nra1941
Enthusiast
Enthusiast

vSphere 6.0U1b can't connect to newly created vVOL ProtocolEndPoint

   I downloaded the latest vVNX to test the vVOL, my lab has 3 hosts with ESXi6 U1b and vCSA 6.0.0.10200. The vVNX is deployed to another esxi 6 and with one virtual HDD ( carved from 240G SSD drive using Thick provision Eager zero) is used to create a single storage pool, I strictly followed the instructions demonstrated in the following video, I did not create extra Storage Profile beyond these required for setting up vVOL. When vVOL is added to the three hosts, everything seems to be fine, but when I look at the storage size, it is 0 B, but when I looked at the storage devices attached to esx1 host, notice it says "inaccessible".   I also chchecked the time, my vCenter, ESXi, vVNX all NTPed to the same server, and their time skew is less than 1 min.

   For these small 3 host lab, I did not configure anything thing such as LUN masking or path selection rules. Also the Protocol Endpoint ip address can be seen from the hosts and vcenter,  When I ssh to any host and type

" esxcli storage vvol protocolendpoint list    " , the result is blank, so it appears that ProtocolEndPoint has some issues,  but yesterday, when I had two storage pools, for 30 min, everything worked ( I later destroyed that vVOL due to other mistakes).

    Does anyone have clue?  I mean the vVOL ver1.0 does seem to have some rough edges.

q.jpg

q2.jpg

EMC World 2015: vVNX VVOL Demonstration - YouTube

8 Replies
vvenu
Enthusiast
Enthusiast

Most likely to do with VASA provider and you may find something in there.

0 Kudos
nra1941
Enthusiast
Enthusiast

This is EMC vVNX, a virtual storage appliance, I have no idea how to install/uninstall VASA, I thought the VASA 2.0 is embedded into the storage appliance. The link u provided also shows no VASA downloadable from EMC. 

0 Kudos
vvenu
Enthusiast
Enthusiast

Haven't done VVOL on vVNX but in general you need it to have communication between vCenter and storage. 

Check this link:

http://lab.piszki.pl/emc-vvnx-testing-vvol-in-vsphere-6-part-two/

0 Kudos
nra1941
Enthusiast
Enthusiast

Thanks, I have already checked this Polish site and other sites, including EMC/VMware video regarding vVNX, really think this might be a bug.  Considering how many Single Point of Failure exist in VMware products, I wonder if their cavalier attitude toward software quality is REALLY designed/planned to generate more support revenues? :smileydevil: :smileydevil:

0 Kudos
nra1941
Enthusiast
Enthusiast

Tonight I tried it again, vVOL now shows the right size, which is 150GB, and it has connection to all three hosts, but only esx3  has sustainable connection, esx1, esx2 quickly lose connection after 305 minutes.

I went to /var/log/vvold.log and I found the following, the same thing was not found in the vvold.log on esx3.  Looks to me there are some issues with VASA, but I don't know how to resolve this issue, as of right now, HA has been turned off, and there is really no different between esx1, esx2 and esx3.

The esxi6 server (Lenovo H50-55 Desktop, with dual-core AMD A8‑7600 CPU) where I installed vVNX has 16G RAM, and it is partially shared with VRAM, and I managed to create an vVNX VM with 12GB RAM, but the RAM is tight.  On the same esxi6 host, I installed another copy of vVNX without VVOL, two iSCSi LUNS were created without any issues. This esxi6 server is NOt managed by the same vCenter managing esx1/2/3, in fact, I am using C# client to manage it.

On esx1 and esx2, when I run command:  esxcli storage vvol protocolendpoint list,   it is still empty

On esx3, this is what I got:

rfc4122.822ebe79-f151-4374-bbd2-e22fbd93e483

    Host Id: e0992541-a79e6176

    Array Id: EMC:VIRT1609T88K43

   Type: NFS

   Accessible: true

   Configured: true

   Lun Id:

   Remote Host: 192.168.1.72

  Remote Share: /NAS_PE

  Storage Containers: 24790a2e-47b8-4d76-aab7-570bcb008dcf

/var/log/vvold.log   content: 

2016-03-08T06:09:29.230Z info vvold[FFFAFB70] [Originator@6876 sub=Default] VasaSession::Initialize url is empty

2016-03-08T06:09:29.230Z warning vvold[FFFAFB70] [Originator@6876 sub=Default] VasaSession::DoSetContext: Empty VP URL for VP (VASA)!

