Redundant HBA

Similar Questions

• The fact to unplug cable adapter HBA on the ESXI host causing VI Client to hang

  We use HP DL380 G5 with the embedded HP version (ESX 3.5i U3 + latest patch)

  This server has all of the available features available (HA, DRS etc..) The server has 2 EMC QLogic 2340 HBA with the latest EMC firmware in it.

  If I look in the VCenter in a disk storage properties it shows me 4 available for SP on our SAN paths. (all MRU settings. we have SAN active/passive)

  Now, here's the problem we have:

  As soon as I unplug the fibrecable, the Vcenter must use another active path. It looks like this, the VM Machines are continue to operate, but only after a new analysis of the hba it also shows that the 2 paths are broken

  But when I close the VI Client and run it then the problems begin:

  • Machines of VM loses connections (rattling irregular and after awhile not available)

  • Customer VI freezes and does not

  In the end, I have reconnected the fibrecable (because each time I start the VI Client it crashes) all problems disappeared and everything works again.

  If I then take a look at the Machines-VM BSOD some are, some have the writeback failed

  Disconnect a fibrecable shall not give any disruption, that's why we have redundant HBA. But it does!

  Anyone have any suggestions?

  HBA QLogic VMkernel pilots understand firmware which (actually) hot-loaded on the HBA driver support.

  See http://communities.vmware.com/message/325181#325181

  So, don't bother with the HBA firmware level.

  See the latest "Fibre Channel SAN Configuration Guide.

  page 99: setting the time-out HBA to failover (value qlport_down_retry)

  Good luck.

• 5.1 ESXi does not see any number of logical unit after restarting both MD3220 and R820

  We restarted our host R820 ESXi 5.1 and table connected to SAS MD3220.  After the reboot, the HBA on ESXi shows no device.  We turned on the MD3220.  WAITED 5 minutes before feeding the R820.

  FYI... the complete Setup includes two R820 connected to two MD3220 through SAS.  The two R820 cannot see all devices on two HBAs after reboot.  We have redundant RAID controllers on the MD3220s and redundant HBA on the R820s.

  On the MD3220, the status of host interface ports appears.

  Anyone know this problem?

  We found the problem.  On the MD3220s, we noticed some "unidentified host identifiers.  For a reason unknown to us, the host port identifiers changed after turning power off and power on all R820s and MD3220s.  To resolve this problem, we replaced the identifiers.  It seems that someone swapped the SAS cables.  I don't think that someone made.

• The correct configuration on the redundancy help

  Hello

  We have a VMware ESXi 3 hosts 4.1 with two Equallogic ps4000 infrastructuer, an AIRLOCK and one with SATA drives.

  It is connected with 2 switches, where everything is connected, so redundancy "should" work.

  On ESXi server, we have 8 network ports, in vCenter, we used 2 NICs for the management and vmotion, 2 for iScsi traffic and 2 for network access VLAN.

  Some time ago, one of our switches failed, and all the VM rebooted because they lose the connection, 2-3 minutes later they came back on the use of the other switch, but it should not happen. I did a little research and discovered that perhaps consolidation of NETWORK cards has not been configured no doubt.

  On vSwitch 2 where will my Iscsi traffic, I used a VMkernel port on both network adapters, it should work, right? What have I done wrong? Someone has an idea, what configuration should be placed on the consolidation of NETWORK cards to do this work.

  After a bit of trouble, I clicked on change properties on vSwitch 2, clicked VMkernel port ISCSi, hget to exepotions political in nic teaming configuration. Failover detection network: Link status notification only, switches: Yes, the failback: No.

  When I disconnected the cable, vmwares worked, when I unplugged the swich, RM of vm. I heard when using network cards, should I create the vmkernel ports on this vswitch using redundancy, is this true? I also thought that the ISCSI port group traffic would inherit the rights of the nicteaming, but this happens for some reason any. Someone has an idea of this network which is the best practice, when you configure the redundancy with multiple cards in vCenter?

  Thx for all the help.

  Bit of an old thread, but in the possibility it has not been resolved, here's your answer:

  With Equallogic units, you will need to configure a vmkernel by physical uplink port.

  1 vSwitch

  2 unique physical network (vmnic3, for example vmnic4) cards

  2 vmkernel ports

  -vmk1-> override set failover of vmnic3 active and vmnic4 unused

  -vmk2-> override unused vmnic4 set failover active and vmnic3

  Bind the vmkernel ports to the iscsi hba software and configure storage.  To improve the performance set up frames on all vswitches, switches and vmk.

• iSCSI HBA + SAN, EqualLogic

  I have a new deployment, I plan.

  I'm looking at a PS6000VX EQL - dual controller with 4 x GBIT by controller cards. In my opinion, it works in an active/passive configuration.

