Up until a few weeks ago, we were using VC port x8 to connect to an iscsi network, leaving x7 as our stacking link ports. We have 10 c7000 enclosures. We've stopped using the iscsi network so, on each c7000, I deleted all iscsi networks and pulled the sfp's from the x8 ports. AFter a few minutes, x8 returned to serving as stacking links. However, one c7000 only shows its stacking link using x8. So apparenly x7 was never participating. I never saw any stacking link warnings prior to removing the sfp's so it was never noticed. So my question is, is there a way to re-enable x7 so it will act as a stacking link port again?
How to re-enable a stacking link
virtula connect c7000
HI
We have two virtual connect f8 modules on c7000 enclosure that contain virtual connect domain . the vc modules are on interconnect bays 1 and 2.
Because we need more network cards for the blade servers we need to add two more f8 modules in bays 3 and 4.
My question is how to add the new modules to the existing domain.
Are they added automatically to vc domain or we need to add them first in the hardware section
The second question is can we directly connect virtual connect uplinks from two c70000 enclosures that have different vc domains . I mean no Ethernet switches exist in the middle
Virtual Connect - Smart Link issue.
Hello all.
We have some issues with the Virtual Connect.
Our infrastructures:
C7000, 2 HP VC FlexFabric 10Gb/24, 4 servers BL460c and 2 Aruba 2920 Switch
The Virtual Connect is connected to two HPe switches 2920 in a stack.
Virtual Connect bay 1 port x5 connect to switch-1 port 10 and port x6 to switch-2 port 10.
Virtual Connect bay 2 port x5 connect to switch-1 port 11 and port x6 to switch-2 port 11.
4 servers BL460 with a network adapter BL650:
All servers have 2 networks:
NETWORK-1 is management and backup VLAN with Shared UPlink and Smart Link.
NETWORK-2 is trunk VLANs, with Shared UPlink and Smart Link.
The problem is - when I boot the Switch 2920 (The Stack) all ports of the blade servers switched to down.
It causes the cluster to lose the network!
We work with Hyper-v 2016, and the management network should be up all the always for the Heartbeat.
My question is - if I will disable the Smart Link on management network that network local will continue to work even if all Shared UPlink is down?
Thank you a lot!
Aleksei.
Change boot from SAN to local for a HPE Blade
Hi.
We need to know if is possible to change the boot for a HPE blade from a SAN to a local disk or USB CD in order to recover data in data local disk, because the SAN that this blade booted not exist more
Thanks in advance
VirtualConnect and configuring LACP on a vpshere distributed switch
I have 3 hosts that are running ESXi 6 in vsphere 6. All three hosts are connected to a dvswitch. Is there anything that should be configured on the dvswitch itself regarding LACP? I remember just adding all four flexnics to the distributed switch and then not configuring LACP on the dvswitch itself. From what ii remember, LACP shouldn't be configiured on the dvswitch because of the two physical nics converting to flexnics. I haven't worked on Flex10' virtualconnects in 4 years now. Really rusty.
This existing configuration has LAGs configured on the dvswitch and we are having problems with packet loss on many VMs.
Synergy Virtual Connect SFLOW
VC profile unable to change port speed
Hello,
I have modified an existing profile and updated it with my new profile requirments. But I am unable to see the speed changes in server. In server all shows same 10GB. Any idea how can I set the bandwidth to required value. I am using VCEM for updating the profile.
->show profile xxxxxxx
Name : xxxxxxxxxxx
Device Bay : enc0:1
Server : Integrity BL860c i2
Status : OK
Serial Number :
Serial Number (Logical) :
UUID : 1D00333E-A39
UUID (Logical) : 2f7aa86d-6381-4b65-907f-0a9
NAG : Default
Hide Unused FlexNICs : false
UEFI Boot Mode : Auto
SR-IOV Mode : Simplified
Ethernet Network Connections
=========================================================================
Port Network Status PXE/IP Boot MAC Address Allocated
Name Order Speed
(min-max)
=========================================================================
1 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-D0 1Gb-10Gb
Network
-------------------------------------------------------------------------
2 SEC_HPX_UT OK UseBIOS/Auto 00-21-5A-9C-46-D2 7Gb-10Gb
L_3279_L
-------------------------------------------------------------------------
3 <Unassigne OK UseBIOS/Auto 00-21-5A-9C-46-D4 -- --
d>
-------------------------------------------------------------------------
4 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-D6 1Gb-10Gb
Network
-------------------------------------------------------------------------
5 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-D8 1Gb-10Gb
Network
-------------------------------------------------------------------------
6 <Unassigne OK UseBIOS/Auto 00-21-5A-9C-46-DA -- --
d>
-------------------------------------------------------------------------
7 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-DC 4Gb-10Gb
Network
-------------------------------------------------------------------------
8 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-DE 3Gb-10Gb
Network
-------------------------------------------------------------------------
9 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-E0 3Gb-10Gb
Network
-------------------------------------------------------------------------
10 <Unassigne OK UseBIOS/Auto 00-21-5A-9C-46-E2 -- --
d>
-------------------------------------------------------------------------
11 <Unassigne OK UseBIOS/Auto 00-21-5A-9C-46-E4 -- --
d>
-------------------------------------------------------------------------
12 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-E6 3Gb-10Gb
Network
-------------------------------------------------------------------------
13 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-E8 100Mb-10Gb
Network
-------------------------------------------------------------------------
14 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-EA 3Gb-10Gb
Network
-------------------------------------------------------------------------
15 <Unassigne OK UseBIOS/Auto 00-21-5A-9C-46-EC -- --
d>
-------------------------------------------------------------------------
16 Multiple OK UseBIOS/Auto 00-21-5A-9C-46-EE 100Mb-10Gb
Network
#######
root@auszogesrusd002:/root # lanadmin -x 0
Speed = 10000 Full Duplex.
