Start a Conversation

Unsolved

This post is more than 5 years old

3093

June 24th, 2016 02:00

Replication with AV/DD

Hello,

I have a AV/DD replicating solution with 2 x M1200 and 2 x DD2500 and want to know how to setup replication between the DDs on a dedicated NIC.

On both DD systems I have created a DDBoost Interface Group for the backup traffic which runs on 2 10GbE optical ports (eth0a/eth0b). The management port is on ethMa. The backup from VMware clients, physical clients and NDMP node runs on the 2 10GbE ports. So far so good

.

But I saw that the repliaction between the DD systems is always running on the ethMa port, which has to be changed to run although on the 10GbE ports.

As I setup the repliaction on the Avamar side, I have no idea how to route the replication traffic on these 10GbE ports.

I have setup a host entry on both DDs where the hostname of the partner DD is routed to one of the 10GbE IP-Adresses, but this didn´t help.

Anyone who has setup this already?

1.2K Posts

June 24th, 2016 13:00

Take a look in the DDBoost administration guide.  In the section on IFGROUPs, it indicates the following:

Note

The managed file replication connection between the Data Domain systems is not part of IFGROUP. A single IP address is used for the destination Data Domain system. EMC recommends excluding one interface from the IFGROUP and reserving it for the managed file replication path between the source and destination Data Domain systems.

This might apply to your case.

In my case, we set one DNS hostname on ethMa (DD4200-mgmt) and one DNS hostname on veth1 (DD4200-10gb).  When registering with Avamar, we used the DNS name of the veth1 interface.  Our backups all use the 10Gb interface, as does our replications:

sysadmin@dd4200-prod-mgmt# ddboost file-replication show active


File-replication active:  06/24 16:18:52

Direction  Remote                        Filename                                                                                                              Pre-comp            Pre-comp  Low-bw-optim  Encryption

            Host                                                                                                                                                    Size        Transferred        Status

---------  ---------------------------  ----------------------------------------------------------------------------------------------------------  -----------------  -----------------  ------------  ----------

Outbound    dd4200-dr-10g.local.com  /data/col1/avamar-1400716054/cur/5363491a8888a90a0c3caa7ccc8666855f60bc4d/1D1CE219C5614C2/container.1.cdsf  4,973,797,323,992  3,850,430,887,681      disabled    disabled

Outbound    dd4200-dr-10g.local.com  /data/col1/avamar-1400716054/cur/8babd4378d6777c8d784fd61ae5518c70f25df5f/1D1CE0770D6AC68/container.1.cdsf  5,080,026,546,252  4,551,523,849,757      disabled    disabled

---------  ---------------------------  ----------------------------------------------------------------------------------------------------------  -----------------  -----------------  ------------  ----------

                                  Inbound            Outbound

--------------------------------  -------  ------------------

Number of files                        0                    2

Total pre-comp size (bytes)            0   10,053,823,870,244

Total pre-comp bytes transferred       0    8,401,954,737,438

Total pre-comp bytes remaining         0    1,651,869,132,806

--------------------------------  -------  ------------------

Let us know if that helps!

Karl

1 Rookie

 • 

20.4K Posts

June 26th, 2016 06:00

Karl,

wasn't MFR and ifgroup replication supported in 5.7.x ?

223 Posts

June 27th, 2016 12:00

Hello,

MFR over IFGROUPS is support (ReleaseNotes 5.7.0.10).

Dynamic interface groups are now available for DD Boost managed file replication (MFR).

Dynamic interface groups with MFR provide greater replication performance via link

aggregation and better resiliency through link failover with these features:

l Support for replication over extended LANs or WANs.

l Support for all MFR replication topologies (one-to-one, one-to-many, many-to-one, bidirectional,

and the like).

l NAT translation support between source and target Data Domain systems.

l Support for IPv4 or IPv6, VLANs and alias IPs

l Available to all DD Boost ecosystem partners that support MFR and DD OS 5.7.

So I am not sure what to do.

I have only 2 x optical 10Gbit and don´t want to take one of them out of the IFGROUP.

Is there any other solution for this?

5 Posts

December 15th, 2016 14:00

Hi Druehl,

Just done a replication configuration with DD and avamar today.

On my DD I have a LACP virtual interface with 2 VLAN (mngt and replication)

DD is connect to avamar with Mngt VLAN and I use host mapping on both DD to trick DD to use replication VLAN.

On DD1,  I add @IP VLAN replication2=hostname Mngt 2

On DD2,  I add @IP VLAN replication1=hostname Mngt 1

It works fine in my case.

Fabien

No Events found!

Top