  What is worth of watching me iSCSI HBA with this sollution? Cost is not a problem at all. Performance/reliability/redundancy is.

  I understand with HBA you can multi-path, but with a sollution active/passive is not treated by the table flipping? In that it switches IP on assets of Eve? (with equallogic?)

  Any comments?

  Do you have more information in this plugin MPIO?

  Is still in beta... so there is no official information.

  But will probably like the MPIO module for the Microsoft iSCSI initiator (which can use all the links on the active controller).

  There will be a big reason to go with the software on a HBA iSCSI initiator?

  It's cheaper

  And you can license to have the company more

  understand my reading that Equallogic doesn't have any special configuration as the SP manages failover.

  Right.

  Just remember to follow the proposed (by Dell) network topology if you use two switches (instead of a main switch with 2 stacked modules).

  André

• Several QLogic QLE2462, HBA in an ESX Server twin?

  We are studying the possibilities of virutalize some of our SAN FC connected servers. These servers are more data intensive, and each of the 10 physical servers 2 paths for our EMC CX SAN. The question comes down to virtualization to them, we will continue to soon use the LUNS of same as physical origin servers used in RAW mode. These LUNS are DATA LUNS only and operating systems will be held in images VMDK on a VMFS LUNS.

  Since each of these servers had a connection to the consolidation of our fabric SAN 3 or 4 of them to a single server VMWare would free many fiber connections on the switch. And the question was raised if once before 4 servers with their own direct SAN connection would be affected having to share a single path to the San.  This makes we wonder if anyone out there has multiple HBAs in a single ESX host? Would you add x 2 or maybe even 3 x QLE2462 HBA in our unique AREA of HOST ESX? Will be ESX using these cards or it always by default is simply having a connection and the rest comes to be considered as standby?

  Thanks in advance.

  Well, do your hosts (at least in this example) less than 500 KB/s and a reasonable number of IOPS / s, 1 card so should be much (although I recommend 2 cars for redundancy).

  -Matt

  VCP, vExpert, Unix Geek

• How many paths with HBAs iSCSI supported - failover?

  Hello

  Some time ago, I read when using Software initiator for iSCSI it is only supported to use a path.

  Now for host bus adapters:

  When I use a hardware initiator dual port HBA QLogic iSCSI adapter, is it supported to configure each port to connect to the storage on a different subnet?

  So that it looks like the FC switch?

  Example:

  HBA port 0: 192.168.1.100

  HBA port 1: 192.168.2.100

  Storage:

  Storage port r: 192.168.1.1

  B: 192.168.2.1 storage port

  I can use it like that?

  Thanks in advance!

  Concerning

  /EGR

  Hello

  Yes it is possible. But that I would use to different switches for redundancy and 2 VLAN different as you describe.

  Best regards

  Lars Liljeroth

  -

• How many virtual machines on a single HBA?

  Hi all

  I understand a VI3 environment for a client, but I doubt if HBA will be a bottle neck for the performance of the virtual machine. I would like to know how much minimum I should recommend? Two enough a primary and another for failover? can you tell us the load balancer for HBA? I read that we can have several number of VMs communicate via a single HBA, however, there will be performance problems? Can anyone share their experiences and their recommendations?

  Thank you

  Rahul

  I've not seen HBA as a bottleneck, especially with the 4 GB HBAs now commonly available.

  I only dual port HBA in our blade servers and servers of U 1, I ran 2 x single port HBA.

  Main SAN issues, I've had have been a doubtful avalanche Transceiver and a motherboard fault on a Server Blade.

  Otehrwise HBAs themselves were enough sound that's why I am not too concerned about the unique dual-port HBA's on the blade.

  A lot of the design is subject to redundancy and how much risk / single points of failure, you are willing to tolerate.

• You would rely on external redundancy for ASM?

  DB version: 10.2.0.4
  Operating system: Sun Solaris 5.10
  
  If you have been setting up a database of critical Production on RAC ASM and equipment storage (Sun or IBM) already has a capacity of RAID10 correct, you would still NORMAL or HIGH redundancy? If so, why?

  no no no.

  A server can have multiple HBA cards. The HBA connects to a storage network that can support multiple storage arrays.

• The upgrade to the will of the Sierra that software redundent

  Upgrade to Sierra will make it redundant software?

  Hello Robert,

  Thank you for using communities Support from Apple!

  I see Your ' e concerned about the software that you have installed on your Mac and its compatibility with macOS Sierra. You will need to check with the developers of the software that is installed to see if the version you have is compatible with macOS Sierra, or if there is an update available for her.  As a precaution, you can back up your Mac with Time Machine and then, if your software is not supported under macOS Sierra, just go back to your previous backup:

  Use Time Machine to back up or restore your Mac

  Take care.

