Too high cpu load with HP 6510b
Hello guys,
I have a very high cpu load on my laptop even if nothing despite htop is running. I guess it's a problem with acpi because when I deactivate acpi completely over the bootline I have no such problems.
How could I fix that?
Thank you in advance.
Hello B,
unluckily noacpi doesn't help but acpi=off. That seems to be the only way to stop the high load.
Is there any patched acpi or something else?
I also think I picked the bad revision with fan problems and some more...
Similar Messages
-
Problem... High CPU load with HDD. Please Help
Hello all,
Im stuck. Ive got a MSI KA790GX which has poor HDD access/transfer, low burst rate and very high CPU load.
As far as i know the Western Digital 640gb WD6401AALS Black should be a fast hard drive. Which according to reviews/HD Tune,
should be:-
Minimum 60mb/sec
Maximum 120mb/sec
Average 95mb/sec
Access 12ms
Burst 110mb/sec
CPU 5%
No matter what I try, i always get poor results...
My MSI KA790GX
Minimum 41mb/sec ~ 54mb/sec
Maximum 104mb/sec ~ 108mb/sec
Average 86mb/sec ~89mb/sec
Access 12ms
Burst 60mb/sec ~ 80mb/sec
CPU 26% ~ 44%!!!!!!!!
I could accept not getting the same high speed results as the review/reference, even if the burst rate sucked.
But one of the cores goes to 50%-60% load during hdd activity. Thats just crazy!!
I have tried...
IDE mode, AHCI, RAID Ready, RAID Single Disk
SATA Port 1, 2, 3, 4 + 5
Different SATA cables
Flashed bios, cleared bios, optimised defaults, failsafe defaults
Windows XP 32bit, Vista 32bit, Vista 64bit. fresh installs
MSI CD Drivers, MSI Web Drivers, AMD/ATI Web Drivers
... and combinations of.
I have also tested the WD640gb Black on another older PC, with Nvidia (55?) chipset and slower (1.9ghz x2) cpu.
Transfer rates were slightly better than my KA790GX but, Burst Rate = 115mb/sec and CPU usage = 13%.
I also tried a Seagate 1.5tb (ST31500341AS) on both PCs,
Nvidia chipset
Minimum 65mb/sec
Maximum 134mb/sec
Average 107mb/sec
Access 14ms
Burst 83mb/sec
CPU 7%
On my MSI KA790GX
Minimum 55mb/sec
Maximum 120mb/sec
Average 100mb/sec
Access 14ms
Burst 87mb/sec
CPU 33%!!!!
Could someone please please help, ive spent hours and hours on this. Its doin my head in.
I just hope its something stupid ive missed, cos i still not used my HD4850 from christmas :-(
Thanks.Sm3K3R
Not sure how to check PIO/DMA mode on a SATA device. IDE devices show DMA enabled.
BIOS version 1.4 - 29 December 2008, and now version 1.51
Fredrik
I have Vista 64, but i did try XP32 and Vista32 and got similar poor results.
I tested HDD speed with, HD Tune 2.55 and HD Tune 3.50 trial.
No auto-start programs really, Windows clean installs + any drivers.
Quadrifoglio Verde
CPU @ default 2800mhz, All mem bios options to auto/defaults
Windows Experience Index, CPU Score = 5.3
AMD Overdrive
Integer Computation = 8220
Floating Point Computation = 5325
Memory speed = 1770
Cache Speed = 10790
System Performance = 7403
Bas
I used HDTune because I could reference to other results. Whats a better program to use?
Windows Experience Index, HDD Score = 5.9.
But I have no reference and cant be sure of cpu load.
Sepher
Bios defaults to C&C and C1E off so i have tried them. Also windows power settings, balanced/high performance.
UPDATE
Reinstalled Vista64 but loaded the AHCI driver from the disk config screen. This seems to have stopped the high cpu usage during disk activity.
HDTune with Windows Power settings @ High Performance,
Min = 52mb/sec
Max = 106mb/sec
Ave = 88mb/sec
Access = 12ms
Burst = 77mb/sec
CPU = 3.4%
HDTune with Windows Power settings @ Balanced
Min = 54mb/sec
Max = 104mb/sec
Ave = 87mb/sec
Access = 12ms
Burst = 65mb/sec
CPU = 12%
So now transfer speed is about 10% slower than reference. But burst speed is terrible.
I also used IBM Feature Tool to change the HDD AAM to 'performance mode' 254 as suggested elsewhere.
Long shot as the hdd performs ok with other hardware. same results.
Also MSI support asked me to try a newer bios, version 1.51. same results.
Anyone know what the AMD 7-Series Chipsets Disk Performance Utility does, and should i run it in Vista64?
Thanks for all the replies, hope i can sort this, not played any PC games since Quake 3 on my ti4600... -
High CPU load with Flash on Mac Leopard
Dear Adobe,
I am always running the latest version of the Flash plugin on my Macbook. Ever since even with the simplest Flash movies running in a browser (no matter whether Safari, Firefox, Camino, Opera, ...) the CPU load jumps oh so high the fans will spin at full speed and the CPU starts to create subtropic temperatures. Years ago I was running a Windows PC where the CPU load wasn't even close to such high CPU loads. Oh so many Apple users would very very much appreciate Flash Player being at least close to as efficient as the Windows version.
Thanks in advance for making Flash movies more enjoyable on the Mac platformAdobe need to realize that once their great product with a lot of penetration into the web ad + ria market is starting loose its ground as they simply can't address ongoing issue on their poor flash plug-in performance other than on windows platform. There are a lot of flash enthusiast out there but this poor support from the Adobe doesn't help this much further. One of the reasons why other software vendors pushing for more open standard RIA platforms other than Flash based...
Only suggestion that I have with Adobe is that if you can not get it right then, it may help to open up flash player source to the community. I am pretty sure that there are a lot of developers out there who can help out to drive better code base for the player and in turn make flash more adopted by non non ms windows user base. At the moment, I have installed "Click2Flash" and "FlashBlock" to save majority of CPU cycle on my MacBook Pro. Imagine the impact on this for future Adobe product. All online ad will slowly move away from flash based and in turn loosing entire market share it has successfully built over the years... Well must note that MS isn't much better either... MS RDP client is permanently turned off from my system, now I'm using CoRD instead..
I sincerely hope to see some major performance improvements in near future. -
Has anyone noticed a high cpu load when transferring over nfs4.
My arch box has a dual core amd with 4 gig of ram. It has a rocketraid 2320 card being used for a 8 1TB mdadm raid 6.
I am transferring from ubuntu -> arch box. I am noticing high cpu on both sides. I dont know if its an nfs issue or what.
On the arch side I am seeing high usage from nfsd , flush-9:0, and an mdadm process. Both cores are sitting around 50%.
Anyone run into this or know if its an issue with nfs4? or maybe different nfs versions of ubuntu and arch.
Thanks.Adobe need to realize that once their great product with a lot of penetration into the web ad + ria market is starting loose its ground as they simply can't address ongoing issue on their poor flash plug-in performance other than on windows platform. There are a lot of flash enthusiast out there but this poor support from the Adobe doesn't help this much further. One of the reasons why other software vendors pushing for more open standard RIA platforms other than Flash based...
Only suggestion that I have with Adobe is that if you can not get it right then, it may help to open up flash player source to the community. I am pretty sure that there are a lot of developers out there who can help out to drive better code base for the player and in turn make flash more adopted by non non ms windows user base. At the moment, I have installed "Click2Flash" and "FlashBlock" to save majority of CPU cycle on my MacBook Pro. Imagine the impact on this for future Adobe product. All online ad will slowly move away from flash based and in turn loosing entire market share it has successfully built over the years... Well must note that MS isn't much better either... MS RDP client is permanently turned off from my system, now I'm using CoRD instead..
I sincerely hope to see some major performance improvements in near future. -
High CPU Load with Alesis Multimix Usb 2.0 mixer
Hi everybody,
i'm still playing around with logic studio 9, mainly mainstage 2.1.1 for a possible future live use.
Today i tried to use my Alesis Multimix 8 usb2.0, just as the output devices, using software instruments with a m-audio keystation.
