello, Daniel.

  Problem solved. It's strange, but all devices have started working after restart 8132 switches. :)

  Thank you for your help.

  Good bye.

• Problem; ESX 5u1, initializing iSCSI on NetApp

  The problem that we see is;  VMware 5u1, even on 5 takes about 90% through the installation/copy of VMware on an 8 GB logical unit number (lun id 0)... .fails with the message "wait 2 start benches, found 0.

  HW configuration as follows;

  FIS 6248

  Blades B-200

  VIC M81KR Palo

  Configured "Device Ports" connection to NetApp 3240-10 GB e1a/e1b

  Firmware 2.0.2M

  Service profile

  assigned iSCSI (cdrom, iSCSI nic overlay) No. MAC startup strategy

  iSCSI boot vNIC, fabric a - no. failover, vLAN 550

  iSCSI_A vNIC, no failover, vLAN 550

  iSCSI_B vNIC, no failover, vLAN 551

  1 Mgmt vNIC fabric failover

  1 VMotion vNIC fabric failover

  VM - A vNIC, fabric failover

  VM - B vNIC, fabric failover

  NetApp 3240 DOT 8.0.2P6 7-mode

  10GB Ports connected to the FI 6248 "device Ports.

  E1A (vlan 550)

  E1B (vlan 551)

  It is a known problem with ESX 5.  At the end of installation to the iSCSI boot LUN, the Setup program tries to re - initialize the adapter iSCSI with a DHCP address.  This fails and kicks on the error you see.  Installation fine filled aura - restart just like you did.

  The VMFS LUNS not created on the start for ESX 5 LUN also is expected behavior.

  Kind regards

  Robert

• Boot from iSCSI SAN

  Let's say I have a blade without a hard drive, I want to install ESXi 4.1 on and an iSCSI storage array connected to my ethernet switches to the North. Can I use SAN boot in this scenario to boot from iSCSI? I can't quite find anything definitive either way.

  If it is possible are there restrictions on the type of card mezz blade would need? Ideally, we would use the adapter of VIC, but again, I don't find any details on whether or not this is possible. Is there a resource anywhere that lists the different adapters iSCSI features?

  See you soon,.

  Simon

  the list of supported iSCSI today is as follows:

  -----

  None

  -----

  SAN boot is FibreChannel only now but I heard iSCSI is on the roadmap.

  Perform a PXE for ESX boot works very well and I know quite a few clients who use it (but officially it is not supported by VMWare)

  Kind regards

  Michael

• iSCSI Boot target IP Question

  Hi all

  Setting up iSCSI starts for the first time and had a question.

  Installation: Blades UCS about a pair of N5K (vPC) with a double NetApp controller connected.  There are 4 UCS vNIC - two for only iSCSI (as the VLAN native) and two for everything else.

  The blades odd dΘmarrez since LUN on NetApp controller and the blades even dΘmarrez since LUN on NetApp B controller.

  So my question.  When I make for an odd Blade (for example) iSCSI targets, do I have both A and B "NIC" s point on A controller's IP?  Or NIC - A point on A controller and IP point of NIC - B to B controller IP?  I was not sure that, in this example, the controller would hold unit number logic that this odd blade is starting from.

  Thank you!

  Hi Ian

  You can add up to two iSCSI vNIC startup strategy. A vNIC acts as the source of primary iSCSI boot, and the other acts as the source of secondary iSCSI boot.

  "point of NIC - A controller - A IP and NIC B points to the IP Address of the controller of B ' YES.

  You would create another strategy of starting for blades, with the primary order / trade school.

• Cisco UCS Direct Attached ISCSI Storage access upstream to storage array

  I'm doing my first installation of UCS with a chassis 5108) and (4) B200 M3 blade and (2) 6248 fabric interconnects. The plan is to connect our EMC VNX directly on the fabric of interconnections as the chassis will be the only system for access to the SAN, after migration from our current system. The fabric interconnects will be transmitted to (2) 4500-X switches configured in a vs.

  In my research of best practices and general deployment guidelines I noticed the need to create (2) separate iscsi VLANS for fabric and a fabric in B. I'm guessing that this requirement is the lack of STP as YEW is configured in EHM, but I was not able to find an exact answer. Any idea would be appreciated.

  My main question and my concern is that if I create a second VLAN ISCSI on our X 4500 and then create both of these VIRTUAL on FI networks A & B I'll be able to connect the SAN to the UCS as well as our former VMware cluster (we are migrating our system HP Compute and storage UCS 5.5 environment) in order to facilitate the vMotions of storage?

  I noticed the need for iscsi separated (2) VLAN a fabric and a fabric B

  Are you referring to

  https://www.EMC.com/collateral/hardware/technical-documentation/h8229-VN...

  p 36

  If I understand your question: can a host attached to 4500 (connected to the UCS FI), access LUNS on the VNX, which is directly connected to the UCS fabric interconnect through lanes (EHM FI) material? YES! I would accept this for a migration, but not forever, because it means that traffic on the fabric of interconnection.