Mapping pWWNs to physical port

I am about to connect up a bunch of V445s with StorageTek PCI-X 4GB HBAs to a SAN solution. Before I can do that, I need to map my pWWNs to the physical support, so our cabling people can connect the bits of wire to the appropriate holes.
I've got the list of pWWNs, now I need to know how to map them to the physical ports. What I want to end up with is a diagram which has the pWWN and an arrow pointing to a particular port on a particular HBA.
Any ideas how I do this?
TIA
Stuart

Stuart,
Give this a try.
....# prtpicl -v | fgrep -e 'scsi-fcp
....> devfs-path
....> wwn'
You will get more than just HBA info but it will be easier to wade through than the entire prtpicl -v. Either way once you've captured the output, search on <scsi-fcp>. Here is a sample using the above filter.....
SUNW,qlc (scsi-fcp, 2e0500000343)
:device_type scsi-fcp
:node-wwn 20 00 00 e0 8b 12 3e a7
:port-wwn 21 00 00 e0 8b 12 3e a7
:devfs-path /pci@1f,4000/SUNW,qlc@2
:_class scsi-fcp
It should be a simple matter to match up the pci slots after that.

Similar Messages

Logical Column Mapping to Multiple Physical?

I am "learning" OBIEE still and a consultant today showed me something that totally blew my mind
In the OBIEE Admin tool (in verson 10) there are 3 layers displayed. The physical, business management and presentation layer.
I was under the impression that columns would be "mapped" one to one from each layer
The consultant proved me wrong and showed me that in fact, when making a logical column that you can have it "map" to two physical column names
Since every minute with the consultant is $$$ I tried to google around to find out why and how is this so but havent been able to find anything
Could someone explain to me why this would be done?
For example I have the logical column named "Home Faculty" and it is mapped to two physical columns "Faculty Short Desc" & "Main Faculty"
How would OBIEE know which one to use when?
Or does it simply "merge" the data between the two columns into one?

In this case equi join.
When you are using OBIEE try to see wrt fact table with Table 1 and 2 and follow as I suggest earlier.
When you use Fact,Table 1 and 2
Physical Layer Joins would be like:
Fact join with Table 1
Fact join with Table 2
In BMM I can make 3 tables
Like
Fact
dim 1
dim 2
Other case would be
Physical Layer Joins would be like:
Fact join with Table 1
Fact join with Table 2
Table 1's A1,A2 equals Table 2's A1,A2
In BMM I can make 2 tables
Like
Fact
dim sourcing 1 and 2
a) 1 mapping with 2
b) dim with LTS 1 and 2
Hope this helps
If I made it more confuse ;) send me email with US mobile# will call you back
If helps mark

Mapping Files to Physical Devices

Hi
I'm reading 9i Oracle Administrator guide and there is a topic called "Mapping Files to Physical Devices", I'm reading it but I don't understand in what situation it can be useful. Please if anybody can give me a simpler explanation.
Thanks.

OK What you mean is that, for example, I have four
disks on my server and my tablespaces have several
datafiles each. so the Oracle utility allow me to
know exactly the disk where my data is store in order
to analize disk contention.
Am I ok?
Thanks.Almost - I mean that you have 4 disks with some partitions say in solaris speak - if you aren't familiar with the names then d0 is the first disk and p0 the first partition on that disk and so on
/dev/dsk/c1t0d0p0
/dev/dsk/c1t0d0p1
/dev/dsk/c1t0d1p0
/dev/dsk/c1t0d2p0
/dev/dsk/c1t0d2p1
/dev/dsk/c1t0d2p2now your lvm has presented some of those disks to you as two mount points for oracle
/dev/vg01/u01
/dev/vg01/u02if you don't have an ability to map back to the device you won't know what physical devices the datafiles /dev/vg01/u01/oradata/orcl/system01.dbf and /dev/vg/u02/oradata/orcl/undotbs01.dbf are physically located on. It's this capacity that the FMON process activated by the FILE_MAPPING parameter gives you for certain LVMs.
Me, I've mostly used SANs and map my /u01 /u02 filesystems to known mount points presented by my handy sysadmin, and she uses a complicated piece of software called excel to record what goes where!
Niall Litchfield
http://www.orawin.info/

Trunk Vs physical port

Guys can someone plz tell me what is the difference between using subinterface and physical interface for example if i have four DMZ and i take ether1 and then subdivide them in four and trunk it to the swicth (with port channel)....th eother scenario is that i take each physical link and make seprate DMZ (apart from redudnacy) what is the difference and what does cisco recommends or what is teh security say on this guy plz jhelp as cant get my head around

Hi Guroo,
As such there are no specific recommendations, it all depends on your requirements, if you have 4 vacant ports available and do not have any future requirements then you can use individual physical ports, but if you have any issues, then you can easily make 4 sub-interfaces on the ASA physical port, just make sure not to use that physical ports to pass the traffic.
Thanks,
Varun Rao
Security Team,
Cisco TAC

WLC2106 - is there a limit of 1 physical port per SSID/WLAN?

