Purpose of GC Threshold

What is the purpose of the GC threshold setting feature in WLS 7.X? JVM GC
documentation and other GC related information resources suggest this is at
best pointless and probably degrading to system performance. We are talking
specifically of a threshold trigger versus a distributed synchronous GC.

Similar Messages

How to set threshold level JPG images

Please I need help with setting the right threshold value for my images. I am new with Photoshop. I need to interpret images consisting of sand oil and water in a glass medium. I need to calculate in pixels the volume of each phase (oil and water) with time. I converted the image to threshold and by default the setting is 128 level. This gives a rather high pixel value for the black background. adjust to other levels like 70 makes the white image more. Does some know how I can determine the correct threshold level to reflect that in the original image?
Also how can set the threshold value to be the same for all the files (example 600 jpg files).
Thanks a lot
Engyy

Also how can set the threshold value to be the same for all the files (example 600 jpg files).
Create an action and use batch processing or import your images as a sequence and treat it as a video in the timeline. The rest is unclear. For all intents and purposes setting a threshold is completely arbitrary and factors like the color of the sand or whatever can influence it since e.g. red sand will naturally accentuate the red channel more than green and blue. Come to think of it, your methodology is rather unscientific in the first place, but providing one of your images as a sample reference might help people to be able to advise...
Mylenium

JMS Server Message Buffer Size & Thresholds and Quotas settings

On WLS10MP1,
For persistent messages:
1.Does "JMS Server Message Buffer" setting serve the same purpose as "Bytes Threshold High" under Threshold ?
2.If no, can someone explain the difference pls.
Many thanx,

Message Buffer Size relates to the number of message the JMS server keeps in the memory. The value of this determines when the server should start paging the message out of memory to a persistence store. So this is directly related with the memory/storage issue and the size of messages.
Bytes Threshold High relates to the performance of the JMS server. When this limit is reached JMS server starts logging the message and may even instruct he producer to slow down the message input.
So the if you get Bytes Threshold High messages that means you should check on your consumer (MDB who is picking up messages from the que), and try to increase its performance.
However if your Message Buffer Size is crossing limits then you should think of increasing the momory so that more messages can be kept in memory and disck IO can be reduce.
Anyone wants to add something more to it?

Paywall Limit/Threshold Clarifications

Hi Gang!
Thanks in advance for your input.
I've read most (I think) of the official documentation out there in regards to Social Sharing and Free Articles within the DPS environment, but a few things are still not 100% clear to me. I'm hoping someone can clarify them for me.
Device Viewer Paywall Threshold & Web Viewer Paywall Threshold: What is the difference between them? I understand that the former pertains to the Device Viewer while the latter pertains to the Web Viewer. I just don't understand why they are independent of each other. Is this just an added option for publishers if they so choose to meter distribution differently between both options, or am I missing something fundemental here?
Metered Articles: They can be viewed by a reader until the Paywall Threshold is reached (whether it be the via the Device Viewer or Web Viewer). Gotcha. But is the threshold reset with each issue? Let me try and explain what I mean. Say we set a Device Viewer Paywall Threshold at 3 articles. Then we publish an issue with 2 metered articles and reader x (who does not have a subscription or other entitlement) reads both articles (2 out of the 3 maximum articles allowed before being cut-off). Next month the reader downloads the new issue, which also has 2 metered articles. Is reader x limited to just 1 more article moving forward (having used only 2 of the allowed 3 articles the previous month) or is he allowed to access both because the threshold of 3 articles as set by the publisher is on an issue-by-issue basis and "resets" each time? I'm guessing it's the latter and the purpose of the threshold number is to limit the niumber of articles shared per issue, while still giving the audience a selection to choose from. I at first thought the threshold number was a sort of lifetime number to prevent someone from getting "free" content every month in perpetuity. I await the official confirmation.
Sorry for the long post.
Cheers!

1. Device Viewer Threshold vs. Web Viewer Threshold - Yes, you're right. A publisher might want to limit the number of articles users can read on a device viewer but not on the web viewer, or vice versa. Or, they might want to apply different threshold settings.
2. Yes, this threshold is reset with each issue.