• Redundant notifications watch OS 3 Countdown Timer

  Watch 3 OS, the countdown now redundant notifications. When the timer counts down to 0, you get the normal "meter made" alert and then 15 seconds later, you also get a notification of alarm named 'COUNT_DOWN_TIME_REACHED '. It doesn't seem to be a sign of timer settings in the application settings of the watch, so I can't understand how to disable this alert redundant and boring.

  Hello

  This seems to be a bug. You can report (as others have fact) to Apple here:

  https://www.Apple.com/feedback/watch.html

• Z210 and SAS HBA: those that will work?

  I have an Z210 with the processor E-1240 and 16 GB of RAM, Win7 64-bit.  I have a few hard drives SAS 15 K HP's and I want to install on this machine - but the Z210 has no integrated SAS controller. I discovered that there is an Avid Certified version of this computer, and it uses a controller LSI SAS SAS9200-8F to hang the external storage drives. LSI SAS9200 - 8 used to be available in a version of the internal interface (LSI SAS 9200-8i), but that is only apparently now deployed for IBM and there are no Windows 7 drivers.

  I want to install an INTERNAL SAS HBA. and I noticed that there is an older model, the LSI SAS3041e, which has drivers for Windows7 64-bit available. So, I think that this cheap SAS HBA could be a good candidate for my Z210.

  Has anyone any how SAS HBA in an Z210 properly installed? Specifically the SAS3041e LSA? I think that the issue could boil down to how you install drivers and access the HBA for the boot BIOS...

  DRM for your collective wisdom!

  Well, after much research, I decided that the LSI SAS3041-e (non-RAID) controller was a likely candidate. They are available on eBay for almost nothing.

  I just installed one in my Z210 and it was as easy as possible: Although the technical docs of LSI a several step process to load the drivers Windows 7 Pro 64-bit with Service Pack 1 installed, the drivers are already present in Windows 7 Pro. I plugged the controller, hooked to a pair of HP SAS 15 K disks and booted from the disc of Windows 7 HP system recovery.

  The controller immediately recognized Z210 installed the BIOS and I was off and running. Installation of Win 7 completed smoothly and when the machine finally started in a new Win 7 system, everything was fine and my two new hard drives SAS have been properly formatted and correctly recognized. You should be aware that you will need a pair of "bizarre" SATA cables to use these discs 'battleship-quality '. Cables have a conenctor unified at one end to connect the SAS drives, and the other end is a plain old SATA drive connector and a SATA power plug.

  Even if SAS 15 K drives are slightly slower than a disc of 6 GB/S SATA III, the practical speed difference is trivial, these drives have a superior quality construction regular SATA HD, I think.

• Store steel 640GO - message data error (cyclic redundancy check)

  Hello

  I had the car store for a few months, so a lot of data on it.

  Today, windows detects it as a local disk and when I try to access it I get a message telling me that the disk must be formatted when I select number, I get a message saying * error (cyclic redundancy check *) data.

  Help.

  I tried two other laptops, both give the same answer.

  Thank you

  Guy

  Hello

  It seems that something go wrong with the HARD disk partition.
  Did you try running chkdsk and to check the external HARD disc ability?
  Do this!

  Startup command box (cmd)
  You can use the command: chkdsk c: /f r + +.
  But c: means that the HARD drive or partition, you need to change this with the same letter that is assigned to the external HARD drive

• Adding redundancy backup with time Capsules

  Are there problems associated with adding a second time Capsule to my system in order to add additional redundancy and safety of my backup options? I have a MacBook Pro and a single Capsule of 3 TB. I am looking for a non-Cloud option add extra security while maintaining the automation of the OSX TimeMachine.

  Adding a second Time Capsule works perfectly.  After that Time Machine performs a full backup of your Mac (s) on the new Time Capsule, it automatically alternates between two time Capsules backups.

  If you have the possibility of adding a USB or FireWire drive directly to your Mac, which would be a bit better set up in terms of reliability of time Machine, but probably not very convenient if you have a laptop that is moving all the time.

• Adding adapter HBA

  I added an HBA LSI9280 adapter that is installed to a server RD 530. I already had a Raid of LSI9240 card on the system for integrated readers. I also used the LSI Mega RAID management software. I wasn't sure if it I'm missing something or do I see the new map 9280 on the server? I enclose the 9280 a new SA120, which I also do not see at this stage. Would it be possible that the new configuration is not connected correctly?

  Or is it a plugin go type installation?

  It seems that riser was the question. I changed my 4 ports NICcard and my hba... and my HBA worked perfectly... While my NETWORK card was not after slot machines. I have a new riser card on the way.