snapmirror multi interfaces not available in PM

pierrek · ‎2009-06-03

Hi,

I received a question from a customer using Protection Manager DR. He used to configure snapmirror manually, using multipath/failover interface specifications in /etc/snapmirror.conf.

The problem is that PM does not seem to allow the use or creation of aggregated network links as snapmirror.conf does.

Is it still a PM limitation, or is there some workaround?

Thanks in advance

Pierre

adaikkap · ‎2009-06-03

Hi Pierre,

No.Its not a limitation of PM.

Use the "hostPreferredAddr1" "hostPreferredAddr2" in the dfm host set command for creating preferred connection

for snapmirror.relatinships.

lnx186-101: # dfm host set
Valid options are
hostLogin                 Login
hostPassword              Password
hostLoginProtocol         Login Protocol
hostPreferredAddr1        Preferred IP address 1
hostPreferredAddr2        Preferred IP address 2
hostPrimaryAddress        Primary IP Address
hostRLMAddress            Remote Platform Management IP Address
hostConsoleAddress        Console Terminal Server Address
cpuTooBusyThreshold       Host CPU Too Busy Threshold (%)
cpuBusyThresholdInterval Host CPU Busy Threshold Interval
hostPingMethod            Ping Method
hostAdminTransport        Administration Transport
hostNdmpLogin             NDMP Login
hostNdmpPassword          NDMP Password
hostAdminPort             Administration Port
hbaportTooBusyThreshold   HBA Port Too Busy Threshold (%)
hostAgentCIFSAccount      Agent CIFS Account
hostAgentCIFSPassword     Agent CIFS Password
perfAdvisorTransport      Performance Advisor Transport
prefsnmpVersion           Preferred SNMP Version
useHostsEquiv             Hosts.equiv Enabled
maxActiveDataTransfers    Maximum Active Data Transfers
perfDataExportEnabled     Performance Advisor Data Export Enabled
perfAdvisorDisabledCounterGroups Perf Advisor Disabled Counter Groups
defaultvFilerInterface    Host vFiler Default Interface
processHostIP             Primary IP Address Inconsistency Action
lnx186-101: #

pierrek · ‎2009-06-03

Thanks a lot, I will test your solution and let you know.

Kind regards

Pierre

reide · ‎2010-10-19

adai,

If you were to set hostPreferredAddr1 & hostPreferredAddr2 for a PM host, would PM define a muliplex or failover SnapMirror relationship? I don't see anywhere to specifty this in the host options.

rshiva · ‎2010-10-19

Hi Reide,

AFAIK, It's a failover based logic. You can check the following link for more details:

https://now.netapp.com/NOW/knowledge/docs/DFM_win/rel40/html/faq/index.shtml#_16.8

Thanks and regards

Shiva Raja

lovik_netapp · ‎2010-10-20

Sorry Shiva, to my experience and knowledge it's multi not failover.

rshiva · ‎2010-10-20

Hi there,

Are you saying that Protection manager would use both interfaces at the same time for a SM transfer?

Thanks and regards

Shiva Raja

lovik_netapp · ‎2010-10-20

Yes.

In our environment we wanted to move all the data over a dedicated vlan so used this feature of DFM. We have set that IP in DFM settings however looking at snapmirror.conf shows that PM used that single IP with multi option to make connection entry.

Don't know why it thinks that with single IP it can create a failover or multiplexing connection but unfortunately that's the reality, hopefully Adai can shed some light on it.

rshiva · ‎2010-10-20

Hi there,

Have you tried specifying both HostpreferredAddr1 and HostpreferredAddr2? If for some reason PM is not able to transfer through HostpreferredAddr1, then it would use HostpreferredAddr2. That’s what I meant in my previous post.

Now coming to your point. Did you import these relationships into PM or were they created through PM?

Thanks and regards

Shiva Raja

adaikkap · ‎2010-10-20

Yes, its a multi, as when hostpreferredAddr1 is set on both the source and destination filer, and when PM creates a VSM, it makes an entry in the snapmirror.conf of the destination filer as follows,

conn_name=multi(IP1,IP2)

and when hostpreferredAddr2 is also set on both source and destination IIRC it is used as failover between hostpreferredAddr1 and hostpreferredAddr2 connection. Though I will have to confirm the same.