Multiple OEM agents installed

Good morning all,
Our test database is a single node db in Sun Microsystems Inc. (SunOS 5.10). The database version is 10.2.0.4. We have an oem agent (10.2.0.4) installed in this db server communicate with a 10g grid control. Everything is working fine. Now we are trying to migrate this db server under 11g grid control. So what we did is installing new version of agent (10.2.0.5) into a different directory and configure this agent to talk to 11g grid control. So basically we have two agents installed in this db server. One is 10.2.0.4 agent which is talking to 10g grid control and another agent 10.2.0.5 which is talking with 11g grid control. Although both agents are running but we experience some issues: First, we noticed that all listeners in the server are in “Status Pending” in 11g grid control but actually all listeners are up. Second, a notification rule has been setup in 11g grid control to monitor one db instance and I purposely set the threshold very low so I can receive the alert email. But the all the alert emails actually comes from 10g grid instead of 11g where I set threshold very low. I am not sure how agent works internally so I am not sure if multiple agents will ever be work in one db server. Does anybody have any thoughts on this? Any insights to share?
Greatly appreciate your help!
Shirley

Hi,
You not there are most of the things about a target which are stored locally on the target rather than on the GC itself. For example, if you set t matrix value for a database target, it is stored on the database itself whether you use 10g or 11g GC . Here in your case, i think your both GC are fighting together to monitor a single target which is configured on both and i think it will not be supported (you can confirm from MOS). So my advise will be to configure your targets with only one GC (either 10g or 11g)
Salman

Threshold Value in Sales Order Processing

Dear All,
I have a question for which I need inputs and suggestions from you.
We have a requirement in the returns processing to do check on the value of goods being returned and if this value goes beyond a certain threshold (pre-agreed value with the customer) then the system will raise a message and should not allow further processing of returned goods.
This is the approach that we think of following:
1) we store the initial Threshold amount as a header condition in the Return Authorization (RA) document.
2) when creating Return Orders (RO) with reference to RA we check the value of all the open RO's against the RA
3) We make reference to RA as mandatory at the time of RO creation
3) if total value exceeds then we raise a error message
Basically we plan to do this check in real-time, whenever a RO document is being entered we do this calculation in real time.
Please advice if you think this is a feasible approach, or in case you have any other suggestions for storing Threshold amounts in SAP and working with.
Thanks a lot in advance.
BR,
Sahadj

dear friend,
your approach looks okay from my point of view...
you would also append VBAK table creating a new custom field for your purpose and adjust the relevant code(s)
good luck!

Performance Counter monitors stay unhealthy even when values drop below thresholds

I'm investigating some (SCOM 2012) alerts on our Exchange 2013 (SP1) environment:
Malware filtering is taking too long (90th percentile)
Mailbox Transport Submission is not keeping up with the work...
Queue Alert: Internal Aggregate Delivery Queue (Normal Priority) exceeds threshold
The total number of messages in shadow queues exceeds 1500
All the monitors are Performance Counter based monitors. When I investigate those performance counters (found via the Crimson Channels in the Eventlogs ../ActiveMonitoring/Monitordefinition) all affected counters have already dropped below threshold values
but the associated Health Sets stay unhealthy
IE: Get-ServerHealth -Identity '<Server>' -HealthSet '<HealthSet>' still reports Total.Shadow.Queue.Length.Above.Threshold.Monitor as UnHealthy while perfmon reports values way below 1500 for that particular server.
How to (re)evaluate the monitors ?

