On my manager I cannot edit any option as everything becomes grey if I click on them. On this exemple, I am trying to add a Network to my erver profile and it just becomes completely grey. I can still go back to the menu and start over but it alway's stops at the same places. I can still connect and do certain things, but the moment I want to edit anything it just freezes like this. 

Hi,

Not only am I new to the forums but I'm also new to Virtual Connect.

Currently I'm in the process of designing/deploying a new environment which consists of:

- C7000 enclosures
- Virtual COnnect swicthes in bays 1, 2 + 5, 6
- BL460c Gen 9 Blades with 1 x 536FLB (LOM) and 1 x 534M in Mez slot 2 (both dual port
- All firmware/drivers deployed from the latest support packs

I define the following NPARs in the server profile:

LOM 1a - vmnic0
LOM 2a - vminc1
MEZ 1a - vmnic2
MEZ 2a - vmnic3
LOM 1b - vmnic4
LOM 2b - vminc5
MEZ 1b - vmnic6
MEZ 2b - vmnic7

That works and the ESXi 6.5 hosts see all the nic's that Virtual Connect is presenting to the blade. The problem is that more often than not, a number of the nic's show as disconnected in ESXi. We did find this link from Jan 2016 http://h20564.www2.hpe.com/hpsc/doc/public/display?docId=emr_na-c04942515 (describes our issue perfectly) but I have not been able to find anything since that states the issue has been resolved.

Is anyone else seeing this with newly deployed kit??? Is there something I'm not thinking about when defining the NPAR NICs???

Thanks in advance

nikrik

Hi everyone.

Over the weekend, I upgraded the firmware on our HP OA, including the VC firmware.  When it was completed, only one of the two VC-FC modules had been upgraded.  The other shows as incompatible and nothing I've tried will get it to accept the newest firmware.  I couldn't find any older versions on HP's web site, and I've tried upgrading from OA, VCSU and HPSUM.

As an added note, even though I fed VCSU an account with all privilleges, it said I didn't have firmware upgrade permissions.  I even tried the Administrator account with the same results.

Any idea how I can get this module up and running again?

Thanks in advance for any suggestions.

Derek

Hi,

New to the forums and to Virtual Connect and already, on a completly new environment being built, I have an issue that I cannot find clarity on.

So, what we have is the following:

- C7000 Enclsoure

- Virtual Connect Swicthes in bays 1, 2  + 5 , 6

- BL460C Gen 9 with 1 x 536FLB (LOM) and 1 x 534M in Mez Slot 2 - both adapters are dual port

I'm looking to use NPAR to define a number of adapters that my ESXi 6.5 hosts will see as different vmnics, so for example:

LOM 1a = vmnic0

LOM 2a = vmnic1

MEZ 1a = vmnic2

MEZ 2a = vmnic3

LOM 1b = vmnic4

LOM 2b = vmnic5

MEZ 1b = vmnic6

MEZ 2b = vmnic7

From what I can tell, all firmware/drivers are current.

The problem comes after applying the Server Profile and rebooting, there are times that the ESXi host shows some of the NICs as disconnected.  I found this  link http://h20564.www2.hpe.com/hpsc/doc/public/display?docId=emr_na-c04942515 which describes our issue, but it is dated Jan 2016 and I have not been able to find anything newer to say that the issue has been resolved.

I guess others have seen this in the past, but is there an actual resolution to the problem as Jan 2016 seems a long time to go without a fix.

Thanks in advance

Nikrik

Hi all,

We have two c7000 chassis with a pair of 10/10D modules and a pair 16gb 24port Fc modules in each chassis all fully populated with SFP's to used for VMware ESXi.

For the 10/10D modules is there any advantage of me using the VC domain stacking (apart from centralised management) and is there a limit on the number of stacking cables I can run (could I run 4 per module between the chassis' and the remaining 6 to uplink) or am I best running all the connections as uplinks to the core switches for throughput?

Any replies appreciated, thanks in advance!

On September 12, version 4.61 of the Virtual Connect firmware was released, it can be downloaded here:

https://h20566.www2.hpe.com/hpsc/swd/public/detail?sp4ts.oid=7271228&swItemId=MTX_83557b0e132046f7811a51a8d1&swEnvOid=4183#tab4

The Release Notes tab advises that "The 4.61 version of HPE Virtual Connect Release Notes can be found in the following URL: http://www.hpe.com/info/vc/manuals

but when I navigate to the URL I am redirected to the Hewlett Packard Enterprise Information Library here:

http://h17007.www1.hpe.com/us/en/enterprise/servers/solutions/info-library/index.aspx?cat=bladesystem&subcat=vc

and there is absolutely nothing there related to version 4.61.  There is documentation for the 4.60 release in July 2017, but nothing so far as I can see for 4.61.   Does anyone know where I can view the VC 4.61 release notes?

