Impact:
The inability to enable a VS2 front-end port in VPlexcli can lead to prolonged loss of redundancy on the Front End (FE) of the VPLEX, and also has the potential to impact a Generational Hardware Upgrade (GenU) of a VS2 to VS6.
Issue:
A VS2 running GeoSynchrony 6.x has a front-end (FE) port listed in the VPlexcli as having a status of
down indicating it is disabled.
Attempts to re-enable the FE port in VPlexcli (using the
set enabled true command) do not return any errors, yet the FE port remains in
down state.
Example for verifying the issue:
1] Check the FE port status by using the following command, and check what FE port(s) are down
(disabled):
In the following sample output below (we are only including the output from director-1-1-B):
- director-1-1-B B0-FC02 is down (disabled)
- this is the port with the issue
also note:
- ports B0-FC00 and FC01 are up (enabled and have connectivity)
- port B0-FC03 is showing as no-link state (enabled, up, yet no connectivity, meaning no light level from the source end)
Checking the status of the ports on all engines and ports
VPlexcli:/> ll /engines/*/directors/*/hardware/ports*
or
VPlexcli:/> ll /engines/**/ports
/engines/engine-1-1/directors/director-1-1-B/hardware/ports:
Name Address Role Port Status
------- ------------------ --------- -----------
B-CMI00 2 - down
B0-FC00 0x50001442902a0a00 front-end up
B0-FC01 0x50001442902a0a01 front-end up
B0-FC02 0x0000000000000000 front-end down <<<< port with issue
B0-FC03 0x50001442902a0a03 front-end no-link
B1-FC00 0x50001442902a0a10 back-end up
2) Attempt to enable the
down port by using the VPlexcli command
set enabled true .
a] First, cd into the context of the FE port in question,
example:
VPlexcli:/> cd /engines/engine-1-1/directors/director-1-1-B/hardware/ports/B0-FC02
b] Then issue the
set enabled true command,
example:
VPlexcli:/engines/engine-1-1/directors/director-1-1-B/hardware/ports/B0-FC02> set enabled true
VPlexcli:/engines/engine-1-1/directors/director-1-1-B/hardware/ports/B0-FC02>
Note: No error will be returned, after the command is run, the prompt returns as shown above.
c] From the 'B0-FC02" context level issue the
cd .. command to change to the
ports context level
of the director, and then run the long listing, ll , command to check the state of the port again. It will
still show with the
down status:
example:
VPlexcli:/engines/engine-1-1/directors/director-1-1-B/hardware/ports/B0-FC02> cd ..
VPlexcli:/engines/engine-1-1/directors/director-1-1-B/hardware/ports> ll
Name Address Role Port Status
------- ------------------ --------- -----------
B-CMI00 2 - down
B0-FC00 0x50001442902a0a00 front-end up
B0-FC01 0x50001442902a0a01 front-end up
B0-FC02 0x0000000000000000 front-end down <<< still shows as down like before
B0-FC03 0x50001442902a0a03 front-end no-link
A VS2 to VS6 GenU can also fail if a VS2 has a FE port in the
down state prior to the GenU.
During the GenU, the state/settings of the VS2 (including that of the FE ports) are transferred to the VS6, and then during the I/O Transfer Phase portion of the GenU, when the VS6 director is being brought up, the FE ports are enabled. If one or more VS2 FE ports are in a
down state (disabled) prior to the GenU, and especially if those ports cannot be enabled in the VPlexcli, it can lead to the I/O Transfer Phase failing. As the step of enabling the FE ports on the VS6 director will silently fail, and then a subsequent step waiting for all FE ports to have the
enabled state (on the VS6 director) will fail due to a timeout error.
Example of the error returned during the I/O Transfer Phase of the GenU:
Continuing I/O transfer with connectivity issues could cause data unavailability and system instability. Connectivity issues must be fixed before proceeding. Would you like to re-check connectivity? (Retry/Abort): retry
Transferring director-2-1-A to VS6: .....................ERROR <<<<<
Failed to complete director transfer to VS6: hardware-upgrade Evaluation of
<<hardware-upgrade transfer-io --vs6 --director transfer-io: director-2-1-A>> failed. <<<<<<
cause: Command execution failed. cause: *Evaluation of <<wait -c
[/engines/engine-2-1/directors/director-2-1-A/hardware/ports/A0-FC00,
/engines/engine-2-1/directors/director-2-1-A/hardware/ports/A0-FC01,
/engines/engine-2-1/directors/director-2-1-A/hardware/ports/A0-FC02,
/engines/engine-2-1/directors/director-2-1-A/hardware/ports/A0-FC03] -a enabled -v*
True -t 240>> failed. cause: Command execution failed. cause: Timed out after 240 seconds
An internal VPLEX component related to the FE port is in a disabled state. The root cause of why this state becomes disabled is not yet known.
If this issue is encountered engage Dell EMC VPLEX Customer Service to implement a workaround to enable the FE port. When contacting Dell EMC Customer Service mention this article.
Permanent Fix:
Dell EMC VPLEX Engineering is currently investigating this problem. Once a fix is available this article will be updated.