2016-03-08T06:09:29.230Z info vvold[FFFAFB70] [Originator@6876 sub=Default] Initialize: Failed to establish connection https://192.168.1.70:8443/vasa/version.xml

2016-03-08T06:09:29.230Z error vvold[FFFAFB70] [Originator@6876 sub=Default] Initialize: Unable to init session to VP VASA state: 0

2016-03-08T06:09:46.117Z info vvold[FFF6EB70] [Originator@6876 sub=Default] VVolUnbindManager::UnbindIdleVVols called

2016-03-08T06:09:46.117Z info vvold[FFF6EB70] [Originator@6876 sub=Default] VVolUnbindManager::UnbindIdleVVols done for 0 VVols

2016-03-08T06:09:59.245Z info vvold[FFD672E0] [Originator@6876 sub=Default] VasaSession::GetEndPoint: with url https://192.168.1.70:8443/vasa/version.xml
2016-03-08T06:09:59.252Z error vvold[FFF6EB70] [Originator@6876 sub=HttpConnectionPool-000000] [ConnectComplete] Connect failed to <cs p:1f07f7a8,
          TCP:192.168.1.70:8443>; cnx: (null), error: N7Vmacore3Ssl18SSLVerifyExceptionE(SSL Exception: Verification parameters:

          --> PeerThumbprint: 3C:A5:90:95:C0:54:C1:5A:7C:C5:2D:BD:3F:25:C3:63:AE:01:A9:BA

          --> ExpectedThumbprint:

          --> ExpectedPeerName: 192.168.1.70

          --> The remote host certificate has these problems:

          -->
          --> * self signed certificate in certificate chain)

2016-03-08T06:09:59.252Z warning vvold[FFD672E0] [Originator@6876 sub=Default] VasaSession::GetEndPoint: failed to get endpoint, err=SSL Exception: Verification parameters:
          --> PeerThumbprint: 3C:A5:90:95:C0:54:C1:5A:7C:C5:2D:BD:3F:25:C3:63:AE:01:A9:BA
          --> ExpectedThumbprint:

          --> ExpectedPeerName: 192.168.1.70

          --> The remote host certificate has these problems:
          -->
          --> * self signed certificate in certificate chain, using default

vvol.jpg

0 Kudos
Marco07
Contributor
Contributor

I have the exact same issue:

VVOLd.log:

error vvold[FFA762E0] [Originator@6876 sub=HttpConnectionPool-000000] [ConnectComplete] Connect failed to <cs p:1f0a33e0, TCP:sgstorage:31443>; cnx: (null), error: N7Vmacore3Ssl18SSLVerifyExceptionE(SSL Exception: Verification parameters:

--> PeerThumbprint: F1:C9:68:55:11:D5:8F:0F:38:D0:D7:8B:D0:29:7B:B2:2B:5D:08:BD

--> ExpectedThumbprint:

--> ExpectedPeerName: sgstorage

--> The remote host certificate has these problems:

-->

--> * self signed certificate in certificate chain)

0 Kudos
Marco07
Contributor
Contributor

I have found a resolution:

Simple remove the esxi host from the inventory and readd it.

After that, the provider on the esxi host are ONLINE and the endpoints are created and working  :smileygrin:

prvnsanpla
Contributor
Contributor

This issue is related to esxi certificate. I tried to remove host from inventory and add again as suggested above. But this didn't solve my problem. Then I removed the host from inventory and add it on another vCenter. While adding vCenter asked to replace host's certificate and clicked on yes. vCenter replaces certificates and I saw the commands[/etc/init.d/vvold ssl_reset] in logs which can be used to reset the certs on ESXi host. After the certificate reset my vasa provider is working fine.

vvold logs on my ESXi host:

[/etc/init.d/vvold: /etc/init.d/vvold ssl_reset, called by pid 2430929]

info vvold[FF8F5B70] [Originator@6876 sub=Default] Reading from pipe returned 1

warning vvold[FF8F5B70] [Originator@6876 sub=Default] VVold caught signal 1

warning vvold[FF8B42E0] [Originator@6876 sub=Default] ReloadSslCert: Resetting all VASA sessions on receiving SIGHUP

info vvold[FF8B42E0] [Originator@6876 sub=Default] VendorProviderMgr::ResetVPSessions: Resetting only SSL sessions

warning vvold[FF8B42E0] [Originator@6876 sub=Default] VVolServiceInstance::SetSSLContext: Skipping CRL check as env variable VVOLD_DO_CRL_CHECK is not set!

signalled for SSL reset]

Thanks,

Praveen K.