Regards

adai

rshiva · ‎2010-10-20

Hi there,

Here's a sample demo:

Primary host: sim733a

Preferred IP address 1: 192.168.34.140
Preferred IP address 2: 192.168.34.141

Secondary host: sim733b

Preferred IP address 1: 192.168.34.141
Preferred IP address 2: 192.168.34.161

Created a relationship through PM:

Portion of job progress messge in PM:

Event Id:      8673
Event Status: normal
Event Type:    job-progress
Job Id:        870
Timestamp:     18 Oct 2010 19:00:16
Message:       sim733a:/mirror: Using connection sim733a_sim733b_1287442815 (192.168.34.140 -> 192.168.34.141,192.168.34.141 -> 192.168.34.161), id = 1
Error Message:

Event Id:                         8675
Event Status:                     normal
Event Type:                       rel-create-progress
Job Id:                           870
Timestamp:                        18 Oct 2010 19:00:17
Message:                          Set connection to sim733a_sim733b_1287442815 (1)
Error Message:
Source Volume or Qtree Id:        1262
Source Volume or Qtree Name:      sim733a:/mirror
Destination Volume or Qtree Id:   1267

Snapmirror.conf entry

#Regenerated by registry Mon Oct 18 23:15:18 GMT 2010

sim733a_sim733b_1287442815=multi(192.168.34.140,192.168.34.141)(192.168.34.141,192.168.34.161)

sim733a_sim733b_1287443717:mirror sim733b:mirror_mirror connection_mode=inet - - - -

sim733a_sim733b_1287443717=multi(192.168.34.140,192.168.34.141)(192.168.34.141,192.168.34.161)

After create relationship failed the Preferred IP address 1 of secondary host

ns0: flags=0x2408866<BROADCAST,RUNNING,MULTICAST> mtu 1500
        inet 192.168.34.141 netmask-or-prefix 0xffffff00 broadcast 192.168.34.255
        inet 192.168.34.150 netmask-or-prefix 0xffffff00 broadcast 192.168.34.255
        ether 00:50:56:05:d4:ee (auto-100tx-fd-cfg_down)
ns1: flags=0x2c48867<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
        inet 192.168.34.161 netmask-or-prefix 0xffffff00 broadcast 192.168.34.255
        ether 00:50:56:06:d4:ee (auto-100tx-fd-up)
lo: flags=0x1948049<UP,LOOPBACK,RUNNING,MULTICAST,TCPCKSUM> mtu 9188
        inet 127.0.0.1 netmask-or-prefix 0xff000000 broadcast 127.0.0.1

Executed the job: portion of job message in PM:

Event Id:      8720
Event Status: normal
Event Type:    job-progress
Job Id:        872
Timestamp:     18 Oct 2010 19:15:17
Message:       Using connection sim733a_sim733b_1287443717 (192.168.34.140 -> 192.168.34.141,192.168.34.141 -> 192.168.34.161), id = 2
Error Message:

Event Id:          8722
Event Status:      normal
Event Type:        snapmirror-progress
Job Id:            872
Timestamp:         18 Oct 2010 19:15:18
Message:           Set connection to sim733a_sim733b_1287443717 (2)
Error Message:
Source Id:         1262
Source Name:       sim733a:/mirror
Destination Id:    1267
Destination Name: sim733b:/mirror_mirror
Bytes Transferred: 0

Job was succesful

So to Adai's point and your point it is multi in the snapmirror.conf, but PM can also perform failover and succesfully complete the job.

Thanks and regards

Shiva Raja

adaikkap · ‎2010-10-20

That’s what I remembered, its failover between the hostpreferredAddr1 and hostpreferredAddr2 connections.

But with just hostpreferredAddr1 connection its multi.

But Shiva I see that IP addr are same.

Primary host: sim733a

Preferred IP address 1: 192.168.34.140
Preferred IP address 2: 192.168.34.141

Secondary host:

Preferred IP address 1: 192.168.34.141
Preferred IP address 2: 192.168.34.161

This is the man pages excerpt from ONTAP.

Each connection definition is a single line giving a name to one or two pairs of IP addresses along with a mode of operation for the connection. The lines are specified in the following format:

name = mode( source_ip_addr1 , dest_ip_addr1 ) ( source_ip_addr2 , dest_ip_addr2 )

name This is the name of the connection you would like to define. This name is to be used as the source filer in relationship definitions.

mode The mode is optional and specifies the mode in which two IP address pairs will be used. Two modes are allowed multiplexing and failover mode and are specified by using the multi and failover keywords. If not specified, multiplexing mode is used.

The multiplexing mode causes snapmirror to use both paths at the same time. If one should fail, it will switch to use the remaining path only and use both again should the failing path be repaired.

Failover mode causes snapmirror to use the first path as the desired path and only use the second path should problems arise with the first path.

So that confirms the behaviour of multi when both are specifed, also failover when one of them is not available.

Regards

adai