Sponsored Content
Special Forums IP Networking Quick Poll: Solaris IP Multipathing Post 44400 by RTM on Friday 5th of December 2003 11:16:28 AM
Old 12-05-2003
Quote:
Whenever a single point of failure can be removed, it should be.
If it is cost effective. In the case of interface cards it is but then add in the cost of using two ports for each network your one system is on...more switches are needed, folks complain about running out of IP addresses...some battles are not worth it. If you had a SUN E10K, you would have to buy a second because of the single point of failure that is internal to it. Quite costly.
 

7 More Discussions You Might Find Interesting

1. Solaris

Quick question about Solaris =)

:D Just curious, I know that solaris 2.0-2.x_present are System V UNIX... But, I have only used Solaris 10, 9 and 2.5... On a 2.5 Sparc box, after you login you get presented with the usual SunOS 5.5.1 then UNIX(R) System V Release 4.0 I know Solaris 9 and 10 do not include this banner... ... (0 Replies)
Discussion started by: ssorel
0 Replies

2. Solaris

solaris multipathing

I have solaris 10 sparc. I installed a Qlogic hba card. This card is connected on a brocade switch and the brocade is connected on 2 different controllers on a Hitachi disk bay. I formated 2 luns. On my solaris system, i have 4 disk. How to configure solaris 10 to fix the dual disk view. ... (4 Replies)
Discussion started by: simquest
4 Replies

3. Solaris

Solaris IP Multipathing (IPMP) Help

Hello All, I work for a Health care company at a local trauma hospital. I maintain a Picture Archiving and Communication System (PAC's). Basically, any medical images (X-Ray, CT, MRI, Mammo, etc) are stored digitally on the servers for viewing and dictation from diagnostic stations. I took over... (10 Replies)
Discussion started by: mainegeek
10 Replies

4. Solaris

Solaris IP Multipathing

Hi, I saw your post on the forums about how to setup IP multipathing. I wanted your help on the below situation . I have 2 servers A and B . Now they should be connected to 2 network switches . S1 and S2. Now is it possible to have IP Multipathing on each of the servers as follows ? ... (0 Replies)
Discussion started by: maadhuu
0 Replies

5. Solaris

Disable Inbuilt multipathing software of Solaris 10

Dear UNIX Gurus I need to know how to disable MPXIO feature (Inbuilt multipathing software of Solaris 10) as we are going to implemented iSCSI EMC based SAN. I am bit new to this feature so kindly suggest how to disable it Regards (4 Replies)
Discussion started by: girish.batra
4 Replies

6. Solaris

Solaris multipathing

Hai we using emc storage which is conneted to M5000 through san switch. we asign 13 luns but in server it is showing 22 luns. i enable the solaris multipathing (MPxIO) #more /kernel/drv/fp.conf in that file MPxio-disable=no #mpathadm list lu it shows ... (2 Replies)
Discussion started by: joshmani
2 Replies

7. Solaris

ZeusIOPS multipathing failed Solaris 11

hi, just installed solaris 11.1 (11.1.9.5.1) on a sun7410 ZFSSA (x4440 + j4400), but when I enable multipathing with "stmsboot -e", only the 1TB SATA drives inside the j4400 are configured as multipathed vhci / mpxio devices, but the logzillas / ZeusIOPS drives are skipped. the server is... (0 Replies)
Discussion started by: pressy
0 Replies
lanqueryconf(1M)														  lanqueryconf(1M)