root@xxxxx:/root # lanadmin -x 1
Speed = 10000 Full Duplex.
root@xxxxx2:/root # lanadmin -x 15
Speed = 10000 Full Duplex.
root@axxxxxx2:/root # lanadmin -x 4
Speed = 10000 Full Duplex.
root@axxxxxxxx02:/root #
Virtual Connect Manager 4.20 hangs - Loading, please wait
Hello
When I want to run connect manager for my HP Virtual Connect Flex-10 / 10G switch, the login screen keeps showing "loading please wait". Nothing else happens. I have tried other browsers (Firefox version 24, IE 11,10 ..) As it is in the recommendations and unfortunately no reaction. This panel is in version 4.20. I read various posts for versions up to 4.01 because others had similar problems but unfortunately it does not help. Can someone help me?
Replacing HP 4Gb VC-FC Module with HP VC 8Gb 24 - Port FC Module
Hi,
i would like to replace my productive environment on Bay 4+5
- 2 x old "HP 4Gb VC-FC Module" 409513-B21 with
- 2 x new "HP VC 8Gb 24 - Port FC Module" 466484-001 / 466539-001
The Firmware of the new HP VC 8Gb matches the requirement of the enclosure (I can update the firmware in a second, non-productive C7000).
Question: How should I replace the modules? What's the right order? Is there a downtime (FC) necessary?
best regards
Danny
upgrade firmware c7000 g3 to version 4.62 and firmware nic
I have a bl460c g9 chassis with virtual connect two module: c7000 g3 with firmware 4.41
My goal is to upgrade to 4.62
If upgrade virtual connect to 4.62 i can have 536flb cards with xxx old firmware.?
My questions:
1) i can upgrade virtual connect to 4.62 and to mantain 563flb with old firmware for compatibility with OS?
2) for total upgrade chassis the correct sequence is? upgrade virtual connet and next step upgrade nic? or
upgrade total nic and next step upgrade virtual connect? ( i not find document for which answers these questions.
thanks for support
Strange not assigned/mapped network interface from VC on Server 2016
I did assign/map 9 networks at VC and now I can see 10 network interfaces at Server 2016... Extra network interface has also "strange" MAC not starting with 00-. I can see this extra interface also at ILO with status unknown. Server 2016 does show interface as Network cable unplugged. I cannot remove that interface because it is not mapped on VC. How I can remove extra interface? I Did attach snip.
VC firmware 4.60, G8 blade, HP FlexFabric 10Gb 2-port 554FLB Adapter, drivers should be latest.
EDIT:
I did notice that all other Blades will show also that same interface as Network cable unplugged.
EDIT2: I did remove 9th interface from VC and now extra interface went away from Server 2016 and also from ILO. Bug does not show up if there is only 8 interfaces... so maybe I do not need to mind about extra interface which does not work... Maybe it will work also if I add enough interfaces :)
Virtual Connect FlexFabric 10/24 Port issues
Hello Everyone! I am currently experienceing some weird issues with a VC 10/24 port FlexFabric interconnect.
Quick Spec's
OA FW= 4.85
571956-B21-VC FW= 4.62, Bay1&2
BL460c G7= Bios-l27 5/5/2011, PMC 1.6, iLO-3 1.90. Emulex NC553i Flex NIC in LOM1-a & LOM2-a
ESXi version 6.5 (Updated)
----------------------------------------------------------------------------------------------
Current configuration:
Server Profile: Ethernet network bound, no special configs
Eth Network: Tunnel Vlan checked
UPlink: X1 using Cisco DAC to upstream switch 10gb port. Configured as Trunk with all vlans allowed
ESXi Host has Vlan tagging on management vmkernel port
------------------------------------------------------------------------------------------------
I can see on the VC that all the upstream switch MAC addresses are being learned on the VC itself. So layer 2 seems to be up and running just fine.
When I assign the ESX host to use the 10GB FlexFabric NIC @ VMnic0, TCP/IP traffic does not pass through the VC domain.
I also tried Disabling all vlan tagging and tunnelling, Configured a single access port in a single vlan and still no traffic goes through.
Verified NC553i is compatible with VC listed above.
I just cant figure out why traffic will not pass from ESX -> VC -> upstream switch. I think the issue is between Blade and VC. I have also tried Shared Uplinks and ensured vlans were tagged on both sides of the equation.
I have also tried installing the latest Emulex NIC drivers from VMware with no luck.
Thoughts?