I always get cpu peak load when playing notes, with every patch, either internal apple instrument or third party au instruments, when idle cpu load goes to about 40%...
If i revert back to internal output / internal input, cpu load drastically goes down to 10% idle and max 50-60% with internal instruments,max 80% with 2-3 layers of third party au instruments (Omnisphere, loungelizard, m-tron)...
I know that Alesis mac drivers for snow leopard are at a beta stage, but am i the only one experiencing this or is it a known problem?
Thanks,
Alex.Gotenks82 wrote:
Hi Pan,
i made a different test today...
I know it's not the same but i tried using a guitar rig on mainstage (guitar on channel1 in the mixer), and even with 256 samples i ALWAYS get 100% cpu usage...
then closed mainstage and opened NI Guitar Rig 3 as standalone, using the alesis as interface, with 64 samples as buffer, i cannot go over 13-14% of CPU usage (Hi quality enabled)... I know that mainstage adds more effects than the ones i selected in guitar rig 3, but the HUGE difference in cpu usage cannot be just that, i mean...
It could be if one of the effects Mainstage was using is Space Designer.
If it is using Space Designer try using a lesser reverb, like Gold.
USB doesn't use that much more CPU so that's not the problem.
You are comparing two separate guitar setups, correct? Or do you mean you're opening Guitar Rig 3 in Mainstage?
pancenter-
As I understand it you are opening completely different setups, correct. -
High CPU load with Library Cache Pin wit event.
Oracle8i Enterprise Edition Release 8.1.6.0.0, 64 bit - Production
HP-UX B.11.00 U 9000/800 605309363 unlimited-user license
Currently CPU load is 100% with very less sessions in the database.
But i am finding Library Cache Pin wait event in top.
If i query v$latchholder i am not getting any rows..and in v$lacth_children
SQL> select SID,EVENT,WAIT_TIME,SECONDS_IN_WAIT,STATE from v$session_wait where EVENT='library cache pin';
SID EVENT WAIT_TIME SECONDS_IN_WAIT STATE
8 library cache pin 0 863 WAITING
90 library cache pin 0 3093 WAITING
89 library cache pin 0 2566 WAITING
57 library cache pin 0 27384 WAITING
54 library cache pin 0 21029 WAITING
53 library cache pin 0 7840 WAITING
50 library cache pin 0 2620 WAITING
47 library cache pin 0 6031 WAITING
46 library cache pin 0 1241 WAITING
41 library cache pin 0 15637 WAITING
145 library cache pin 0 910 WAITING
133 library cache pin 0 5124 WAITING
111 library cache pin 0 15077 WAITING
98 library cache pin 0 6563 WAITING
94 library cache pin 0 2088 WAITING
93 library cache pin 0 4592 WAITING
92 library cache pin 0 14705 WAITING
91 library cache pin 0 14798 WAITING
36 library cache pin 0 1533 WAITING
33 library cache pin 0 1491 WAITING
28 library cache pin 0 13970 WAITING
25 library cache pin 0 7630 WAITING
11 library cache pin 0 12169 WAITING
12 library cache pin 0 2352 WAITING
14 library cache pin 0 19748 WAITING
SQL> select addr,name from v$latch where name='library cache';
ADDR NAME
C00000006971D460 library cache
SQL> select latch#,name from v$latch_children where name='library cache';
LATCH# NAME
105 library cache
105 library cache
105 library cache
105 library cache
105 library cachePlease help me out to find actual cause of these latches and fix ?
-Yasserbump
No-one?
The problem seems to have 'resolved itself' over the weekend incidentally (although I don't believe anything truly resolves itself in this manner - something must have changed). -
High CPU load with no proof from processes
I have a problem with my CPU load monitor.
In Activity Monitor, the processes that are showing do not reflect my actual CPU load.
At any given time the highest process shows 3% cpu load, but the percentages under the cpu tab show that the cpu is running at 50% (basically one full core processing SOMETHING).
Now I need to figure out what this "Something" is and kill it?