For investigation purposes I've added some perfmoncounters in SCOM
Another server now alerts 'Total number of messages in shadow queues exceeds 1500'. On the specific server the treshold never reached 1500; it was about 5 at the time the alert fired ?!?
Now it seems the alerts fire for no reason and cannot be reset (I want Managed Availability to reset the alert, not scom)
The XML of Eventlog\Applications and Serviices/Microsoft/Exchange/ActiveMonitoring/MonitorDefinition/
- <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
- <System>
<Provider Name="Microsoft-Exchange-ActiveMonitoring" Guid="{ECD64F52-A3BC-47B8-B681-A11B7A1C8770}" />
<EventID>7</EventID>
<Version>0</Version>
<Level>4</Level>
<Task>7</Task>
<Opcode>0</Opcode>
<Keywords>0x4000000000000000</Keywords>
<TimeCreated SystemTime="2014-07-08T12:08:40.460212000Z" />
<EventRecordID>15316181</EventRecordID>
<Correlation />
<Execution ProcessID="38980" ThreadID="30424" />
<Channel>Microsoft-Exchange-ActiveMonitoring/MonitorDefinition</Channel>
<Computer>XXXX</Computer>
<Security UserID="S-1-5-18" />
</System>
- <UserData>
- <EventXML xmlns:auto-ns2="http://schemas.microsoft.com/win/2004/08/events" xmlns="myNs">
<Id>577</Id>
<AssemblyPath>C:\Program Files\Microsoft\Exchange Server\V15\Bin\Microsoft.Office.Datacenter.ActiveMonitoringLocal.dll</AssemblyPath>
<TypeName>Microsoft.Office.Datacenter.ActiveMonitoring.OverallConsecutiveSampleValueAboveThresholdMonitor</TypeName>
<Name>Total.Shadow.Queue.Length.Above.Threshold.Monitor</Name>
<WorkItemVersion>[null]</WorkItemVersion>
<ServiceName>HubTransport</ServiceName>
<DeploymentId>0</DeploymentId>
<ExecutionLocation>[null]</ExecutionLocation>
<CreatedTime>2014-07-08T12:08:40.4602120Z</CreatedTime>
<Enabled>1</Enabled>
<TargetPartition>[null]</TargetPartition>
<TargetGroup>[null]</TargetGroup>
<TargetResource />
<TargetExtension>[null]</TargetExtension>
<TargetVersion>[null]</TargetVersion>
<RecurrenceIntervalSeconds>0</RecurrenceIntervalSeconds>
<TimeoutSeconds>30</TimeoutSeconds>
<StartTime>2014-07-08T12:08:40.4602120Z</StartTime>
<UpdateTime>2014-07-08T12:04:53.7722193Z</UpdateTime>
<MaxRetryAttempts>0</MaxRetryAttempts>
<ExtensionAttributes>[null]</ExtensionAttributes>
<SampleMask>EDS/Performance Counter/MSExchangeTransport Shadow Redundancy Host Info\Shadow Queue Length\_total</SampleMask>
<MonitoringIntervalSeconds>600</MonitoringIntervalSeconds>
<MinimumErrorCount>0</MinimumErrorCount>
<MonitoringThreshold>1500</MonitoringThreshold>
<SecondaryMonitoringThreshold>1</SecondaryMonitoringThreshold>
<ServicePriority>2</ServicePriority>
<ServiceSeverity>0</ServiceSeverity>
<IsHaImpacting>0</IsHaImpacting>
<CreatedById>50</CreatedById>
<InsufficientSamplesIntervalSeconds>28800</InsufficientSamplesIntervalSeconds>
<StateAttribute1Mask>[null]</StateAttribute1Mask>
<FailureCategoryMask>0</FailureCategoryMask>
<ComponentName>ServiceComponents/HubTransport/High</ComponentName>
<StateTransitionsXml>[null]</StateTransitionsXml>
<AllowCorrelationToMonitor>0</AllowCorrelationToMonitor>
<ScenarioDescription>[null]</ScenarioDescription>
<SourceScope>[null]</SourceScope>
<TargetScopes>[null]</TargetScopes>
<Version>65536</Version>
</EventXML>
</UserData>
</Event>

Configuration of Cost threshold for Parallelism

Hi all,
I want to adjust the 'Cost Threshold for Parallelism' on the SQL server running OLTP databases.
MSDN says that this is an estimate query duration - in seconds.
Is this estimate available in a DMV? I want to look at my server's query workload & assess which queries I want to be considered for parallel processing and which ones I don't.
Some potential candidates include:
dm_exec_query_stats.max_elapsed_time / dm_exec_query_stats.execution_count
I don't want to use this as this actual - where the setting is an estimate (on some arbitrary server).
I thought it might be the Estimated Subtree Cost in the query plan? Though I've read this has nothing to do with an estimated number of seconds.
Any assistance greatly appreciated.
Cheers, Clay

