I found the reason and a solution. In in the vbackupd-vddk.log there are these messages which show, that there ports were blocked through a firewall.It was determined that the ESXi hosts did not have port 902 open, to allow vProxy communication for backup/recovery operations.
VDDK: INFO: CnxOpenTCPSocket: Cannot connect to server :902: Connection timed out VDDK: INFO: CnxAuthdConnect: Returning false because CnxAuthdConnectTCP failed VDDK: INFO: CnxConnectAuthd: Returning false because CnxAuthdConnect failed VDDK: INFO: Cnx_Connect: Returning false because CnxConnectAuthd failed VDDK: INFO: Cnx_Connect: Error message: Failed to connect to server :902 VDDK: WARN: [NFC ERROR] NfcNewAuthdConnectionEx: Failed to connect to peer. Error: Failed to connect to server :902 VDDK: INFO: NBD_ClientOpen: Couldn't connect to :902 Failed to connect to server :902 VDDK: INFO: DISKLIB-DSCPTR: : "vpxa-nfc://[xio_lun22_bo1] / .vmdk@ :902" : Failed to open NBD extent. VDDK: INFO: DISKLIB-LINK : "vpxa-nfc://[xio_lun22_bo1] / .vmdk@ :902" : failed to open (NBD_ERR_NETWORK_CONNECT). VDDK: INFO: DISKLIB-CHAIN : "vpxa-nfc://[xio_lun22_bo1] / .vmdk@ :902" : failed to open (NBD_ERR_NETWORK_CONNECT). VDDK: INFO: DISKLIB-LIB : Failed to open 'vpxa-nfc://[xio_lun22_bo1] / .vmdk@ :902' with flags 0x1e NBD_ERR_NETWORK_CONNECT (2338). VDDK: INFO: VixDiskLib: Unable to locate appropriate transport mode to open disk. Error 13 (You do not have access rights to this file) at 4570.
Logon to your vCenter and right click on the issued VM, navigate to Tags & custom Attributes click edit custom attributes, delete register EMC vProxy Session and hit ok. Go back to networker, start the the job again and it should work. It work for me.
since our VMware admins upgraded from vSphere 6.5 to 6.7, we have frequently run into failed VM backups due to this attribute not automatically getting removed. when the cluster was still at v6.5, from the CLI of the vproxy server, running the vmconfig -c "enable-migration" used to clear the lock
we are using networker v19.1 and we are running the same version for the vproxies.
is there an issue (or conflict) with vsphere v6.7 ?
the ProxyHC tool is also very helpful to solve some issues that can occur with vproxy based backups by being able to perform various checks from the vproxy end towards the backup infra and the VC specifically.
You out the tool onto a vproxy and run it as root. it can be used to check connectivity towards VC, datadomain and the ESXi hosts, VC permissions, get rid of orphaned processes, get rid of vmdk's still assigned to the vproxy and more.
Been using it for some time already and also reached out directly to the Dell developer whose contact info is stated in the tool output. On request for example he added functionality to use it in a scripted way instead of only interactive.
onestoned
1 Rookie
•
49 Posts
1
June 13th, 2017 05:00
Hello all,
I found the reason and a solution. In in the vbackupd-vddk.log there are these messages which show, that there ports were blocked through a firewall.It was determined that the ESXi hosts did not have port 902 open, to allow vProxy communication for backup/recovery operations.
VDDK: INFO: CnxOpenTCPSocket: Cannot connect to server :902: Connection timed out
VDDK: INFO: CnxAuthdConnect: Returning false because CnxAuthdConnectTCP failed
VDDK: INFO: CnxConnectAuthd: Returning false because CnxAuthdConnect failed
VDDK: INFO: Cnx_Connect: Returning false because CnxConnectAuthd failed
VDDK: INFO: Cnx_Connect: Error message: Failed to connect to server :902
VDDK: WARN: [NFC ERROR] NfcNewAuthdConnectionEx: Failed to connect to peer. Error: Failed to connect to server
:902
VDDK: INFO: NBD_ClientOpen: Couldn't connect to :902 Failed to connect to server
:902
VDDK: INFO: DISKLIB-DSCPTR: : "vpxa-nfc://[xio_lun22_bo1]
/ .vmdk@ :902" : Failed to open NBD extent.