Anyone have any suggestions on how to find what bogging my system down?hey, i know someone linked the other post i created, however ill give you the rundown quickly with out my babble (i was hyper at the time :p)
Go into your sharing control panel, services, turn OFF windows sharing. It seems for some reason teh netbios daemon is loading over and over again, most likely due to an error. it is restarting a few times a second creating the loads, in my acse atleast.
let me know if it works for you, if so im going to contact apple just to inform them since the tech I spoke to had no idea what it was.
good luck! -
Hi, i am currently using macbook pro and i am not sure why when i run skype the cpu is eat up to 90% and the fan start to make noise. is that any way to fix this issue. thanks alot. P.S i use mac os 10.7
Yes I was trying to pinpoint the action that causes Skype to go 100% When it does it uses two cores and the battery is drained increadibly fast.
I will let you know if I find out. Otherwise restarting and/or removing the $HOME/Library/Skype data does the trick of fixing it. -
X-Fi + 2 GFX-Cards = high CPU-Load, slow Mouse, hanging system
>X-Fi + 2 GFX-Cards = high CPU-Load, slow Mouse, hanging systemA Hey guys,
I'm running 2 GF 8800 GTs (PCIe .) cards on a MSI Eclipse SLI board. My Audigy2 ZS PCI runned fine in this system.
I just replaced the Audigy by a PCIe X-Fi (EMU20K2 audio processor - not the xtreme audio from MSI).
Now I'm dealing with the following issue:
- System startup is slow
- in Windows, Mouse is slow, stuttering
- high CPU-Load
- very slow system response
!!! Attention !!! I had this problems allready at the st boot with the X-Fi, when the driver from creative wasn't installed (the card was recognized as "HD Audio" bei Windows 7). Installing the drivers didn't help.
If I install only one graphics card, the system runns fine and I can use the X-Fi and it's Software without problems.
If I replace the X-Fi by the PCIe Xtreme Audio from MSI, System runns fine.
If I use Win XP with the two GFX cards and the X-Fi, System runns fine.
Can you help me?
Regards from snowy Germany
User....Hi there,
got the same problem if i use my two Nvidia GTX260.
My system:
Asus P6T
I7 920 @ 3,8GHz
2GB ram
Gigabyte GTX260 OC Edition
Nvidia GTX260
X-Fi XtremeGamer
-System (Win7x64) takes ~0min to start.
-Stuttering mouse, keyboard everything >.<
Deactivating the onboard sound doesnt help.
Newest Nvidia and Creative driver too =(
The system runs fine without the X-FI and two GTX260 no errors ect...
One GTX260 runs also fine with the X-FI
But two graphic cards and the X-FI..... **bleep** =/
I use debian as my main OS and it run still fine with both graphic cards and the x-fi...
so i think its just a driver problem.
Hope it will be fixed soon =/
If someone have any sugestions...please help us.
Greetings from Germany also =D
Syro -
VI causes high CPU load after several days of running
Hello,
I have a strange problem. I have craeted a test bench control VI and it runs on five PCs, controling five test benches. All PCs have almost the same hardware, same motherboard, same CPU, etc.
Normaly the CPU load is around 50% when I run the VI, but at one PC the CPU load climbs up to 85% after several days (maybe one week) of continuous running and the PC is almost tied up - the VI execution becomes very slow (10 times slower) and the mouse only moves very slowly.
First I thought that the PC has a problem but the problem dissapears as soon as I stop and restart the VI!
I don't even have to terminate LabVIEW - only my VI and then everything is okay again.
The problem does not seem to occur at the other four PCs, but I can not tell for sure.
The task manager says only LabVIEW causes the CPU load - swap file utilisation is little.
Is there a way to analyze what causes the high CPU load?
I Use LabVIEW 7.1 professional
Operating System: Win XP (SP 2 or 3 I don't remember)
P.S. I have no loops without wait-functions.
Greetings Johannes
Using LabVIEW 7.1 and 2009 recently
Solved!
Go to Solution.Do all PCs have the same amount of RAM?
What is the memory use over time as seen by the windows taks manager. Does it increase gradually?
An increase in resource usage if often cause by sloppy programming.
Do you grow arrays forever or constantly resize arrays in memory? (built array, insert into array, delete from array)
Do you often create new references and never close them?