Setup:

1 - HP Bladecenter C7000

9 - bl465 g7 and bl685 g7 blades w/ standard dual emulex 10gb flexnics

1 - HP  1/10gb VC-Enet Module

I am running this enclosure from my home.  So I am using a cablemodem/router from my cable company with a small netgear gigbit unmanaged switch in between.  I have network cables plugged into the first 2 ports on the Interconnect card.

ive set up virtual connect manager to the best of my knowledge.  I am able to get an ip just fine on my first blade that is running windows server 2016.  My second blade running a linux live cd w/ correct nic drivers installed just wont pull an IPv4 address.  I just show 100% dropped Rx packets uder ifconfig.

Ultimately What I am wanting set up is a fully populated blade center. The first server will house a hard drive and a pxe server and I want to network boot a live cd to ram (diskless) for the other 15 servers.

Any help would be greatly appreciated, I have been fighting with this for several days and I need to get this setup.  I am not sure if maybe I need a better switch or router?

Thanks in advance

There are two VC switch.

My VC-SW-1 is connect to HP-sw-1 and VC-SW-2 is connect to HP-sw-2.

Now when I poweroff the one of the VC-sw from OnBoard Administrator GUI my blades become unreachable.

I have enabled smart links for Ethernet Network as well as Associated Network of Shared Uplinks.

Kindly, share your expirence, if there is something I missed to enable to make the VC-Switches redundant.

I have two flex-10 VC switches in a c7000.  I have a snmp v3 user set up on them.  On 20 September, I was able to run an snmp query and see the status of the external ports.  Today, when I try, I get "invalid user."  I've logged back in and checked.  The setup looks correct.  Any ideas on why it is not responding to a valid snmp query it did just a short time ago?  Any suggestions on how to troubleshoot?

Can anyone share knowledge of the root cause(s) of the VC Manager Major error 6030 - "Profile could not be assigned to powered up server".   I got bit by this recently while doing Virtual Connect firmware upgrades (HP VC Flex-10/10D Module - firmware 4.45 to 4.50) and several servers dropped off the network after the interconnect modules reset.   Reviewing the VC Manager system log shows 6030 Major error - while the VCSU log doesn't indicate any issues with the firmware upgrade itself.   Power-cycling the affected servers brings them back on the network, but a firmware upgrade on the interconnect modules should be non-impacting to the servers if the server profiles provide redundant network ports and the server O/S properly handles the temporary link loss on each port.

-----------------------------------------------------------------------
Record    : 3136
Date/Time : 2017-10-06 01:04:26-04:00
Info      : PRO:ServerProfileBay10:6030:Major
Message   : Profile could not be assigned to powered up server : Bay 10
-----------------------------------------------------------------------
Record    : 3137
Date/Time : 2017-10-06 01:04:29-04:00
Info      : PRO:ServerProfileBay11:6030:Major
Message   : Profile could not be assigned to powered up server : Bay 11
-----------------------------------------------------------------------
Record    : 3138
Date/Time : 2017-10-06 01:04:32-04:00
Info      : PRO:ServerProfileBay12:6030:Major
Message   : Profile could not be assigned to powered up server : Bay 12
-----------------------------------------------------------------------
Record    : 3139
Date/Time : 2017-10-06 01:04:34-04:00
Info      : PRO:ServerProfileBay15:6030:Major
Message   : Profile could not be assigned to powered up server : Bay 15
-----------------------------------------------------------------------

Hi All,

i have 4 HPE Virtual connect FlexFabric 10G/24 Module's in my environment , i have patched the latest VC firmware 4.60  to fix SSH Server CBC Mode Ciphers Enabled  but it  fixes only 3des-cbc,blowfish-cbc and rest of the things are not fixed 

aes128-cbc
aes192-cbc
aes256-cbc
rijndael-cbc@lysator.liu.se

 cast128-cbc

Here is my questions 

1. is there any ssh configuration file in VC to disable the above encryption algorithms 

2. if there is no ssh configuration file in VC is it possible to  disable  ssh login in VC

It would be highly appreciated if you could help me on this.

Thanks and Regards

Vinothsaran

Hi All,

Here is my question regarding SSL certificate in VC flex fabric 10g/24 module

1. is it possible to create self signed cerificate for VC

2. if it possible to create the self signed certificate with strong hashing algorithm 

3. what are the steps to be done before and after importing the self signed certificate in VC

Thanks in advacnce for your help

Regards

Vinoth

Hello All,

First (Bay 3) HP Virtual Connect 4Gb FC Module (409513-B22) lost communication with Onboard Administrator (No IP assignment) and as a result the Enclosure got degraded status. The general  VC FC module  status was still OK (see below)

Status Status  OK Thermal Status  OK Powered On

Diagnostic Information Device Identification Data   OK Temperature   OK Overheat Check   OK I/O Configuration   OK Power Allocation Request   OK

A couple of reset, power off/power on operations did not change the situation. We decided to replace suspicious VC FC module by using the same module model from another HPE Blade Enclosure. Unfortunately there is still  "No communication" error and additionally used Uplink ports have "NOT-LOGGED-IN" connector status. For the sake of completeness  the firmware version is the same on both VC FC modules.

It seems that simple replacement of VC 4Gb FC Module is unsuccessful. Could someone explain what can be a reason of "No communication" error and whether any additional steps are needed in case we replace VC FC module?

Is there any special procedure in order to add new VC FC module to the existing VC domain?  We use HP Virtual Connect Enterprise Manager.

Regards

Hello,

one of our customers has a new c7000 enclosure with BL460 G9 Blades. The enclosure has 2 FlexFabric 20/40 F8 Virtual Connect modules (FW 4.50). The blades have 630FLB Adapters. LOM1:1-b and LOM1:2-b are configured as FlexHBAs and have factory-assinged WWNs.

The Fabric Uplinks are connected using Ports X5 and X6. Both ports of Bay 1 are connected to one fabric, both ports of the Bay 2 are connected to another fabric. The SAN Switches are Brocade 5100 running FOS 7.4.1 . The SAN Ports are NPIV-capable and enabled and have Fill-Word 3 set. The uplinks are online and in VC manager I can see the WWN of the connected SAN switches.

But no FC NPIV device logs into the SAN. All I can see are the Port WWNs of the FlexFabric Ports (e.g. 20:04:00:11:0a:04:7e:80). The port should be reporting the number of NPIV devices, but there are none logging in. I've tried disabling/enabling the ports and running a redistribute on VC manager.

Running a nodefind command in the SAN for the WWPN of the HBAs returns nothing. I don't see these devices anywhere in the SAN. The Server Adapter port reports not logged in.

Any ideas as to what is going wrong here?

Thanks in advance.

Steve

Hi everyone,

I've configured what I thought was an active/active setup from side A and B, but when testing failures, I've found that my F8 modules are in active/standby mode which causes an issue with chassis that have one QSFP from side A and one QSFP from side B to my 5940 switches.

Just for a little overview, I have two 5940's in a IRF stack, which I'll refer to as switch 1 and switch 2, but it's important to note they are a single fabric. So in the above exmaple I have chassis side A F8 module connected to switch 1, and chassis side B F8 module connected to switch 2. Side A F8 module is currently active, and side B F8 module is currently in standby mode. VC is in full stacking mode.

When I simulate a power failure on switch 1 in the above scenario, because there is now only a single link between chassis side B F8 module and switch 2, my servers now can't talk. Within the shared uplink set it looks like everything is ok, the links are in an active state, but traffic is not being passed. I'm assuming is because of this active/standby module scenario.

This leads ot my question; should I be configuring external stacking links between the modules to facilitate the configuration I want, or should I change VC to horizontal stacking mode?

My shared uplink sets are in a active/active configuration. My ideal scenario is to have internal chassis server communication to happen across the modules in the chassis, and chassis to chassis comminucation to go to my 5940s. I have 6 chassis, but possibly I should be daisy chaining these modules together so inter-chassis comms don't even go to my 5940s? Also my chassis are all going to be part of their own domains. 

Potentially this isn't even my problem. I've played around with stacking now and I still can't get one side to work. The config is identical to a chassis I do have working, although that has multiple QSFP's from each F8 module to switch 1 and switch 2, so when one switch does down each module still has a path to the remaining switch, where in the above scenario there isn't.