VDDK: INFO: DISKLIB-LINK : "vpxa-nfc://[xio_lun22_bo1]
/ .vmdk@ :902" : failed to open
(NBD_ERR_NETWORK_CONNECT).
VDDK: INFO: DISKLIB-CHAIN : "vpxa-nfc://[xio_lun22_bo1]
/ .vmdk@ :902" : failed to open
(NBD_ERR_NETWORK_CONNECT).
VDDK: INFO: DISKLIB-LIB : Failed to open 'vpxa-nfc://[xio_lun22_bo1]
/ .vmdk@ :902' with flags 0x1e
NBD_ERR_NETWORK_CONNECT (2338).
VDDK: INFO: VixDiskLib: Unable to locate appropriate transport mode to open disk. Error 13 (You do not have access
rights to this file) at 4570.
thanks for your help
Karthik_N1
25 Posts
0
June 12th, 2017 08:00
Most of the VADP cases I have seen this error.
We used to just disable "Client Direct" of clients which have this problem.
Keeping this in mind, check connectivity from vProxy to Data Domain(Ports, lookups) etc.
Also I see that the transport mode used is NBD, you can try hotadd and see how it goes.
Thanks,
Karthik N
onestoned
1 Rookie
•
49 Posts
0
June 12th, 2017 09:00
thanks a lot.
please keep in mind, it's a VMWare backup with Networker vproxy. There is no possibility to disable "Cleint direct" as we are doing a backup of a VM.
Name resolution is working properly.
Hotadd is not possible, as no disks are visible by the vproxy. So we are using nbd.
any more thoughts?
thanks a lot
Dave_Hannover
3 Posts
0
September 5th, 2020 10:00
I had the issue and this is how I got mine fixed.
Logon to your vCenter and right click on the issued VM, navigate to Tags & custom Attributes click edit custom attributes, delete register EMC vProxy Session and hit ok. Go back to networker, start the the job again and it should work. It work for me.
cheers
David
Pezdaddy
1 Message
0
April 27th, 2021 06:00
since our VMware admins upgraded from vSphere 6.5 to 6.7, we have frequently run into failed VM backups due to this attribute not automatically getting removed. when the cluster was still at v6.5, from the CLI of the vproxy server, running the vmconfig -c "enable-migration" used to clear the lock
we are using networker v19.1 and we are running the same version for the vproxies.
is there an issue (or conflict) with vsphere v6.7 ?
Dave_Hannover
3 Posts
0
April 28th, 2021 02:00
crazyrov
4 Operator
•
1.3K Posts
0
April 28th, 2021 02:00
@Pezdaddy, According to the compatibility matric VMware 6.7 is compatible with NW 19.1 with vProxy 4.0.0-x, 4.3.0-x. Check this link for more details - https://elabnavigator.emc.com/vault/pdf/NW_All_19_1.pdf?key=1612451110407
Dave_Hannover
3 Posts
0
April 28th, 2021 02:00
barry_beckers
393 Posts
0
May 3rd, 2021 13:00
the ProxyHC tool is also very helpful to solve some issues that can occur with vproxy based backups by being able to perform various checks from the vproxy end towards the backup infra and the VC specifically.
ftp://ftp.legato.com/pub/NetWorker/NVP/Tools/ProxyHC
the ProxyHC tool:
https://youtu.be/dlutigOJLGg
You out the tool onto a vproxy and run it as root. it can be used to check connectivity towards VC, datadomain and the ESXi hosts, VC permissions, get rid of orphaned processes, get rid of vmdk's still assigned to the vproxy and more.
Been using it for some time already and also reached out directly to the Dell developer whose contact info is stated in the tool output. On request for example he added functionality to use it in a scripted way instead of only interactive.