What does the program actually do? In my opinion, a constant 50% CPU load seems too high for my taste. It means it would not even be able to keep up on slightly older hardware.
LabVIEW Champion . Do more with less code and in less time . -
Hello!
On C3825 with IOS 15.1(4)M5, I faced with issue:
I see high cpu load by Crypto IKMP proccess:
CPU utilization for five seconds: 92%/11%; one minute: 87%; five minutes: 86% PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process 295 96696680 625524 154590 76.42% 70.87% 71.46% 0 Crypto IKMP 65 7719140 608642 12682 1.08% 1.32% 1.33% 0 Per-Second Jobs 6 838308 93487 8967 1.00% 0.16% 0.12% 0 Check heaps 139 1534084 6029064 254 0.75% 0.69% 0.62% 0 IP Input
But we haven`t a lot of traffic ~10 Mbit/s.
We have 1 GRE tunnel to branch, and crypto map for remote clients and some ASA`s
I didn`t find any information how I can tshoot this situation.
How I can deal with this? Why CPU load so high?
Router1#show crypto engine accelerator statistic
Device: Onboard VPNLocation: Onboard: 0 :Statistics for encryption device since the last clear of counters 1325 seconds ago 370618 packets in 370618 packets out 78673429 bytes in 77224979 bytes out 279 paks/sec in 279 paks/sec out 474 Kbits/sec in 466 Kbits/sec out 171515 packets decrypted 199103 packets encrypted 24579216 bytes before decrypt 52645763 bytes encrypted 15116445 bytes decrypted 63556984 bytes after encrypt 0 packets decompressed 0 packets compressed 0 bytes before decomp 0 bytes before comp 0 bytes after decomp 0 bytes after comp 0 packets bypass decompr 0 packets bypass compres 0 bytes bypass decompres 0 bytes bypass compressi 0 packets not decompress 0 packets not compressed 0 bytes not decompressed 0 bytes not compressed 1.0:1 compression ratio 1.0:1 overall Last 5 minutes: 69175 packets in 69175 packets out 230 paks/sec in 230 paks/sec out 380964 bits/sec in 376846 bits/sec out 2652816 bytes decrypted 9459770 bytes encrypted 71697 Kbits/sec decrypted 255669 Kbits/sec encrypted 1.0:1 compression ratio 1.0:1 overall Errors: 0 pkts dropped 0 ppq full 0 tx parts overflow 0 rx parts overflow 0 replenishment failure 0 zero len 0 flow inputs bad 0 cmd invalid 0 IPV4 len 0 IPV6 len 0 algor invalid 0 bad shadow particle 0 algor disabled 0 pre tx fail 0 dma error 0 dbit miss 0 pipeline abort 0 failsafe timeout 0 reserv 0 bad sz count 0 bad shdw 0 bad flow tx 0 spi mismatch 0 bad flow rx 0 auth fail 0 udm fs fail 0 pad fail 0 addr limit fixup fail 0 seq fail 0 quad fix sp 0 quad fix mp 0 quad fix cont
Thanks!#show crypto isa sa
IPv4 Crypto ISAKMP SA
dst src state conn-id status
109.x.x.x 79.105.x.x QM_IDLE 1196 ACTIVE L2L
109.x.x.x 77.41.x.x QM_IDLE 1157 ACTIVE VPNClient
109.x.x.x 85.88.x.x QM_IDLE 1162 ACTIVE L2L
109.x.x.x 2.61.x.x QM_IDLE 1179 ACTIVE L2L
109.x.x.x 178.212.x.x QM_IDLE 1178 ACTIVE L2L
109.x.x.x 178.210.x.x QM_IDLE 1191 ACTIVE VPNClient
109.x.x.x 178.210.x.x QM_IDLE 1186 ACTIVE VPNClient
109.x.x.x 178.210.x.x QM_IDLE 1182 ACTIVE VPNClient
109.x.x.x 178.210.x.x QM_IDLE 1180 ACTIVE VPNClient
109.x.x.x 178.210.x.x QM_IDLE 1177 ACTIVE VPNClient
109.x.x.x 194.186.x.x QM_IDLE 1187 ACTIVE L2L