Help :)

Greetings. Newbie to HP C7K.

I have the following configuration:

1. BL460c blades with 536 FLB and a 560M in Mezz2.

2. VC Flex-10/10D in bay 1 & 2.

3. Passthru in 5 & 6 for 560M (Unused for now)

I am trying to create a configuration where I add 4 (Flex)NICs to the blade. 2 ports (w/ bonding) to carry untagged OAM traffic (vNet_10_61 in attached image1). And 2 ports to carry VLANs. A combination of Scenario 1 & 7 in the Cookbook.

However when I attach the profile to the server, it maps port 3 & 4 to Mezz2 rather than to the LOM. This is even when I pull out the passthru modules. Somehow I cannot see the detailed flex-nic mapping after clicking the server profile (image2), but the CLI confirms the mapping.

I tried Factory v/s VC MACs, capping max speed limits etc.

I added two more ports and kept port 3 & 4 unassigned. And that worked! But I cannot find such a recommendation in the User Guide.

Q1: Is that the recommended method of forcing a mapping?

Q2: Does leaving such ports unassigned have any other effect? For e.g. their usage with passthrus?

VC Firmware: 4.60 (also seen on 4.50)

Guys,

Were currently running VC's running version 4.41 and we have problems collecting logs from these, apparently a known issue on this firmware. So ideally we need to upgrade these to a newer version.

We currently have 2 chassis in 1 VC Domain, and from my understanding if we upgrade these it will need to stop servicing blades for a while while the upgrades completes.

I'm not going to get a chance to power off both enclosures to allow the upgrade, so can anyone else suggest another idea? If we got 4 new VC's running a newer o/s would we get away with putting out config on these and swapping the VC's 1 at a time. IOR would the chassi complain that the running VC has older firmware than the newly inserted one?

Thanks

Hello i need your help.

We have a C7000 with two Virtual Connect Module with ESX 6.0, and we configured two shared links, with four nics as multiple networks, and the configuration works fine with the VMware and shows four vmnics: vmnic0, vmnic1(vSwitch0), vmnic2, vmnic3(vSwitch1), but when we try to do the test for active/active virtual connect, only the vSwitch(vmnic0 and vmnic1)  works, the vSwitch(vmnic2, vmnic3) does not work.

So the vSwitch0 can works when the virtual connect was turn-off, but the vSwitch1 fails all the vmnics at different virtual connect.

So we need your help, why this happen?, how can we fix this issue?

Hi,

I have created shared uplink sets named VC_A and VC_B. VC_A is connected to port 1.1 on switch A and VC_B is connected to port 2,1 on switch B and  both are in seperate lacp. Both are showing as linkes Active on Virtual connect.

But while testign failover. When i disconnect VC_B , traffic keeps flowing through VC_A without any issues.

But as soon as i disconnect VC_A cable, traffic stops and network goes down., I have enable smart link and made sure that fast cache option is disabled in advance settings as well.

Please suggest, i can share more details if required.

Greetings. I am a newbie to HP C7K. I have the following configuration:

1. Single HP C7K. With six BL460c blades. In bay 1-3 and bay 9-11.

2. VC Flex-10/10D in bay 1 & 2. VC1 has 2 ports in LAG to switch 1 and VC2 has 2 in LAG to switch 2.

I sometimes see that "Domain Status" indicates that "Stacking Links" degraded. Trying to narrow it down. I have three probable candidates:

1. When I power on bays in particular order? Or change server profile config without un-assigning (only powered off: that's allowed right?)

2. When the OS bonding config is not complete/correct.

3. VC detecting a loop in my configuration?

Am I even making sense? Where can I read more about how the internal stacking links work? The X11-X14 are essentially uplink ports, right? In one LAG to the next bay peer, in my case?

->show stackinglink                 

Connection Status : Failed
Redundancy Status : Degraded

Stacking Mode : Full

No stacking links to display

Earlier:

->show stackinglink 

Connection Status : OK
Redundancy Status : OK

Stacking Mode : Full

===========================================
Link  Speed  Connected From  Connected To  
===========================================
1     10Gb   enc0:1:X11      enc0:2:X11    
2     10Gb   enc0:1:X12      enc0:2:X12    
3     10Gb   enc0:1:X13      enc0:2:X13    
4     10Gb   enc0:1:X14      enc0:2:X14    

I am not sure, but I think--amongst the things I have tried--VC reset fixes this.

Thanks in advance