Unsolved

This post is more than 5 years old

1 Rookie

 • 

61 Posts

8123

March 3rd, 2014 17:00

Celerra to VNX File system replication

Hi All,

I am trying to setup File System replication between Celerra(Source) and VNX(Destination) using Celerra Replicator.

Both the boxes are in same site. I have configured the replication interface on both Celerra and VNX in same subnet. However, CIFS and NFS subnet are different from replication subnet.

When I try to create a data mover interconnect on VNX for Celerra I get below error:

1) Check if the Data Movers on the specified remote system are accessible.

2) Ensure that the difference in system times between the local and remote Celerra systems or VNX systems does not exceed 10 minutes. Use NTP on the Control Stations to synchronize the system clocks.

3) Ensure that the passphrase on the local Control Station matches with the passphrase on the remote Control Station.

4) Ensure that the same local users that manage VNX for file systems exist on both the source and the destination Control Station. 5) Ensure that the global account is mapped to the same local account on both local and remote VNX Control Stations. Primus emc263860 provides more details.

When I try to create a data mover interconnect on Celerra for VNX I get below error:

1) Check if the Data Mover on the specified remote system are accessible. \n

2) Ensure that the difference in system times between the local and remote Celerras does not exceed 10 minutes. Use NTP on the Control Stations to synchronize the system clocks. \n

3) Ensure that the passphrase on local Control Station matches with the passphrase on remote Celerras. \n

Both the boxes can ping each other's replication interfaces. So network does look to be an issue here.

Passphrase is same. There is a nasadmin local account on VNX, similarly there is a nasadmin local account on Celerra.

Now since, replication interfaces on both the boxes are in same subnet, I dont think we need to setup a static route.

What am I doing wrong?

11 Legend

 • 

20.4K Posts

 • 

87.4K Points

March 3rd, 2014 19:00

data/time on datamovers and control stations correct ?

6 Operator

 • 

8.6K Posts

March 3rd, 2014 19:00

Did you try both ping and server_ping ?

6 Operator

 • 

8.6K Posts

March 3rd, 2014 19:00

Not enough

Both control stations need to be able to see each other

Both data mover need to be able to see each other

1 Rookie

 • 

61 Posts

March 3rd, 2014 19:00

I tried to ping using data movers from gui.

Regards

Nishant Kohli

Storage - Guitar Center, ITO | Desk: (818) 735-8800 ext 2835

Mobile – 818 584 5956

E: Nishant.Kohli@Guitarcenter.com

1 Rookie

 • 

61 Posts

March 4th, 2014 09:00

Step 1. Tried to ping from source datamover using individual interfaces(cifs, nfs, replication). All Celerra interfaces can ping all VNX interfaces and Vice versa.

Step 2. Thanks for sharing that command. I tried and found time is exactly same on both devices and same NTP server configured.

Step3. Verified that passphrase is exactly same.

So what’s next??

Regards

Nishant Kohli

Storage - Guitar Center, ITO | Desk: (818) 735-8800 ext 2835

Mobile – 818 584 5956

E: Nishant.Kohli@Guitarcenter.com

1 Rookie

 • 

61 Posts

March 4th, 2014 09:00

I am able to ping all IPs on VNX from Celerra datamover or Control station and Vice Versa.

1 Rookie

 • 

61 Posts

March 4th, 2014 09:00

How do you want me to verify. Everything is pingable.

Regards

Nishant Kohli

Storage - Guitar Center, ITO | Desk: (818) 735-8800 ext 2835

Mobile – 818 584 5956

E: Nishant.Kohli@Guitarcenter.com

1 Rookie

 • 

61 Posts

March 4th, 2014 09:00

yes

Regards

Nishant Kohli

Storage - Guitar Center, ITO | Desk: (818) 735-8800 ext 2835

Mobile – 818 584 5956

E: Nishant.Kohli@Guitarcenter.com

1 Rookie

 • 

61 Posts

March 4th, 2014 09:00

Could this happen if standby datamover on VNX is not cabled to the network switch?

We are short of available ports on switches as of now.

We have cabled up the 2*10G datamover ports on primary DM(DM2) to the network, however we are yet to cable the 2*10G datamover ports on standby DM(DM3).

1 Rookie

 • 

61 Posts

March 4th, 2014 16:00

So as per that primus this is what I need to do according to my scenario.

VNX5100/5300/5500/5700/7500 (Integrated systems) replicating with any NAS 6.0 system:

  1. Create a new global user account on the VNX system. 
  2. Log in to the File side of Unisphere and find the name of the local account mapped to that global account. 
  3. Log in to the 6.0 system and create a local account with the same name as the local account on the VNX Integrated system.

Step1: I created a global user named 'reptest'

Step2: I think what they refer as the mapping of local and global account. They should have mentioned mapping of local group to global account. Because I dont see how local account and global account can be mapped in user management.

I logged in to file side found global account 'reptest' was mapped to fullnas and nasadmin.

Step3: I created a local account named reptest and mapped groups fullnas and nasadmin to it.

Then I tried to login with reptest to both VNX and Celerra and tried to create interconnects. Still no go.

I guess I am unable to understand Step2 and Step3.

11 Legend

 • 

20.4K Posts

 • 

87.4K Points

March 5th, 2014 03:00

did you try command line while logged-in as "nasadmin" to Control Station ?

1 Rookie

 • 

61 Posts

March 5th, 2014 11:00

$ nas_cel -interconnect -create SourceInt -source_server server_2 -destination_system xxxxx.com -destination_server server_2 -source_interfaces ip=x.x.x.x -destination_interfaces ip=x.x.x.x

Warning 17716815878: HTTP communication not permitted: Time skew between local and remote system xxxxx.com may exceed 10 minutes OR nas_cel passphrase mismatch exists.

Error 10248: HTTP communication problem: Time skew between local and remote system xxxxx.com may exceed 10 minutes OR nas_cel passphrase mismatch OR network issue.

Verified:

1) No firewall rule implemented on network end.

2) I found time difference of more than 10 mins on Source standby DM and fixed it yesterday. Still no go.

3) Did a nas_cel -update id=”cel_id” on both source and destination.

4) Did set a simple passphrase on both source and destination. No go.

Well before I could send this email it finally worked after doing below steps:

Deleted the existing nas_cel entry from both source and destination and recreated. ☺

Regards

Nishant Kohli

April 16th, 2014 04:00

I had the same problem. I solved logging on celerra and vnx with nasadmin scope local.

Regards

Nadia Pozzoli

6 Operator

 • 

4.5K Posts

April 18th, 2014 13:00

Was your question answered correctly? If so, please remember to mark your question Answered when you get the correct answer and award points to the person providing the answer. This helps others searching for a similar issue.

glen

1 Message

January 27th, 2017 06:00

fujimed Your answer works for me! Thanks a lot!

0 events found

No Events found!

Top