Thanks for the replies SQL24 and Ashwin -
I am looking at a server which is struggling to keep up with a substantial increase in volumes.
My top 3 waits are:
- CXPACKET @ 55%
- LCK_M_IX @ 25%
- SOS_SCHEDULER_YIELD @9% (25% signal waits)
The production server has only 2 CPUs - the current MAXDOP and Cost Threshold for Parallelism are the default, 0 and 5.
Though 2 CPUs is a tiny amount of CPU - I'm hesitant to request more until I can prove CPU is struggling (my
In regards to "issues relating to parallelism" - my guess (I'm not a DBA) is
that too many queries are being parallelised - I'm basing this guess on the high level of waits on CXPACKET, the fact that it is primarily managing small transactions (+ plus some manifesting/grouping of those transactions each afternoon) & the
fact that I currently only have 2 CPUs!
To my original question - MSDN states "The
cost refers to an estimated elapsed time in seconds required to run the serial plan on a specific hardware configuration"
http://technet.microsoft.com/en-us/library/ms188603(v=sql.105).aspx
It is the 'specific hardware configuration' part of this description that renders all of the actual times (be it a max or average) that I can obtain from
dm_exec_query_stats worthless.... for this purpose.
My guess (again) was that it could be compared to the query plans estimated subtree cost - but before I attempted extraction of this info from query plans - I wanted to confirm.
Thanks again.

Binary threshold - U16 Image

I have a simple example which I have attached (Vision 2013 but can save for previous if need be) in which I would like to apply binary thresholding to a range of values (minimum --> mean) in a U16 image. This example is based on the particle analysis binary threshold example included with Vision for U8 images. The thresholding does not appear to be working properly (to hilte the range of values red) and areas outside of the range appear to be getting thresholded as well. Ultimately, what I would like to be able to do is actually display the gray scale (rather than binary) image and have the thresholded range show in red against the gray scale image.
Any thoughts.
Solved!
Go to Solution.
Attachments:
U16 Threshold Example.llb ‏4 KB

Bruce -
Thanks for taking the time to reply. The first question that needs to be answered is if the binary thresholding VI is properly working to hilite in red the range of values I have specified (non-zero min --> mean) for a U16 image. The U8 version of this example VI works just fine. I wanted someone to run my VI and see that what is getting hilited does not seem to match up with the non-zero min --> mean range.
In terms of thresholding on top of an actual gray scale image, another idea is to just find the pixels that fall within the range I have specified, and draw an overlay of points on top of the image. That should work fine. However, this is mainly for visualization purposes.
I really need the thresholding itself to work properly for 16-bit images in order to do a particle analysis. So I needed a sanity check to see if somehow I am doing something wrong in the thresholding (this is a pretty simple VI, not sure what could be going wrong, so am wondering if it is not properly operating on U16 data).
Sincerely,
Don

GD Threshold Issue

We are getting this message:
<Sep 23, 2003 10:21:05 AM PDT> <Warning> <WebLogicServer> <000335> <Percent
of s
tarting server free memory is now at "4%". This is less than the configured
thre
shold of "5%". We will try to garbage collect, but you may consider
rebooting.>
WLS 7.X has a GC threshold page. The benefit of a forced GC is that you are
distributing the time required for a single GC across several pre-emptive
GC calls. I have seen no indication or information that would indicate that
threshold drive GCs are anything but a bad idea.
Consider what is being done here. When there is memory available there is no
need to run the GC. When there is memory available your request for memory
will be fulfilled. When a request cannot be fulfilled the GC will run and
release memory. If and only if all memory is held by non-collectible objects
will you run out of memory.
Setting a threshold means that you are lowering the trigger point so the GC
runs when there is no point in doing so. Consider what happens when you get
to the threshold of, say, 10% and there legitimately there is only 10% free
memory: everything else is reachable and by definition it cannot be
collected. The GC will start to run continuously! You have effectively taken
10% of your memory out of service. It seems as though setting artificial
memory thresholds is counterproductive.
I added a memory watchdog for diagnostic purposes a year ago. We determined
that we there was no legitimate out of memory error as the GC would always
return to about the same baseline. Running GC at regular intervals fixed did
not solve the crashing problem and out of memory problem we were trying to
solve. Apparently there is a known problem with WLS and hotspot but WLS
support has precisely zero information on that a year ago.
Question: what is the purpose of the GC threshold and can anybody point me
to a credible information source that actually recommends this. This is not
the same as running the GC at regular intervals which has the benefit of
distributing the effort over time. Can anybody explain what the threshold is
all about and whether I am missing some important information that has led
me to the wrong conclusion about this?
TIA