Since SSID/WLAN are assigned to a VLAN/Interface, and VLAN/Interfaces are assigned to a physical port, does that mean, on a WLC2106, a WLAN is limited to 100 mbs, the speed of a single port?
How can I add more bandwidth per WLAN? What's the workaround?
On a WLC4400, it is not an issue since the two 1-GIG ports can be made into an etherchannel.
thanks

Few things...
Given that one AP will never really push more than 18Mbps of real throughput anyway (yes, even if all clients are running at 54Mbps), I'd be surprised if you actually have any contention worth worrying about.
Create multiple interfaces on the WLC, and using Group Mappings ("AP Groups VLAN"), you can then assign each AP to a different group. Presuming you set each group to use a different VLAN for a given SSID, and also presuming that each VLAN is configured for a different physical interface, you then get more than enough bandwidth (ie, 100Mbps per AP).
Other option is to use H-REAP, which is a mode of operation that allows your LWAPP AP to behave like an IOS AP, ie, client traffic is dumped stright on to the network without being forced thruogh the controller. This has obvious knock-ons about what the WLC can / can't do w.r.t. monitoring / controlling client traffic.
Regards,
Richard.

Programmatically identifying com ports by their physical port number

I have a system with a NI PXI 8420/16 card. Once in a while a glitch in my PXI system occurs such that my physical RS-232 ports get re-assigned different com port numbers. Then I have to go into MAX and reassign my aliases to re-align my LabVIEW application with my port hardware. Is there a way to stop these glitches from happenning? OR is there a way to programmatically identify a physical port's com number?
Solved!
Go to Solution.

See if this helps: http://forums.ni.com/ni/board/message?board.id=170&message.id=490257#M490257

SG 300-10 . Mapping fiber to ethernet ports (vlans?)...

Hi,
the incoming fiber on ports 9 and 10 are on different subnets. I need to map the subnet on port 9 to ethernet ports 1-9 and port 10 is on its own. I have the device IPv4 address set to a static address on the same subnet as port 9. I don't know if there's an easier approach, but I attempted to map the ports using vlans (please see the attached screenshots). We don't yet have the fiber link established for port 10, so I haven't had a chance to test, but I wanted to confirm that my configuration is sound. I used the default vlan for ports 1-9, because I need to manage through that subnet. I added vlan 10 for port 10, but I don't know if I have it configured correctly... Please advise. Thanks..!

Hi Greg,
Fiber ethernet is most likely running at 1000Mbits/sec and any Copper Ethernet interfaces on the switch runs at 1000Mbit/sec. personally the only differentiation I make between the copper or fiber are the distances they can run.
I love your screen captures it shows me that ports 1-9 are untagged in VLAN 1 (default VLAN) and port 10 is untagged in VLAN 10.
It tells me that both fiber runs are running untagged Ethernet frames on both fiber runs.
You have associated a IP address onto VLAN1 so you can manage the switch.
but the next question is;
What are you trying to achieve ?
regards Dave

Not enough physical ports - error message??

Hi
I just started to get an error message dialog box when I open Soundtrack Pro on my quad core G5. The message says "the current output device does not have enough physical ports for this document. Some outputs may not play back as a result".
I don't understand this - can anyone help. If I click OK then for sure, not audio comes out of STP.
Running version 1.1 on OSX 10.4.10
This hasn't happened before. It happens either when I open a new STP file with nothing in it, or when sending from FCP.
Thanks
Neil

First of all, thanks for the replies.
I gave it another try today; I deleted the old files and ripped them to my computer again, this time using Windows Media Player instead of iTunes. I tried transferring them to my iPod again and this time it worked. As always iTunes had to convert them to files that it could read, but then it took just a few seconds and all of them made it to my iPod.
I don't know what the problem was the first time around. I checked the file sizes, and the largest was 4.2 MB. There were about 11 songs. It was just a typical audio CD and its total length was 44 minutes.
I guess from now on I'll just import CDs with Windows Media Player and then let iTunes convert them so that I can have even more files on my computer. Oh well. If that's what it takes... Sometime when I have a lot of time to fool around I'll try to see if something is going wrong when I'm importing them with iTunes.
Thanks again for the responses!

