Best Practices for multi-switch MDS 9124 Impelementations

Hi,
I was wondering if anyone had any links to best-practices guides, or any experience, building mutli-swtich fabrics with the Cisco MDS 9124 or similar (small) switches? I've read most of the FibreChannel books out there and they all seem pretty heavy on theory and FibreChannel protocol operations but lack when it comes to real-world deployment scenarios. Something akin to the Case Studies sections a lot of the CCIE literature has, but anything would be appreciated.
Regards,
Meredith Shaebanyan

Hi Meridith
www.Whitepapers.zdnet.com has links to good reading. It has links to items like:
http://www.vmware.com/pdf/esx_san_cfg_technote.pdf is probably a typical SAN environment these days. It's basic and just put your 9124's in where the switches are.
http://www.sun.com/bigadmin/features/hub_articles/san_fundamentals.pdf is for bigger SANs such as DR, etc.
Things to consider with 9124's are:
They can break so keep a good current backup on a tftp/ftp/scp server.
Consider that if you have all the ports used, the two 8 port licences are not going to work on a replacement switch as they are bound to your hostid. The vendor that sold the switch should be able to get replacements quickly but you will lose time with them.
Know exactly what the snmpserver command does as if you have your 9124 replaced and you load your backup config and you use Fabric Manager, it won't be able to manage the 9124 unless you change the admin password with snmpserver.
9124/9134's don't have enough Buffer Credits to expand beyond about 10 km.
Any ISL's used between switches should always be at least two and use Port Channels where possible.
The 9124 or 9124e or 9134 are great value based switches. I keep a spare for training and emergencies. We use them in a core/edge solution and I am very satisfied with them. I have only had one failure with Cisco switches in the last 5 years and it was a 9140 that sat around for far too long doing nothing. The spare meant we were up and running in 30 minutes from the time we noticed the failure and got to the data centre. As there were two paths, no one actually noticed anything. My management system alerted me.
Remember to make absolutely sure that any servers attached to the SAN have multipathing software. The storage array vendors (HDS, EMC, etc) can sell you the software such as HDLM or Powerpath. You can use an independent solution such as Veritas DMP. Just don't forget to use it.
Follow the guidelines in the two documents and get some training as the MDS training is very good indeed. 5 days training and you will be confident about what to do in any sized SAN including Brocade and McData.
A small SAN is just as satisfying as a large one. If in doubt, get a consultant to tell you what to do.
Is that what you was after? I hope it was not too simple.
Stephen

Similar Messages

Best practice for multi-language content in common areas

I've got a site with some text in header/footer/nav that needs to be translated between an English and Spanish site, which use the same design. My intention was to set up all the text as content to facilitate. However, if I use a standard dialog with the component's path set to a child of the current page node, I would need to re-enter the text on every page. If I use a design dialog, or a standard dialog with the component's path set absolutely, the Engilsh and Spanish sites will share the same text. If I use a standard dialog with the component's path set relatively (eg path="../../jcr:content/myPath"), the pages using the component would all need to be at the same level of the hierarchy.
It appears that the Geometrixx demo doesn't address this situation, and leaves copy in English. Is there a best practice for this scenario?

I'm finding that something to the effect of <cq:include path="<%= strCommonContentPath + "codeEntry" %>" resourceType ...
works fine for most components, but not for parsys, or a component containing a parsys. When I attempt that, I get a JS error that says "design.path is null or not an object". Is there a way around this?

BPC 7M SP6 - best practice for multi server setup

Experts,
We are considering purchasing new hardware for our BPC 7M implementation. My question is what is the recommended or best practice setup for SQL and Analysis Services? Should they be on the same server or each on a dedicated server?
The hardware we're looking at would have 4 dual core processors and 32 GB RAM in a x64 base. Would this adequately support both services?
Our primary application cube is just under 2GB and appset database is about 12 GB. We have over 1400 users and a concurrency count of 250 users. We'll have 5 app/web servers to handle this concurrency.
Please let me know if I am missing information to be able to answer this question.
Thank you,
Hitesh

I don't think there's really a preference on that point. As long as it's 64bit, the servers scale well (CPU, RAM), so SQL and SSAS can be on the same server. But it is important to look also beyond CPU and RAM and make sure there's no other bottlenecks like storage (Best practice is to split the database files on several disks and of course to have the logs on disks that are used only for the logs). Also the memory allocation in SQL and OLAP should be adjusted so that each has enough memory at all times.
Another point to consider is high availability. Clustering is quite common on that tier. And you could consider having the active node for SQL on one server and the active node for OLAP (SSAS) on the other server. It costs more in SQL licensing but you get to fully utilize both servers, at the cost of degraded performance in the event of a failover.
Bruno
Edited by: Bruno Ranchy on Jul 3, 2010 9:13 AM

Is there a best practice for multi location server setups including mac mail server?

Afternoon all,
Last year I setup a client with Snow Leopard Server including hosting his mail on the server with Mac Mail Server and calendaring. He now has plans to open other sites with the same setup, how can this be done using Mac Server? The implementation of a new server on the new site would be straight forward my concerns / question are:
1. How would the two servers communicate?
    a.)Do they need to communicate?
3. How will mail across the two sites work?
    a.) How will DNS work for email internally?
    b.) How will DNS work for emai externally?
4. How will calendaring work across the two sites?
Is Mac Server the best platform for moving ahead with this type of business need?
Any help or direction would be greatly appreciated.
Anthony

Camelot,
many thanks for the speedy reply. Your comments are very helpful thank you, if I may I will give you some for information and some other questions.
The offices will be from 5 miles to 25 miles apart, the new office and ones that follow will be considered as branches of the main office for example the company name is Sunflower and it serves area 1, then new will office will serve area 2 and so on. So in theory I could leave the main server domain and mail mx record as sunflower and then add further servers as 2.sunflower.com, 3.sunflower.com as domains and mx records? This would then provide unique mail records and users within the organisation such as [email protected] and [email protected], which doesnt look to good I would prefer all users to be name@sunflower how can this be acheived?
With regard to user activity in the whole users will be fixed but their will be floaters such as managers and the owners that may at times float from one office to the other and would beneift from logging into any machine and getting their profile.
I have thought about VPN's as I have acheieved this type of setup with Microsoft server products, I have found speed issues in some cases, how can this be acheived using OS X, are there any how to docs around? In the Microsoft setup I acheived this using netgear VPN Firewalls which ofcourse adds an overhead, can this be acheived using OS X propietary VPN software?
So ultimatley I would prefer to have the one domain without subs "sunflower.com" and users to be able to login to their profiles from either location. The main location now will remain as headoffice and the primary location and the new will be satelites.
I hope that covers all of your other question, again many thanks for your comments and time.
Kind Regards
Anthony

What is best practice for multi camera edit with differing lengths?

I have three cameras that took video of an engagement party. Camera A and B took it all (with some early extra material exclusive to each camera). Camera C took some, then stopped, then took more.
So I have four sets of clips - A and B which are complete, then C then D.
Should I create sequence 1 with A, B and C synchonised, then create sequence 2 with A, B and D synchronised, then sequence 3 with sundry early non-multi camera clips, plus sequence 1 plus sequence 2 then late non-multi camera clips?
Or can I synchronise A, B and C, then on the same timeline synchronise A, B and D? I'm concerned that the second synchronisation will put C out of sync.
What is the best way to approach this?
Thanks in advance.

A and B which are complete, then C then D.
I think you're looking at this in the wrong way. You have only three cameras, A, B and C, but you don't really have a D camera, as those are just other clips from camera C. You might call them C1 and C2 if you like, but calling them D seems to be confusing the issue, as it's still only three cameras, and three shots to choose from when cutting the sequence. (Except for the gap between C clips, where you will have only the A and B shots to choose.)
You can absolutely sync all the clips form camera C on the same sequence as A and B. And it will probably be easier to do so.

Best practice for multi-page site

Hello!
I am working on a website build entirely in Flex 3.
What is the best way to implement multiple pages as I don't want all loding at the same time?
Should I make a 'parent' mxml, say main.mxml (swf later on) which loads the other pages in a VBox where they should show?
Is there a better way?
Thank you!

Split your app into components, use a viewstack for this, not states, and consider modules if there are many views most users never see.
Here are some resources to help you:
http://livedocs.adobe.com/flex/3/html/help.html?content=navigators_3.html
http://livedocs.adobe.com/flex/3/html/help.html?content=modular_2.html
The ChiaraDev tutorials and sample apps are also useful:
http://www.chikaradev.com/tutorials/index.html
http://www.chikaradev.com/learning/flex3/dragdrop/basic_dragdrop/index.html
http://www.chikaradev.com/learning/flex3/general/SampleApps1/index.html
If this post answers your question or helps, please mark it as such.
Greg Lafrance
www.ChikaraDev.com
Flex Development and Support Services

Best practice for unmanaged switch to cisco switch

In our environment, I have to allow some users to have a unmanaged switch which is connected to access port.
I put this configuration for each port which is connected to unmanaged switch (Netgear 8 port)
interface GigabitEthernet1/0/47
switchport port-security maximum 3
spanning-tree guard root
end
port-security maximum 3: only allow 3 mac
spanning-tree guard root: just in case to protect root bridge if someone put managed switch with lower bridge ID.
I connected one cable from unmanaged switch to another port to make a loop for test.
It showed that switch got "Loop-back detected" and put err-disable port automatically. So I don' t need to worry about this.
Apr 7 18:33:01.370: %ETHCNTR-3-LOOP_BACK_DETECTED: Loop-back detected on GigabitEthernet1/0/47.
Apr 7 18:33:01.370: %PM-4-ERR_DISABLE: loopback error detected on Gi1/0/47, putting Gi1/0/47 in err-disable state
Apr 7 18:33:02.373: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/47, changed state to down
Apr 7 18:33:03.379: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/47, changed state to down
LAB_HQ_Fiber(config-if)#
What are the option do you use usually to protect from unmanaged switch?
I am not able to use "spanning-tree bpduguard" because it will block a port
I can use "spanning-tree bpdufilter" to protect a STP area, but I don't think this is a big matter.

Hello
If you need to attached these kind of switches/hubs - Make sure you shutdown all unused ports on the managed switch so to limit any further unauthorized attachments looping back into the network from the unmanageable device, This way you can managed these unmanageable devices to a certain extent.
As for the stp root, you should manually set your stp priority on the managed switch to a low level anyway so as to not allow any other new device negotiate its self to become the root, and for the ports you are aware of that will have these devices attached, i would disable portfast and also advise against using bpdufilter as this negates the stp process.
Int range fa0/x -xxx
description unmanaged devices
no cdp enable
On all managed switches and on all ports you DONT expect to have unmanaged hubs/switches I would suggest to apply
spanning-tree loopguard default
udld enable
udld aggressive
int range fa0/x -xxx
description access ports
switchport port-security
switchport port-security aging type inactivity
switchport port-security violation restrict/shutdown
switchport port-security maximum 2
spanning-tree portfast
spanning-tree bpduguard enable
spanning-tree guard root
no cdp enable
One last thing I also wouldn't enable error recover either, as you would want to know the reason why your ports are erroring and not go chasing your tail as the reason why your having intermittent network issues.
res
Paul

Best Practice for multi lengual site

I have a site: www.powerhouse.com that we want to create a spanish version for. I decided to basically duplicate the site at url: www.powerhouse.com/espanol. The espanol site links to all the original css. I am running in to spacing issues given that spanish has more characters than english. For example: The main drop down menu bar across the top looks fine in english yet in spanish, 'About Us' ends up being 2 lines, screwing up my spacing on the bar. I am assuming that I should duplicate the css file and adjust it accordingly? I just want to make sure I am on the right track. It basically looks like I am going to have 2 sites to maintain.

Split your app into components, use a viewstack for this, not states, and consider modules if there are many views most users never see.
Here are some resources to help you:
http://livedocs.adobe.com/flex/3/html/help.html?content=navigators_3.html
http://livedocs.adobe.com/flex/3/html/help.html?content=modular_2.html
The ChiaraDev tutorials and sample apps are also useful:
http://www.chikaradev.com/tutorials/index.html
http://www.chikaradev.com/learning/flex3/dragdrop/basic_dragdrop/index.html
http://www.chikaradev.com/learning/flex3/general/SampleApps1/index.html
If this post answers your question or helps, please mark it as such.
Greg Lafrance
www.ChikaraDev.com
Flex Development and Support Services

Best practice for optimizing processing of big XMLs?

All,
What is the best practice when dealing with large XML files ... (say couple of MBs).
Instead of having to read the file from the file system everytime a static method is run, what would be the best way in which the program reads the file once and then keeps it in memory. So the next time it would not have to read and parse it all over again?
Currently my code just read the file in the static method like ...
public static String doOperation(String path,...) throws Exception
try{
String masterFile = path+"configfile.xml";
Document theFile = (Document)getDocument(masterFile);
Element root = theFile.getDocumentElement();
NodeList nl = root.getChildNodes();
// ... operations on file
Optimization tips and tricks most appreciated :-)
Thanks,
David

The best practice for multi-megabyte XML files is not to have them at all.
However if you must, presumably you don't need all of the information in your XML, repeatedly. Or do you? If you need a little bit of it here, then another little bit of it there, then yet another little bit of it later, then you shouldn't have stored your data in one big XML.
Sorry if that sounds unhelpful, but I'm having trouble imagining a scenario when you need all the data in an XML document repeatedly. Perhaps you could expand on your design?
PC²

Best Practices for MDS Monitoring

We are deploying Solarwinds and would like some tips on what would be useful to monitor on MDS switches. Of course, the normal port utilization, cpu, interface counters are already set up, but thought I would check to see if anyone has any suggestions on any other OID's I should look into.
Thanks!

Hi Mike,
Did you try the following link ?
If not, it may be useful to some extent:
http://service.sap.com/bestpractices
---> Cross-Industry Packages ---> Best Practices for Solution Management
You have quite a few documents there - those on BPM may also cover Solution Monitoring aspects.
Best regards,
Srini
Edited by: Srinivasan Radhakrishnan on Jul 7, 2008 7:02 PM

What is best practice for remotely managing bank of switches over POTS

I need to be able to have a back door into several catalyst switches and ASA.
What is the best practice for accessing them remotely. ?

Just place a modem into any console port. Ideally you use a terminal server, but is not always really needed.

Best Practices for CS6 - Multi-instance (setup, deployment and LBQ)

Hi everyone,
We recently upgraded from CS5.5 to CS6 and migrated to a multi-instance server from a single-instance. Our current applications are .NET-based (C#, MVC) and are using SOAP to connect to the InDesign server. All in all it is working quite well.
Now that we have CS6 (multi-instance) we are looking at migrating our applications to use the LBQ features to help balance the workload on the INDS server(s). Where can I find some best practices for code deployment/configuration, etc for a .NET-based platform to talk to InDesign?
We will be using the LBQ to help with load management for sure.
Thanks for any thoughts and direction you can point me to.
~Allen

Please see if below metalink note guides you:-
Symmetrical Network Acceleration with Oracle E-Business Suite Release 12 [ID 967992.1]
Thanks,
JD

Networking "best practice" for setting up a farm

Hi all.
We would like to set an OracleVM farm, and I have a question about "best practice" for
configuring the network. Some background:
- The hardware I have is comprised of machines with 4 gig-eth NICs each.
- The storage will be coming primarily from a backend NAS appliance (Netapp, FWIW).
- We have already allocated a separate VLAN for management.
- We would like to have HA capable VMs using OCFS2 (on top of NFS.)
I'm trying to decide between 2 possible configurations. The first would keep physical separation
between the mgt/storage networks and the DomU networks. The second would just trunk
everything together across all 4 NICs, something like:
Config 1:
- eth0 - management/cluster-interconnect
- eth1 - storage
- eth2/eth3 => bond0 - 8021q trunked, bonded interfaces for DomUs
Config 2:
- eth0/1/2/3 => bond0
Do people have experience or recommendation about the best configuration?
I'm attracted to the first option (perhaps naively) because CI/storage would benefit
from dedicated bandwidth and this configuration might also be more secure.
Regards,
Robert.

user1070509 wrote:
Option #4 (802.3ad) looks promising, but I don't know if this can be made to work across
separate switches.It can, if your switches support cross-switch trunking. Essentially, 802.3ad (also known as LACP or EtherChannel on Cisco devices) requires your switch to be properly configured to allow trunking across the interfaces used for the bond. I know that the high-end Cisco and Juniper switches do support LACP across multiple switches. In the Cisco world, this is called MEC (Multichassis EtherChannel).
If you're using low-end commodity-grade gear, you'll probably need to use active/passive bonds if you want to span switches. Alternatively, you could use one of the balance algorithms for some bandwitch increase. You'd have to run your own testing to determine which algorithm is best suited for your workload.
The Linux Foundation's Net:Bonding article has some great information on bonding in general, particularly on the various bonding methods for high availability:
http://www.linuxfoundation.org/en/Net:Bonding

What are the best practices for using the enhancement framework?