I've had the same question on my mind....
"dealmein" <[email protected]> wrote in message
news:[email protected]...
We are getting this message:
<Sep 23, 2003 10:21:05 AM PDT> <Warning> <WebLogicServer> <000335><Percent
of s
tarting server free memory is now at "4%". This is less than theconfigured
thre
shold of "5%". We will try to garbage collect, but you may consider
rebooting.>
WLS 7.X has a GC threshold page. The benefit of a forced GC is that youare
distributing the time required for a single GC across several pre-emptive
GC calls. I have seen no indication or information that would indicatethat
threshold drive GCs are anything but a bad idea.
Consider what is being done here. When there is memory available there isno
need to run the GC. When there is memory available your request for memory
will be fulfilled. When a request cannot be fulfilled the GC will run and
release memory. If and only if all memory is held by non-collectibleobjects
will you run out of memory.
Setting a threshold means that you are lowering the trigger point so theGC
runs when there is no point in doing so. Consider what happens when youget
to the threshold of, say, 10% and there legitimately there is only 10%free
memory: everything else is reachable and by definition it cannot be
collected. The GC will start to run continuously! You have effectivelytaken
10% of your memory out of service. It seems as though setting artificial
memory thresholds is counterproductive.
I added a memory watchdog for diagnostic purposes a year ago. Wedetermined
that we there was no legitimate out of memory error as the GC would always
return to about the same baseline. Running GC at regular intervals fixeddid
not solve the crashing problem and out of memory problem we were trying to
solve. Apparently there is a known problem with WLS and hotspot but WLS
support has precisely zero information on that a year ago.
Question: what is the purpose of the GC threshold and can anybody point me
to a credible information source that actually recommends this. This isnot
the same as running the GC at regular intervals which has the benefit of
distributing the effort over time. Can anybody explain what the thresholdis
all about and whether I am missing some important information that has led
me to the wrong conclusion about this?
TIA

Metrics - Thresholds - Best Practices

All,
I installed em grid control 11g and configured the targets and notification rules. Now i am trying to set up thresholds, is there a best practice threshold values that someone can share for various metrics at weblogic level. I know this is a generic question and i can tune the thresholds for my environment/application usage. But there must be a ball-park threshold document somewhere for alert purposes.
Thanks in advance,
Prasad.

Hi Prasad,
There is no document giving recommendations on threshold values. Setting thresholds really depends on your environment. I recommend looking at performance history...looking at for instance a timeframe that had heavy load on the application/server but performance was good. Then set thresholds according to that....either adjust up or down as needed.
Thanks,
Nicole

QoS - Min and Max Thresholds

Hello Everyone,
I'm studying QoS and I'm wondering how does a Cisco Engineer find out what the best Minimum and Maximum Threshold rates are when using WRED?
Where would one find these values?
I'm under the impression that it requires a study that's based on the size of the queues, is that correct?
But then, I come to the conclusion that even if the queues are big enough, application critical data like (VOIP) would suffer from delay..
So, how do you calculate the min and max threshold values..
Thanks ahead of time..
PR

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
Setting WRED parameters, for optimal performance, is (IMO) very tedious. Lots of variables involved.
For starters, it shouldn't normally be applied to traffic that's not rate-adaptive, such as VoIP.
It works best for multiple large bulk data transfer TCP streams that are using dropped packets for their principle flow control. Then it can be optimized to maximize their "goodput".
To answer your question, you might study RED literature, and its many variants, found on the Internet.
You also want to study, if you're not already very familiar, with how adaptive flow rate protocols, such as TCP, behave when packets are dropped. (With TCP also be mindful of the different "flavors", i.e. Tahoe, Reno, New-Reno, etc.)
If you have specific questions, post again to this thread, and I might be able to help.