Management port in Cisco Switches (are they really physical port)

Hi all,
I have been taught to console into my cisco switch for configurations through console cable + putty (serial terminal).
Then I have been taught to configure a management ip and gateway on the cisco switch.
Switch# conf t
Switch(config)# interface vlan 1
Switch(config-if)# ip address 192.168.1.11 255.255.255.0
Switch(config-if)# no shut
Switch(config-if)# exit
Switch(config)# ip default-gateway 192.168.1.1
All the while, i thought this is the way to remote in to the switch via putty/telnet through the network to configure the switch, until i saw the picture below (cisco catalyst 2960)
=======================================
There is a physical port call ethernet management port. What is it ? What is the difference between this port and the earlier example of setting a management ip in VLAN 1 ?
If i set an IP on this particular interface and I ssh in, will i see the same screen/display/console from the earlier example in which i set a management ip in VLAN1 and I ssh in ?
Regards,
Noob

Hi Leo,
Sorry if you find it hard to explain to me.
I have understood to think of the ethernet management port as a separate entity from the original switch.
Maybe with the help of the diagram below, can you let me know if i have understood correctly ?
*please assume connected port is a management port separated from the normal switch ports
q1) does the ethernet management port need to be connected to another switch ?
I have thought of it as a device on the network and it is mentioned by you previously that it will be connected to a switch
"he traffic goes up the cable connected to the Management port and up a switch. Now that switch holds all the information because it is a switch. "
q2) In the current setup then, terminal B will be able to access the management port - am i right ?
q3) you mentioned that the management port is not able to set any gateway, (which is the router fe0/5 - 192.168.0.3 in my illustration), in that case do you mean that terminal A will not be able to access the management port remotely and it can only be accessible locally ?
Please do correct me if my understanding is wrong.
Thank you so much for your advices.
Regards,
Noob

Port-channel "bouncing" but physical ports are not

Refer to the attached PDF for topology diagram and configuration information.
Here's the issue. According to the logs on DC2-5548-P1 and DC2-5548-P2, port-channels Po3107 and Po3108 are going down numerous times. However, in the same timeframe, the physical interfaces are not reporting any down/up events at all. This issue began just a few days ago. Before that, the systems were operating fine.
NOTE: Po29 is the vPC peer link between P1 and P2.
UPDATE: I totally missed this in the logs somehow but the 5Ks are showing that the physical ports are bouncing. "show interface" output shows the "interface reset" counter incrementing on the physical ports. This problem started in two separate environments 9 minutes apart at approx 01:52 AM on the 16th of March. I'm wondering if there was a power event of some sort which caused physical damage to the hosts. Seems kind of odd that 3 servers started having similar issues within 9 minutes of each other in two 5K environments.

Hi,
Putting the EtherChannel to the "on" mode will force the ports to become bundled unconditionally but the true problem, then, is truly seeing if it works. The Cisco switch will happily keep the ports bundled and will even transmit data over these ports but how do we know if Oracle is happy with that and does the same?
Personally, I would not recommend using the "on" mode precisely because of lack of any indication that things are operating just as they should, unless it can be shown without doubts that Oracle runs without LACP and uses a static EtherChannel.
Best regards,
Peter

Mapping to multiple contact ports...

The last holdout feature that Timbuktu performs that ARD hasn't learned (that I know of) is connecting to multiple clients through a NAT'd firewall. In TB2 you assign a "tcp contact port", say in the 4000 range to each TB2 client machine. Then you setup the forwarding rules on your firewall/router and then from the outside you merely type the IP xxx.xxx.xxx.xxx:4000 or 4001 or 4002..etc to connect to those Macs on the inside of the firewall. Anyone have thoughts on whether this can be hacked into function with ARD? From my current viewpoint, it's a no go, but then I'm always willing to be proven wrong.
Thanks.