NAME
lanqueryconf - query LAN interface configuration information for LAN_MONITOR mode link aggregates, also called failover groups SYNOPSIS
ascii_file] ascii_file] DESCRIPTION
The command discovers the set of LAN interfaces that can form a failover group and then maps that information to the ASCII configuration file. Options supports the following options: Find existing LAN Monitor failover group configuration from the kernel, and write it to the ASCII configuration file. Specifies the name of the ASCII configuration file into which configuration information is written. See the "ASCII Configuration File" section for information about the file parameters and syn- tax. By default, information is written to the file. Queries the system for the potential failover group configuration, and write it to the ASCII configuration file. Displays verbose output. ASCII Configuration File The default ASCII configuration file is It is similar to the Serviceguard file. This file contains the fields below. Note: the term is used for a physical link or a link aggregate with or mode. o The aggregate name for a LAN Monitor failover group, where XX is a decimal number, starting with 00 (lan900). may generate failover group name where X is the PPA of the physical link. Such failover group will be ignored unless it is modified manually. Each can have the following keywords: o This field can be used to specify the preconfigured IP address on the primary port. o Each failover group must have one primary port. The primary port is usually assigned the highest priority or lowest cost value, so it will be the port that carries traffic initially. The primary port is also the port that may have IP addresses configured. After lanapplyconf creates the failover group, the IP addresses on the primary port (if any is present) are moved to the failover group. Prior versions of APA maintains that the primary port must have IP address configured and must be the highest priority/lowest cost. APA 11i v3 observes neither of these rules: the primary port may be assigned any priority/cost values and it may not have IP addresses configured. Port that carries traffic depends on the priority/cost. HP recommends that the primary port always be assigned the highest priority or the lowest cost in a failover group. o Specifies a standby port for the A standby port is a port that replaces the primary port when the primary is incapable of carrying traffic. Both the and keywords have the following fields: o The LAN interface name (for example, lan0 and lan1). o The port priority that will be assigned to the port. o A cost value for the LAN interface preceded by a colon (:) and a space. This is an optional field. A port with the lowest normalized cost (cost divided by port's speed) will carry traffic. When ports have the same normalized cost, the one with the higher priority is preferred. For each failover group, either all or none of the ports should have the cost values assigned. If none of the ports has cost value assigned, then the failover group uses the default failover behavior based on priority. See the "Proactive Failover" section and the for more information. o The number of microseconds between polling messages. Polling messages are sent between ports in the specified interval for monitoring the health of all the ports in the failover group. The default is 10000000 (10 seconds). This keyword can be specified multiple times in the configuration file. A failover group's polling interval is set to the last specified interval. o The number of polling packets that are missed before deciding to send a log message to the user that indicates the port might have prob- lems and the network should be checked. The default is 3. o Enables (on) or disables (off) the ability to transmit gratuitous ARP messages at intervals shorter than 5 seconds. By default, is on. When the MAC address of a failover group changes, a gratuitous ARP packet is transmitted. To ensure that other clients and servers receive the new IP-MAC address mapping, APA/LAN Monitor transmits more gratuitous ARP packets for 1 minute. When this option is enabled, LAN Monitor will transmit gratuitous ARP at every for a total of times immediately after the MAC address changes. After completing the process, gratuitous ARP packets will be transmitted every 5 seconds until one minute has elapsed since the first gratuitous ARP was transmitted. When this option is disabled, LAN Monitor will transmit gratuitous ARP at every 5 seconds for 1 minute. o Number of microseconds between rapid gratuitous ARP messages. The range of valid values is 1000000-4000000, inclusive (1 second to 4 seconds). The default value is 1000000 (1 second). Specify a whole number of seconds as the value you specify is automatically rounded up to the next whole number of seconds. The value must be a valid integer. If the value is not in the valid range and is on, the interval value last processed with the command is used. If is off, this value is ignored for the corresponding failover group. o Number of gratuitous ARP packets sent rapidly. The valid range is 5-60, inclusive. The default value is 10. The value must be a valid integer. If the value is not in the valid range and is on, the count value last processed with the command is used. If is off, this value is ignored for the corresponding failover group. Proactive Failover By default, the port in a failover group with the highest priority is the active port. With proactive failover, the port that is the most efficient at carrying traffic is the active port. Efficiency is determined by assigning a cost to each port in a failover group. This cost is divided by the port's current speed to yield a normalized port cost; port speed is the number of links in a link aggregate multi- plied by the link speed, or in the case of a physical port, only the link speed. The lower the normalized port cost, the higher the port's efficiency. During certain LAN Monitor events, the normalized port cost might change on the active or standby ports. When these events occur, the nor- malized port cost of the active port and the standby ports are compared. If a standby port has a lower normalized port cost than the active port, the standby port becomes the active port. Sample File The file specifies failover groups and with total number of 2, 4, 2, 2, and 1 ports in each, respectively. The first network interface listed is the primary and the subsequent ones are the standby interfaces. Failover groups and have polling interval of 5 seconds and dead count of 2; and have polling interval of 10 seconds and dead count of 3. is enabled (on). There are IP addresses preconfigured on the primary ports of lan900 and lan905. There are no IP address preconfigured on the primary ports of lan901 and lan902. is a cost-based failover group. The cost values are preceded by a colon and a space. The primary port has the lowest cost. is also a cost-based failover group. The primary port and the standby port have the same cost, but the primary port has higher priority. When the failover group is created, the normalized cost determines which port will be the active port; it might not be the one designated as in the ASCII configuration file. When the failover group is cleared, the IP address switches back to the port specified as in the ASCII file if the primary port preconfigured IP address when creating the failover group. is a single-port failover group, no aggregate is allocated for it. POLLING_INTERVAL 10000000 DEAD_COUNT 3 LM_RAPID_ARP on LM_RAPID_ARP_INTERVAL 1000000 LM_RAPID_ARP_COUNT 10 FAILOVER_GROUP lan1 STATIONARY_IP 193.12.14.54 FAILOVER_GROUP lan900 STATIONARY_IP 194.12.14.55 PRIMARY lan2 5 STANDBY lan3 3 FAILOVER_GROUP lan901 PRIMARY lan4 5 STANDBY lan5 4 STANDBY lan7 3 STANDBY lan6 3 POLLING_INTERVAL 5000000 DEAD_COUNT 2 FAILOVER_GROUP lan902 PRIMARY lan903 5 : 2 STANDBY lan904 3 : 3 FAILOVER_GROUP lan905 STATIONARY_IP 192.12.14.56 PRIMARY lan907 9 : 3 STANDBY lan908 7 : 3 AUTHOR
was developed by HP. SEE ALSO
lanapplyconf(1M), lancheckconf(1M), landeleteconf(1M), lanscan(1M), netstat(1M). User manuals at lanqueryconf(1M)
All times are GMT -4. The time now is 07:39 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy