Best QOS Practice for a congested Uplink Port?

I have an MPLS uplink port to a carrier that carries both voice and data.
For example, Customer 3750 Switch 100MB Uplink--->100MB Uplink Carrier Router---Carrier Router 6MB MLPP Voice/Data MPLS Uplink with QOS configured for Voice subnet on carrier side too.
The port occasionally suffers from overutilizition and spikes to the full 6MB
I have a centralized CUCM that has phones that occasionally reset due to TCP 2000 timeouts (usually during the period of high utilization)
So that I can avoid most phone resets during high utilization I have prioritized all voice traffic (signalling and RTP streams) to EF
My question: What is best practice configuration for a congested uplink port? I'm going to assume the answer is it depends (the all great technical answer )
Here are my thoughts on how to configure the 3750 uplink port so far:
apply mls qos trust dscp
apply priority-queue out
(Here's where I'm looking for help)
apply some sort of policing or bandwidth statement on the interface to protect the voice traffic: What are the recommendations and what would those configurations look like?
I would apply these these configurations to the uplink ports at the edge site as well as the central site.
Any thoughts as to the best way to accomplish this?

Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
You want to prioritize VoIP traffic at congestion points, which for you, looks to be whenever your LAN bandwidth hits your 6 Mbps WAN.
If the carrier takes your 100 Mbps, and doesn't just generally police all traffic to 6 Mbps, but instead has different priority queues for the 6 Mbps, all you should need to do, is insure you traffic is correctly processed by your WAN vendor. This, though, might require marking your traffic for the WAN provider.
If you carrier first polices all your traffic at 6 Mbps (many do), then you need to shape the traffic (with you own prioritization) before the provider "sees" it. If you need to "shape" on the 3750, there's a command to limit a port's bandwidth utilization, as a percentage. However it's not exact, so you may need to "shape" slower than 6 Mbps to insure 6 Mbps won't be exceeded. (BTW, for 6 Mbps, if you need to do QoS, you would be better off with an ISR.)
You also mention 6 MLPPP, but it's unlclear what your device is for that.
How to configure 3750 QoS is involved. Basically when you enable QoS, each port has four dedicated egress queues. By default, different traffic markings go into one of the four queues, each queue has same share of the bandwidth, and almost same share of the buffers. The PQ command you noted, enables the first queue to always transmit its packets first. Normally, you'll want to do that for VoIP bearer packets, which you've (insured are) directed to that queue. You also want to insure that VoIP signally packets are not likley to be dropped and, more or less, are forwarded quickly.
Cisco has multiple papers on QoS configurations, including papers for 3750s, so instead of my trying to present that, the above is intended as an overview. Feel free to post additional questions, the more specific, the more likely you'll get an answer.

Similar Messages

Best/Easiest Practice for Distribution of Webhelp files for Fat(ish) Client Application

So another in my long line of questions in trying to change the implementation of the help systems at my new employer.
So, the pressure is to convince the existing guard to toss out the old method and go with a new method... One of the issues has come up for CSH Webhelp (via RH10)...
Apparently, in the old method (where every help topic was it's own CHM - no, I'm not kidding). They kept it that way so that any time a help file was updated, they could just send out that one chm to update that one topic.
Now they are worried that if we switch to webhelp using topics in one single big project that we can't just send out a single updated file any more...
So I'm trying to get ammunition as to how updates to the help can be distributed without it being a big hassle...
We do have small updates that go out about every week, and they are afraid that going to the webhelp (1 project with lots of topics) method will require more time and effort for the techs because they will have to complete the update of a huge help system every time.
One solution we have suggested is that we only update help files on major releases and add any changes to help procedures in the Release Notes or in a separate PDF and also include a note that the help files "will be updated to reflect this change in the XX/2015 XYZ Release).
Does ANYONE have any other ideas of how distribution could be done efficiently so that we don't have to continue this "project per topic" fiasco?
HELP!!! PLEASE!!!!

Amebr - you hit the nail on the head... i want to apologize to everyone for all this mass hysteria...but this has been a freaking rollercoaster... one day they are happy with the plan, the next day they decided they want to complicate it more... So now my job is to find out how viable and how tricky and perhaps how dangerous it will be to send out JUST the files that have been changed. Especially for CSH!
Anyone have any experience with this or have any advice? Peter? I suspect you might know how this might work... are they any pitfalls to watch out for.. I just have this fear of sending out a few files from an entire project... Altho, it appears when I publish, ONLY the files that have been edited in some way show a new date.. that includes non-topic files - so i'm assuming sending all those will keep the TOC, Index, and Search features working properly... as well as incoming and outgoing links from an edited topic?
Another question would be how to handle new images in a project... as the image folder tends to be the largest, i don't want to have to send the entire image folder... Thoughts...???? Advice? Jeff? Peter? Amebr? Bueller? Bueller? Bueller....?
THANKS!Best/Easiest Practice for Distribution of Webhelp files for Fat(ish) Client Application
OH... and Jeff... my apologies for the misnomer of using the term "Fat-ish" client in reference to this... I totally misspoke and should have said Mobile App... but in my mind, any time you have to download something to use it, it's a fat client... sorry! :-/

Whats the BEST Deployment Practice for appsTier (2 Node) with RAC Database

Hi All,
Whats the BEST Deployment Practice for appsTier on Shared APPL_TOP between two application tier Nodes and a RAC Database.
Regards
Satish

Satish,
Could you elaborate; are you looking for how to distribute the tiers(forms, web, conc., etc.); are you asking about PCP (parallelt concurrent processing)..
Regards,

Best SEO practices for text frames in content areas

Hello,
Is it best practice for SEO to keep all text (H1, H2, H3 <p> etc.) on a web page in one text frame and adjust with spacing between paragraphs to add additional elements (Video, widgets etc.) or is it better to use separate text boxes and place elements between them?
Thanks! Below is an example. Should the H1 and both paragraphs be in one text frame space out to fit video or two separate text frames with the video between them?
What is Lorem Ipsum?
Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.
Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.

In the example you give, I would put both sets of text in a single box with a blank spacer between them for the video which would then be layered over the top to appear in the middle.
The video takes longer to download and may also be blocked by the users settings so having it separate allows my text to load (and appear) quickly even if the video encounters issues.
I would only separate the two text sections if they are clearly different in style or content and that’s really only helpful if you want a phone and/ or tablet version of your pages… it makes your life easier then to have the text in separate containers when you want to sync them across the different layouts.

What are the best security practices for your forms and databases

What are some of the best security practices to follow to
ensure your database isn't attacked with injections, forms abused
with snippets of code, and mass spam DB inserts on your
forms?

On Sat, 29 Mar 2008 17:17:01 +0000 (UTC), "jsteinmann"
<[email protected]> wrote:
>What are some of the best security practices to follow to
ensure your database isn't attacked with injections, forms abused
with snippets of code, and mass spam DB inserts on your forms?
David Powers books on php for Dw give detailed advice on how
to do
forms, first with html, and then validate them with secure
php
scripts.
http://foundationphp.com/
~Malcolm N....
~

Best Backup Practice for Solaris 10 running Zones

What is the best practice for backing up a solaris 10 server with zones?
I have solaris 10 installed here on a v240 running 4 zones. All services run off the local zones. We use Veritas backup exec 9, and I have the solaris agent. Should I backup the Data on each of the zones as if they were their own system , or should I only run the backup exec agent on the global zone?
It would make sence to just backup from the global zone, as it is only one system to backup.
We are working on our backup procedures now, and I want to make sure that we are doing this the best way possible.
any thoughts?

Hey there,
Have a look at this excellent guide. I think that you may find what you need here. If not drop an email back. To begin with you need an extra disk(s) since you will mirror your data. This is sort of a pre-requisite.
Good luck,
Pierre
http://www.adminschoice.com/docs/solstice_disksuite.htm

Best Storage Practice for Canon PIXMA MG5420 Printer?

Hello,
My Canon PIXMA MG5420 Printer will be inactive for several weeks. What is the recommended best practice for preventing the ink cartridges and printhead from drying out and clogging due to inactivity?
Thanks,
Bill

Hello BSquared18,
The printer can be stored with the ink tanks and print head installed, make sure the printer is powered off properly so that the print head is capped. If you have print quality issue once the printer is out of storage, follow the steps listed HERE to perform a print head cleaning.
Did this answer your question? Please click the Accept as Solution button so that others may find the answer as well.

Best General practice for performance and tuning.

Hi ,
Can some body let me know the General Best practice for performance and tuning for Oracle Application 11i and 10g.
that can be implemented and suggest in a new environment.
Regards,

Hi,
Please see the following documents/threads.
Note: 744143.1 - Tuning performance on eBusiness suite
Note: 864226.1 - How Can I Diagnose Poor E-Business Suite Performance?
Note: 362851.1 - Guidelines to setup the JVM in Apps Ebusiness Suite 11i and R12
Note: 216205.1 - Database Initialization Parameters for Oracle Applications Release 11i
EBS, performence issue
Re: EBS, performence issue
Oracle Apps Tuning
Re: Oracle Apps Tuning
Regards,
Hussein

Best care practices for macbook pro

Hi I would like to be more active in preventing issues on my macbook. I use it constantly for school and work, and I rely on it heavily, and so I want to keep it running as best as possible for as long as possible, understanding that while some degradation is inevitable, there is a lot that is preventable. I apologize if there is another thread like this, i did search, but didnt find anything. thanks in advance for any feedback from those much more knowledgeable than myself. I'm runnin:
MacBookPro8,1
Os 10.7.2
2.7 ghz Intel Core i7
4 GB 1333 MHz DDR3
Boot ROM Version: MBP81.0047.B24
SMC Version (system): 1.68f98
I'm basically looking for tips on the best diagnostics software for my system, and the best implementation of the software (run it off an external? etc) and best practices for the keeping the computer running smoothly. Thanks again. -dL

Make redundant backups, keeping at least one off-site at all times. One backup is not enough. Don’t back up your backups; make them independent of each other. Don’t rely completely on any single backup method, such as Time Machine.
Keep your software up to date. Software Update can be set to notify you automatically of updates to the Mac OS. Some third-party applications have a similar feature, if you don’t mind letting them phone home. Otherwise you have to check yourself on a regular basis.
Don't install crapware, such as "haxies," “add-ons,” “toolbars,” “enhancers," “optimizers,” “accelerators,” “extenders,” “cache cleaners,” “defragmenters,” “firewalls,” “guardians,” “defenders,” “protectors,” most “plugins,” commercial "virus scanners,” or "utilities." With very few exceptions, this kind of software is useless, or worse than useless. The more actively promoted the product, the more likely it is to be junk. The only software you should install is that which (a) directly enables you to do the things you use a computer for: creating and editing documents, communicating, playing, and so on; and (b) does not modify the way other software works. Never install third-party software of any kind unless you know how to uninstall it.
The free anti-malware application ClamXav is not crap, and although not needed by everyone, may be useful under some circumstances, such as being on a mixed Mac-Windows enterprise network.
Beware of trojans. A trojan is malicious software (“malware”) that the user is duped into installing voluntarily. Such attacks were rare on the Mac platform until recently, but are now increasingly common, and increasingly dangerous. There is some built-in protection against downloading malware, but you can’t rely on it -- the attackers are always one day ahead of the defense. You can’t rely on third-party protection either. What you can rely on is common-sense awareness -- not paranoia, which only makes you more vulnerable. Never install software from an untrustworthy or unknown source. If in doubt, do some research. If a web page seems deceptive, it is deceptive. Any page that prompts you to install a “codec” or “plugin” that comes from the same site, or an unknown site, is untrustworthy. Software with a known corporate brand, such as Adobe Flash, must be acquired directly from the developer. No intermediary is acceptable, and don’t trust links unless you know how to parse them. Any file that is automatically downloaded from a web page without your having requested it should go straight into the Trash. Any website that tells you that you have a “virus” or that something is wrong with your computer is rogue.
Relax, don’t do it. Besides the above, no routine maintenance is necessary or beneficial for the vast majority of users, specifically not “cleaning caches,” “zapping the PRAM,” “rebuilding the directory,” or “repairing permissions.” Such measures are for solving problems as they arise, not for preventive maintenance. The very height of futility is running an expensive third-party utility called “Disk Warrior” when nothing is wrong, or even when something is wrong and you have backups, which you must have. Don’t waste money on Disk Warrior or anything like it.

Dashboard for only Monitoring uplink ports ~~~~~SCOM 2012 R2

Hi Guys,
I want to monitoring signal interface of switch like uplink port. And same i want create dashboard.
Please help me out for creating dashboard for only signal uplink interface port.

If you want to monitor the interface status on a switch, you'll have to build a custom monitor. (Unless you use supported devices e.g. Cisco Catalyst).
Go to authoring > monitors. Under the class 'Node' you can make a new monitor. Choose for SNMP and follow the wizard.
You'll need the OID to monitor. Usually you can find this OID in the MIB's (using a MIB browser) of your device.
MIB's can be found on the vendor website.

Best organizing practice for Apple Loops

Did a search for this, but to no avail.
I recently bought some big fish apple loops. They import beautifully using drag and drop, except for one thing:
On the big fish DVD, each folder of the loop is called something non-descript, like 'Eflat40bpmfull'. So in the loops browser, it's not really possible to search for some of these things, unless you have the original DVD open and know what you're looking for.
So how best to import these - or am I just not using the loops browser in the right way? One idea I had was to change the names of the loops before importing them, so they would be easier to locate, but I'm unsure if that destroys any of the looping tags on them.
Thoughts?

Hi Peter,
I don't think renaming them will destroy the metadata (tags).
If its a hassle renaming every one, maybe a workaround to aid searching, assuming they've come in various sub-folders depending on their content would be to drag & drop them onto the loop browser whilst inside their various folders, giving the folders descriptive names e.g. 'bigfish guitars' & 'bigfish drumfills' - then although the samples would still have their slighty useless names at least you could choose a folder from the jampack organiser drop down menu in the loop browser & flick through a lesser number of slightly uselessly named loops that are at least in the right ballpark for the type of thing you're looking for?
I do find it annoying when sample banks aren't practically named - although samplemagic do seem to have got it right on this count from what I've seen
Cheers
Adam

Best workflow practices for iBooks Author re Enhanced Books?

Hi
I have already published one Enhanced iBook in iTunes (The Wandering Gorillas) about my research in Rwanda. I am now working on several others which will contain much more interactive media - photos and video.
I have vast amounts of unique photos and video - which would be useful for education purposes. So, 'where fools rush in...' I began to creat a new Enhanced iBook with lots of rich media.
However, I immediately started to have serious problems as the fille sizes grew very large - due to the many images and especially because iBooks Author creates versions etc. Thanks to solutions provided on another post by Pondini & Linc Davis these problems have now been resolved and I have regained my 'lost' 100 GB - which had been gobbled up into the folder DocumentREvisions and private/var - you can read their solutions here
https://discussions.apple.com/message/22800295#22800295
Now, I wish to avoid the same problems - and try to get the best possible workflow to use when creating Enhanced iBooks - which contain lots of rich media files such as photos and video.
And so I am asking advice form anyone who has created such iBooks using iBooks author - or who is an expert in the finer workings of this software.
First - Since Apple restrict the final iBook size to 2 Gb (and recommen smaller) I am considering creating my new iBook via separetly published chapters - so that each will be less than the max 2GB. Obvioulsy these will have to have some 'stand alone' usefulness!
But, is this a good practice? has anyone done this? Is there any way for the separate chapters to be later combined - by the user - in order to have a complete, coherent book?
Second, re image file sizes. For best quality on their new Retina Display iPads Apple suggest images of (No larger than ) 5000 x 5000. I used this - and it produced image file sizes of 20 -30 MB. I did some other trials - using images of 1024 x 1024 max - and on my iPad 2 (non Retina) my failing eyes could not see any difference? Their file size was reduced to 2- 300 KB
Would the difference be obvious on a Retina Display though? Should we use the largest possible resolution (and file size) - but this has serious implications for the eventual overall iBook file size.
What is the best compromise?
Third, I have an iMac with an i7 250 Gb Flash drive and a separate 1TB hard drive. All my 'media assets' are on the 1TB hard drive. The still images are exported from my various Aperture libraries - as jpegs of the size mentioned above (and into a newly created folder) . Same with video - to Apple's recommendations for video replay on iPads.
So far I have kept and used the actual iBook Author file - for the new book - on the desktop of my 250GB i7 Flash Drive.
But, as I reported in the other post, this soon caused me huge problems as my disk space was rapidly 'eaten up' by duplicate versions etc.
I am now considering using a separate, 1Tb Firewire External drive for the actual iBook Author file (with backup copies onto another 1TB drive) I am certain this will be slower than using my internal flash drive - but is this 'good practice?' Can it avoid the problems of all 'lost' storage space which caused me such hassles when using my flash drive? Or, will this be minimised if I create separate iBooks from separate chapters?
And, delete uneccessary versions of course!
I do feel that this is an emerging issue which many will face - as we incorporate more and more rich media into our iBooks.
Any help or suggestions are gratefull received!

Thunderstorm played havoc with box from my ISP - and I am ruuning on a reserve and internet access flaky!
but, here goes with one reply. Will have to wait later re tests for your other suggestions when weather calmed down
Re 'as Apple suggessted' I should have said - no larger than 5000 x 5000 pixels (the 3240 was my vertical proportional dimension from choosing 5000 horizintal dimension)
this is from Ibooks Author help
Add and edit photos and other images
You can add a photo or other image (in a supported file type) to a page in your book by dragging the photo from theMedia Browser, the Finder, or another application.
After adding a photo to your book and positioning it on the page, you can use the tools in iBooks Author to mask (crop) the photo; change its brightness, contrast, and other image qualities; or erase its background.
Important: When you export your book for distribution outside the iBookstore, images are automatically scaled to a width of 2048 pixels, which is optimized for iPad with Retina display. Readers can view these images on other iPad models, but the viewing experience might be slower. You can improve the viewing experience on iPad models without Retina display by using images no wider than 1024 pixels.
ShowMask (crop) a photo
ShowChange a photo’s brightness, contrast, and other settings
ShowRemove the background or unwanted elements from a photo
Important: Images can’t be larger than 25 megapixels (5000 x 5000 pixels) or 50 MB.
SEE ALSO
Add an accessibility description to an object
Add a title, label, or caption to an object
Was this page helpful?Send feedback.
Copyright © 2012 Apple Inc. All rights reserved.

Best AF practice for video

I'm using a EOS 7D for shooting 1080p HD video. New to the camera and DSLR for video. I've looked at several tutorials but still don't have a good understanding of what the best practices are for focusing.
If I put the camera in AI Servo will it auto focus if the subject moves away or toward me with out me depressing the back AF-ON or shutter button? Or must I press one of those buttons to make the camera continue focus?
Do you know of any tutorials or video teachings that go through using focus with video?
Thanks in advance!
Solved!
Go to Solution.

The 7D is not going to perform continuous auto-focus in video. AI Servo mode is for still photography. The 70D _will_ perform continuous auto-focus during video and does an outstanding job of it (due to it's ability to do phase-detect AF directly on the camera sensor.)
Every camera *other* than the 70D (including all other brands -- the 70D is unique right now) have to do "contrast detect" auto-focus. Contrast detect has to analyze contrast differences between adjacent pixels, make an adjustment and then evaluate if the contrast increased or decreated. That tells the camera if it's even focusing in the right direction... then it has to re-focus, re-evaluate, re-focus, re-evaluate... until it finally optimizes contast and that means the image is finally focused. This results in a game of "focus hunt". It's slower and anyone watcing the video will see the camera hunt for focus.
The better way is to manually focus the camera. If you were making a more serious video you'd even pre-plan the focus changes -- and wrap a piece of tape around the focus ring and use a marker to place the focus points on the ring (you pre-focus the camera to each position you plan to use during that clip). Having done this you can smoothly move focus from subject to subject just by looking at the marks you made on your focus ring. This isn't going to help you in ad-hoc video clips -- it's what you'd do if you were thinking about film-making.
The 7D will re-focus if you press the AF-ON button... but it will focus hunt when it does this.
Tim Campbell
5D II, 5D III, 60Da

Best external drive for backup using USB port on TC

My Time Capsule drive is too small and I would like to use a USB hub and plug it into the port on the device. I will have a printer and external drive plugged into the hub, which will be plugged into the Time Capsule.
What type of external drive is best for this, considering it will be on all the time?
Thanks.

FWIW, I have a number of the NewerTech miniStack external hard drive enclosures. You can get these pre-equipped with drives or add your own. They run very cool and quiet and I've not experienced any problems with them and have been operating them 24x7 for quite some time.

Best settings practice for Cisco 1600 Standalone WAP.

Hi,
I'v just begun self training for CCNA wireless certification, but I'm ineed of some assistance right away for practical purposes. We have several 1600's in our environment. Are there any best practice / settings guides available for getting the fastest speed out of our 1600's?
So far, the fastest speeds (using speedtest.net for testing) I've been able to get is 10down/10up on the 2.4ghz radio and 20 down/20 up on the 5ghz radio. Our ISP service is provided at 200mbps down/20 up.
On both, I'm looking to kick up the speeds as fast as possible. We have several of these WAPs in our work areas., so physical distance between WAP and client is about 50ft, maximum.
Any suggestions on how to balance out the various setting would be great.
Thanks!

There are only a few things you can influence with configuration. In the end there simply needs to be space in the RF and enough clean and strong signal. This depends on the access-point placement, antenna selection and very much the client. This is the reason we always advice to do a site survey, which I will advice you to do so as well.
Besides the survey you also should check the following things:
Check your channels (stick to a 1,6,11 plan for b/g/n)
Try to only enable AES encryption your SSID (which is needed for the n and ac speeds anyway)
Tune your data-rates (make 12Mbit/s mandatory and set the rest to supported)
Tune the configured power levels on your access-point (you need input from your site survey for this)
Please rate useful posts :-)

Best QOS Practice for a congested Uplink Port?

Similar Messages

Maybe you are looking for