You can select what port in the 5900 range the ARD administrator application uses to connect to a client, but you cannot set what ports the client agent uses (and there's no way to "hack" a change). So you can use ARD to observe and control systems behind an NAT router by mapping ports 5900, 5901 etc. (the VNC ports) to the various systems and then connecting as appropriate. You cannot, however, use the other functions behind an NAT router since they only use port 3283 which of course can only be mapped to a single system.
The other option is to have two copies of ARD, one behind the NAT to adminster all the systems on that side of the NAT router, and map the ports in the router to that system. Then you take control of that copy of ARD from the outside copy.
Remember that ARD is not just an observe/control system. It's much more a management system, so it's strengths play to being used for working with local systems. If all you need is observe/control, then Timbuktu, the new PC Anywhere 12, or the freeware VNC may be better choices.
Message was edited by: Dave Sawyer

Assigning a map to an outbound port

Hello together,
I'm trying to syndicate several records using syndication server. I defined my mapping in the MDM Syndicator (5.5 SP5) and saved it. Then I tried to create a port in the console. I took a remote system type inbound/outbound. I defined the port as outbound and expected that I could select my map from the syndicator.
Could you please tell me what I do wrong?
Thanks in advance
René

As far as I understood your problem, you want to syndicate file to the port.
For this -
1) Go to destination properties in Syndicator, select the type, then select your <b>client system</b>.
2) Map the fields and save this map as say MAP1.
3) Now in your Console under Admin node select the port.
Here select your outbound port and the associated client system. In the Map field detail select your map you have created above i.e. MAP1 and save it.
4) Load the repository.
5) Go to Syndicator and open your map i.e MAP1
6) Select Destnation properties.
This time select type as port, corresponding client system and port.
7)Execute the syndication.
Your file will get posted to the respective port.
Hope this helps.
null

Mapping URI with physical location

Hi All
in my web service, i have one method called getBase() which returns a URI
eg: http:://myserver:7001/myapp/data
Using this URI my clients are going to upload some file, my intention is that when they upload that file, the file has to be stored in some location which should be configurable. i mean when they upload some file, it has to go to some location ( can be other machine also). but the URI looks same...where can i do this mapping?
how can this be done?
Plz help!!!
Thanks,

in a database

Problem 2950 arp address not mapped to a destination port

I have a problem loosing connectivity to some switch connected directly to my 2950.
I have the mac address in the arp table but when not working the mac-address is not linked to any port in the show mac-address table.
Please help

I think this could be simply a question of aging. The ARP tables are aged out after 4 hours. By contrast, the entries in the MAC forwarding tables are aged out after 5 minutes. All it means is that the host has not sent any frames to or through this particular switch in the last 5 minutes.
There can be other explanations, but they are rare. For example, if your host is actually a service address in a WLBS cluster configured in unicast mode, then the host will source its frames from a different address from the address it gives in response to an ARP request.
Another is if you have port security configured with a timeout shorter than your ARP table. You may have reached the maximum permitted MAC addresses. That would also account for your intermittent connectivity.
But, as I said, these are very particular cases.
Kevin Dorrell
Luxembourg

Impact of changing VLAN bandwidh if my physical port is 10G?

My monitoring system is reporting that VLAN X is 100% utilization in my 6509. That VLAN is used to pass traffic between two 10G access ports in that VLAN X. It seems that by default the VLAN bandwidth is 1G, however i am passing more traffic than that in that 10G connection. Is it safe to just change the bandwidth to the VLAN without impact? I think it is but would like to know if someone has some experience with this.
Thank you,
Francisco.

Hi,
There should not be any issue by changing it accordingly to the actual speed of the interfaces. Typically the interface bandwidth command would be used to communicate the speed of the interface to Interior Gateway Routing Protocols (IGRP), so they would then calculate their metircs based on that value, the bandwidth value should be accurate, otherwise there would be cases where a routing protocol would rely on wrong values (please see below output) so would consider wrongly a path as equal as another and use them both equally. Please keep in mind also that the bandwidth command does not affect the actual speed of the interface. One more use of the bandwidth command would be in the adjustment of the initial retransmision parameters of tcp.
Switch#sh run int f0/9 | in speed
speed 10
Switch#sh int f0/9 | in BW
MTU 1500 bytes, BW 100000 Kbit, DLY 1000 usec,
In this example, an IGRP would calculate its metric considering 100Mbps as the speed of the interface when the real speed is 10Mbps, that's because the bandwidth is set to 100000 Kb and it is the value that would be used by that IGRP.
Regards,
Aref

Mapping pWWNs to physical port

Similar Messages

Maybe you are looking for