Peak detection (threshold detector) from spreadsheets.

Hello!
I am working on VI that reads two signals, detects peaks that are above a specific treshold and then estimates time difference between these peaks in samples and in seconds.
Recently I have obtained two waveforms using an oscilloscope in a laboratory. Then I saved these two graphs in a spreadsheet in Excel format (first number is time, second is amplitude). Time difference between them is really small, just a few nanoseconds according to the oscilloscope. I think treshold should be 0.3.
My question is how to make Threshold Detector.vi read values from two excel files? Or maybe I need to use another instrument for this purpose. I am quite confused.
I attached all materials I have.
Any help is highly appreciated!
Thank you.
Attachments:
Inception 2.vi ‏55 KB
position S1-S2a.jpg ‏3004 KB

I tried Read From Spreadsheet File.vi, but have not got any positive result. Probably it reads my spreadsheets, but peak detection does not work properly. When I set up any threshold in Threshold Detector.vi that is more than zero (for example, 0.01), final results turn to zero. The same occurs when indexes in Index Array are not zeroes.
I attached my vi saved for LV2009. Please, have a look.
Thanks.
Attachments:
Inception 2.vi ‏49 KB
Peak detection.jpg ‏108 KB

About QoS configurable thresholds...

Hey,
I was reading through QoS guides and got confused about the threshold things.......
According to document,
you can use following command to configure maximum memory allocation for egress queue.
mls qos queue-set output qset-id threshold queue-id drop-threshold1 drop-threshold2 reserved-threshold maximum-threshold
you can use following command to configure allocated memory for egress queue.
mls qos queue-set output qset-id buffers allocation1 ... allocation4
So now questions:
1. Why queue-set? There will be 4 queue for egress anyway...
2. If queue-set is useful feature, Why no queue-set for ingress?
3. Why multiple thresholds? For ingress, there will be one threshold per queue.
4. When is drop-threshold1 used? When will drop-threshold2 used?
Thanks,
/S

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
Or, perhaps this isn't the best forum for this question.
See if this document helps: https://supportforums.cisco.com/docs/DOC-8093

MLS QoS Thresholds

Hi all,
I wanted to get to know more about QoS and thought that there was no better way than document and understand the output of auto qos.
However I want to verify some things to check that I understand it correctly.
The 2 commands I want to check on are the following:
mls qos queue-set output 1 buffers 15 25 40 20
mls qos queue-set output 1 threshold 4 60 150 50 200
mls qos queue-set output 1 buffers 15 25 40 20, this command divides the port buffer space between the 4 egress queues.
1. Does it always need to add up to 100%? Or if it doesn't, will the remaining percentages be added to the common pool?
mls qos queue-set output 1 threshold 4 60 150 50 200
Are the following statements correct?
2. packets belonging to queue 4 threshold 1, will be dropped when the port buffer for queue 4 is 60% full (60% of 20 buffers = 12 buffers)?
3. packets belonging to queue 4 threshold 2, will be dropped when the port buffer for queue 4 is 150% full (150% of 20 buffers = 30 buffers)?
3b. how does this work? since this are more buffers than reserved or available for the queue? what if no more buffers remain in the common pool?
4. packets belonging to queue 4 threshold 3, will be dropped when the port buffer for queue 4 is 100% full (100% of 20 buffers = 20 buffers)?
5.the queue can steal up to 200% buffers from the common pool, can this ever be reached? since we will be dropping at 60% (th1), 150% (th2) and 100% (th3) so we will max reach 150% and not 200%?

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
#1 It's recommended sum of all four queues buffers equal 100%. I don't know what happens if they they don't.
#2 Yes and no, it's 60% of 20%.
#3 Yes and no, it's 150% of 20%.
#3b If not buffers to borrow from common pool, it will drop.
#4 No, it's 200% of 20%.
#5 Yes, buffers are borrowed are (attempted) borrowed from common pool when reserve buffers are all used (and WTD not reached). Th3 is 200%, not 100% of 20%.
Great document on 3560/3750 buffering: https://supportforums.cisco.com/document/31581/egress-qos

Purpose of GC Threshold

Similar Messages

Maybe you are looking for