Login or Register to Ask a Question and Join Our Community


AIX

LACP between AIX 7.1 and HP procurve switches

Login to Discuss or Reply to this Discussion in Our Community

 
Thread Tools Search this Thread
Operating Systems AIX LACP between AIX 7.1 and HP procurve switches
# 1  
Old 04-12-2012
LACP between AIX 7.1 and HP procurve switches
Hi.

On a standalone AIX server, 7.1 version, I have 3 connected gigabit ethernet ports.
I made an etherchannel like this :

Code:
# lsattr -El ent4
adapter_names ent1,ent2 | EtherChannel Adapters True
alt_addr 0x000000000000 | Alternate EtherChannel Address True
auto_recovery yes | Enable automatic recovery after failover True
backup_adapter ent3 | Adapter used when whole channel fails True
hash_mode src_dst_port | Determines how outgoing adapter is chosen True
interval long | Determines interval value for IEEE 802.3ad mode True
mode 8023ad | EtherChannel mode of operation True
netaddr 0 | Address to ping True
noloss_failover yes | Enable lossless failover after ping failure True
num_retries 3 | Times to retry ping before failing True
retry_time 1 | Wait time (in seconds) between pings True
use_alt_addr no | Enable Alternate EtherChannel Address True
use_jumbo_frame no | Enable Gigabit Ethernet Jumbo Frames True

On my HP procurve switch, I set a LACP aggregation link between "ent1" and "ent2" ports of the server.

Here are the result of our tests :
  1. Send data from this server to 2 others servers -> seems to have a real aggregation, with a 2Gbps bandwidth.
  2. Send data from 2 others servers th this one -> seems to have only 1 Gbps bandwidth of network activity on the server.
Have you any idea to have 2 Gbps in both ways ? Does LACP support the aggregation in both ways ? Is there any other way to have 2 Gbps bandwidth ?

Tks.
Last edited by zaxxon; 04-22-2012 at 06:26 PM.. Reason: code tags
# 2  
Old 04-21-2012
Ultimately, when a packet has to be transmitted over the link, the sending device (in your example both the switch and the server) has to make a decision to send it over one of the available physical interfaces. This decision is not made by flipping coin, but it is actually calculated based on very strict rules. This decision can be made on the MAC address, the IP-address or the portnumber of either the source or the destination host; or a combination of these.
Usually, by default the decision is made based on a combination of source and destination MAC address. In many switches you can change this load-balancing hash to better suit your needs.

You can imagine if your equipment is doing the load-balancing based on only the destination IP- or MAC-address, you will by definition only use one of the links for all your traffic that is going towards the server. Traffic from the server will usually be loadbalanced, because of the higher number of destinations.

Every now and then, I am involved in a project where the topology is so 'funny' that trunking does not reliably create the neccessary bandwidth, so an upgrade to physical 10Gbps is needed. For example, when you need enough bandwidth to transfer lots of data from a large fileserver to a backupserver every night, as fast as possible. In that case both of the MAC- and IP-addresses are the same, as is are the portnumbers, so the hash will always resolve to the same physical interface within the trunk.


Both LACP and the 'trunk' type support use of the total aggregate bandwitdh in both directions,provided you are using the appropriate hashing in both directions that are suitable for your specific environment.

This explanation is valid for all trunking/aggregation/channeling technologies in all brands that bundle more links into one bigger-bandwitdh logical interface...


For your HP E5400 series switches, download the Network Management and Configuration Guide from the ProCurve website and find the Outbound Traffic Distribution Across Trunked Links section at page 4-28 for more explanatory details and the information on how to change the hashing algorythm in this switch.

As far as I know, this switch also includes the source into the calculations, so if you test this with a more statistically relevant number of hosts, you will probably find that both links are being used.
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. Solaris

Link Aggregation without LACP

Hi, I'm not from the Solaris world and some of these things are new to me. Can someone tell me if it is possible to configure link aggregation without using LACP? I am told etherchannel was setup without LACP. (3 Replies)
Discussion started by: techy1
3 Replies

2. Red Hat

Bonding a Bond with LACP

Does anyone know if it's possible to bond two bonds together? My situation is I have two older Cisco switches that cannot carry a LACP (bond level 4) aggregated between them, but separate aggregates can be setup on the switches themselves. In order to have redundancy of two switches I would... (0 Replies)
Discussion started by: christr
0 Replies

3. Solaris

Delete LACP in Solaris 10 10/09 s10s_u8wos_08a SPARC

My Customer has the requirement to use IPMP instead of LACP cause of giving Switch-Management to others (Outsourcing). I deleted the aggregation with And created an IPMP-Configuration: with all Entries under /etc/hosts /etc/netmasks and /etc/defaultrouter But i end up with one interface... (3 Replies)
Discussion started by: etomer_sk
3 Replies

4. Solaris

LACP Policy - Which is best?

Hello, I have these S10 boxes with LACP using the L4 policy. Does the L4 policy create a lot more overhead as opposed to using the L2? I'm noticing that my traffic does not seem to be very well load balanced accross the NIC's, and I am wondering if the policy I'm using has anything else to do... (1 Reply)
Discussion started by: BG_JrAdmin
1 Replies

5. Shell Programming and Scripting

echo switches

Hello All, I am writing an shell script but abruptly its not able to recognize switches in echo statement. #!/bin/bash top -n 1 -b>ankit host=`hostname` time=`cat ankit|grep load|tr -s " "|cut -d " " -f3` load=`cat ankit|grep load|tr -s " "|cut -d "," -f4|cut -d ":" -f2` ... (3 Replies)
Discussion started by: ajaincv
3 Replies

6. Infrastructure Monitoring

Nagios monitoring for switches.

Hi, I have configured check_snmp plugin and using the plugin i am able to monitor the uptime of the switch. But the following fails # /usr/local/nagios/libexec/check_snmp -H 10.10.10.1 -C ready -o ifOperStatus.1 -r 1 -m RFC1213-MIB SNMP CRITICAL - *down(2)* | # Any suggestions ? (1 Reply)
Discussion started by: uxadmin007
1 Replies

7. IP Networking

LACP aggregation on separates switches

Hello, I'm working on LACP architecture. I would like to know if it's possible to aggregate two links on two separate switches. Here an example I want : Aggregation of link1 and link2 to obtain a logical 2 gbit/s link. Also have redundancy, if one of them is down, the traffic goes through the... (1 Reply)
Discussion started by: jbemonet
1 Replies

8. IP Networking

bonding lacp and link aggregation

Hello, I am trying to get clarity on a few things and am looking for some info. In every article I have read about link aggregation and lacp, it can be used combine physical links to create 1 logical link for increased bandwidth. But what it doesn't say is if this is limited by source/dst. ... (1 Reply)
Discussion started by: jaredo
1 Replies

9. UNIX for Advanced & Expert Users

Link Aggregation and LACP

Hi there I have a requirement to provide failover to our customer boxes in case of interface / switch failure, I have been looking at Solaris Link Aggregation with LACP and I wanted to ask a question Ive seen multiple websites that say the following Does this also mean that if the... (1 Reply)
Discussion started by: hcclnoodles
1 Replies

10. UNIX for Advanced & Expert Users

Tar switches!!!

Hi, If i want to write my data on several tapes, (more than one tape), what switch(s) i need to use with tar. In other word if my data needs the sapce more than one tape & i don't wanna to compress or ... my data. so is it possible to write up to the end of the tape & it asks to put another... (1 Reply)
Discussion started by: nikk
1 Replies
Login or Register to Ask a Question