Hello enhancement framework experts,
Recently, my company upgraded to SAP NW 7.1 EhP6. This presents us with the capability to use the enhancement framework.
A couple of senior programmers were asked to deliver a guideline for use of the framework. They published the following statement:
"SAP does not guarantee the validity of the enhancement points in future releases/versions. As a result, any implemented enhancement points may require significant work during upgrades. So, enhancement points should essentially be used as an alternative to core modifications, which is a rare scenario.".
I am looking for confirmation or contradiction to the statement "SAP does not guarantee the validity of enhancement points in future releases/versions..." . Is this a true statement for both implicit and explicit enhancement points?
Is the impact of activated explicit and implicit enhancements much greater to an SAP upgrade than BAdi's and user exits?
Is there any SAP published guidelines/best practices for use of the enhancement framework?
Thank you,
Kimberly
Edited by: Kimberly Carmack on Aug 11, 2011 5:31 PM

Found an article that answers this question quite well:
[How to Get the Most From the Enhancement and Switch Framework as a Customer or Partner - Tips from the Experts|http://www.sdn.sap.com/irj/scn/index?rid=/library/uuid/c0f0373e-a915-2e10-6e88-d4de0c725ab3]
Thank you Thomas Weiss!

Best Practice for FlexConnect Wireless roaming in MediaNet environment?

Hello!
Current Cisco best practice recommendations for enterprise MediaNet design, specify that VLANs be local to a switch / switch stack (i.e., to limit the scope of spanning-tree).
In the wireless world, this causes problems if you want users while roaming to keep real-time applications up and running. Every time they connect to a new AP on a different VLAN, then they will need to get a new IP address, which interrupts real-time apps.
So...best practice for LAN users causes real problems for wireless users.
I thought I'd post here in case there's a best practice for implementing wireless roaming in a routed environment that we might have missed so far!
We have a failover pair of FlexConnect 7510s, btw, configured for local switching for Internal users, and central switching with an anchor controller on the DMZ for Guest users.
Thanks,
Deb

Thanks for your replies, Stephen and JSnyder.
The situation here is that the original design engineer is no longer here, and the original design was not MediaNet-friendly, in that it had a very few /20 subnets bridged over entire large sites.
These several large sites (with a few hundred wireless users per site), are connected to an HQ location (where the 7510s in failover mode are installed) via 1G ethernet hand-offs (MPLS at the WAN provider). The 7510s are new, and are replacing older contollers at the HQ location.
The internal employee wireless users use resources both local to their site, as well as centralized resources. There are at least as many Guest wireless users per site as there are internal employee users, and the service to them consists of Internet traffic only. (When moved to the 7510s, their traffic will continue to be centrally switched and carried to an anchor controller in the DMZ.)
(1) So, going local mode seems impractical due to the sheer number of users whose traffic bound for their local site would be traversing the WAN twice. Too much bandwidth would be used. So, that implies the need to use Flex / HREAP mode instead.
(2) However, re-designing each site's IP environment for MediaNet would suggest to go routed to the closet. However, this breaks seamless roaming for users....
So, this conundrum is why I thought I'd post here, and see if there was some other cool / nifty solution I wasn't yet aware of.
The only other (possibly friendly to both needs) solution I'd thought of was to GRE tunnel a subnet from each closet to the collapsed Core / Disti switch at each site. Unfortunately, GRE tunnels are not supported in the rev of IOS on the present equipment, and so it isn't possible to try this idea.
Another "blue sky" idea I had (not for this customer, but possibly elsewhere in the future), is to use LAN switches such as 3850s that have WLC functionality built-in. I haven't yet worked with the WLC s/w available on those, but I was thinking it looks like they could be put into a mobility group, and L3 user roaming between them might then work. Do you happen to know if this might be a workable solution to the overall big-picture problem?
Thanks again for taking the time and trouble to reply!
Deb

Best Practices for multi-switch MDS 9124 Impelementations

Similar Messages

Maybe you are looking for