Centos 7.0.1406 does not see FlexFabric NIC's during install
Hello,
We have an BladeSystem c7000 Enclosure G3 populated with Gen8 and Gen9 ProLiant BL460c blades. Each blade has HP FlexFabric 10Gb 2-Port 534FLB Adapter installed and we setup 8 NICs on each server. One of my blades that I am trying to add to my oVirt installation does not see all eight NIC's during the CentOS 7.0.1406 installation. I have recently built three other blades, all the same configuration and ROM versions without issue.
The HP OA sees them fine, and the server profile has them added - however when I go to the networking tab of the CentOS installation, only ETH0 and ETH1 are showing.
Are there any settings in the BIOS of the blade that could prevent it from seeing the NIC's during OS installation? Any where else I should be looking to resolve this?
I have asked the company that I purchased the blade / adapter from to RMA the adapter first to see if it may be faulty.
System Info:
Blade Part No: 735151-B21
Blade ROM version: I31 06/01/2015
Virtual Connect for the blade reporting 'Configured for Virtual Connect'
Thank you for your assistance.
Mark Steele
Networking for HPE C7000 and Cisco Nexus switches
Good morning everyone!
I'm looking at getting a new C7000 chassis with dual 10GB VC modules connected up to a pair of Cisco Nexus 9000 series switches.
Right now, I've got one port from each VC module connected to each of the Nexus switches, with a VPC peer link across the switches, as below:
However, there are a couple of things I'm unclear of:
- Should I change the two individual links from each VC module to a LACP/VPC link across the two Nexus switches?
- I notice the ports on the VC are broken down into two pairs and a group of four. Does it make any difference which ports I use on the VC modules, for LACP or otherwise?
Thanks in advance for any help with this.
change the enclosure ID in the Domain Settings (C7000)
Hi,
Is it possible to change the Enclosure ID in the Virtual Connect?
We started to monitor our devices, and we have 5 enc0 (because we have 5 C7000 bladesystems.)
Thank you for any advice!
Zoli
FlexFabric and Virtual Connect
Hi,
Can anyone explain the difference between FlexFabric and Virtual Connect?
Help me to find virtual connect
c3000 mixing VC and non-VC - Not in adjacent bays
Hi,
I have found two setup and installation guides that are more or less the same:
http://h20628.www2.hp.com/km-ext/kmcsdirect/emr_na-c04731477-1.pdf
http://h20628.www2.hp.com/km-ext/kmcsdirect/emr_na-a00018702en_us-1.pdf
and they are not conclusive (page 15) on the following re supported configurations:
In a C3000, can I I have two VC Flex-10 modules in Bay 1 and Bay 2, and have two non-VC (i.e. GBE2c switches) in Bay 3 and bay 4 at the same time? I know I cannot mix in Bay 1/2 or Bay 3/4.
One of the PDFs says that both for a C3000 or a C7000 "*"Other" indicates any VC interconnect module for c-Class BladeSystem".
The other PDF says that this is only true for a C3000 and that for a C7000 "In the following tables, "Other" indicates any c-Class interconnect module including a VC, Pass-Thru, or switch"
Anyone has a config running with VC modules in Interconnect Bay 1 and 2 and non-VC in Bay 3 and 4?
Thanks,
Ruben
Virtual Connect Server Profile - Mapping Lost
We've had two BL460c Gen9 servers in c7000 G1 & G2 enclosures temporarily lose mappings to the Ethernet Adapter Connections and FC HBA Connections under their server profiles; the status was listed with a warning icon. Shutting down and reseating the servers in the device bay allowed the server to restore connectivity and the status for those mappings went green.
Based on the previous troubleshooting we've done with HPE, this might be symptomatic of a compatibility issue between the BL460c Gen9 and the G1 & G2 enclosures that are running older VC modules (1/10Gb VC-Enet & VC 4Gb FC).
Our long-term fix is to retrofit our G2 enclosures with VCFF-20/40 F8 modules, replace the oldest G1 with G3 enclosures outfitted with VCFF-20/40 F8 modules, then migrate the newer BL46x Gen8/Gen9 servers to the G2 or G3 enclosures.
Has anyone out there been experiencing a similar intermittent mapping loss in their server profiles? Thank you.
HP VC Flex-10 Enet - Active/Active not working
Hi,
I am trying to achieve the Active/Active setup from example #11 in the Flex-10 cookbook, but always only one link is set to active the other remains in standby. By the way LACP to the connectet HP 5900 isnt working too - only the static method.
What i did:
1. I assigned X5 and X6 from VC 1 to a SUS and connected them to the uplink switch on a static LAG.
2. I created a server profile and assigned the created SUS to it.
The result is only one active link - the other in standby.
Shouldnt both of the links be in actice state?
Why ist dynamic LACP not working? When enabling dynamic LACP with stp edged-port in the 5900, both links remain in unselected state.
I read many posts in this forums - some said that actice/active ist only possible with a SUS builded from two VC bays (expample #11 does the opposite). Then i read on some posts that because VC behaves not like a switch dynamic LACP will not work.
So what is the correct way for setting up an active/active connection to an uplink switch with dynamic LACP enabled?