109.x.x.x 62.32.x.x QM_IDLE 1146 ACTIVE L2LIPv6 Crypto ISAKMP SARouter1#show crypto isa stats
ISAKMP Process Packet Stats
IKE Received Packets.......276181
IKE Transmit Packets.......278526
IKE Int Q Depth [0]........0
IKE Int Q Peak [0].........0
IKE Int Q Depth [1]........0
IKE Int Q Peak [1].........0
IKE Int Q Depth [2]........0
IKE Int Q Peak [2].........1
IKE Int Q Depth [3]........0
IKE Int Q Peak [3].........12
IKE Int Q Depth [4]........0
IKE Int Q Peak [4].........0
IKE IPC Q Depth............0
IKE IPC Q Peak.............1
IKE P1 Retransmitted.......13
IKE P2 Retransmitted.......143
IKE P1 Rcvd Retransmit.....12
IKE P2 Rcvd Retransmit.....11
IKE Dup Retransmit.........0
Pak too long in queue......0
Packets too long in queues
dropped by IKE Dispatcher..0
NAT Keepalives Received....16
IKE call reenqueue 0
IPSec Node Dead Reasons:
(errored reason mark with *)
No reason ...0
*By Error ...0
By User Command ...0
By Expired Lifetime ...0
No Error ...197
Informational (in) state 1 ...273268
Informational (in) state 2 ...0
Done with xauth request/reply exchange ...117
Transaction mode done ...81
Saved QM no longer needed ...2
IKMP_NO_ERR_NO_TRANS ...0
P2 Re-tx timer expired (CONF_ADDR) ...0
Config mode cleanup ...0
QM done ...0
QM done (commit) ...0
QM done (await) ...944
IKE deleted ...22
*Delete Larval ...1
*Phase 2 err count exceeded ...11
*Decrypt_payload failed ...0
*Invalid payload ...0
*No IV for Transaction ...0
*DELAYED_QM_TIMER expired ...0
*QM no hash ...0
*QM bad hash ...0
*QM not authenticated ...0
*QM rejected ...16
*QM not accepted ...0
*No ke payload ...0
* ...0
* ...0
* ...0
* ...0
* ...0
* ...0
* ...0
* ...0
* ...0
* ...0
* ...0
* ...0
*Invalid reason code ...0IKE SA Dead Reason:
No reason ...111
*By error ...9
BY user command ...132
BY expired lifetime ...0
No error ...0
Delete no delete ...0
P1 delete notify (in) ...0
VRF removed from profile ...0
Death by tree-walk ...0
End of ipsec tunnel ...0
IKE SA Lifetime Exceeded ...0
*Receive initial contact ...0
*P1 errcounter exceeded (PEERS_ALIVE_TIMER) ...18
*Needed xauth ...18
*XAUTH fail ...0
*Client cancel xauth prompt ...0
*XAUTH not complete 1 ...8
*XAUTH not complete 2 ...0
*Fail to allocate ip address ...0
*Failed to allocate a connection id ...0
*Phase1 SA policy proposal not accepted ...52
*Recevied fatal informational ...0
*SA err counter exceeded (info) ...0
*Death by retransmission P1 ...0
*Death by retransmission P2 ...0
*Death by retransmission throw ...0
*Encrypt failure ...0
*Delete_me flag/throw ...0
*IKMP_ERR_NO_RETRANS ...0
*gen_ipsec_isakmp_delete but doi isakmp ...0
*QM_TIMER expired ...0
*IKE Fragmentation Failure ...0
* ...0
*Invalid reason code ...0 -
SpamAssassin and too much CPU load
I have been testing my messaging server and SpamAssassin integration. Everything seems fine except too much CPU load coming from spamd (SpamAssassin's daemon). I am testing them on a UltraSparcII-i, 256MB, Solaris 10 box. When I loaded it with 4msg/min containing 50K ascii text each on the receiving side, the spamd process went up to 90% of CPU load, and then just died silently. My questions;
1) Is there any performance tuning that I should do to avoid that much CPU load? Or is this just normal?
2) I have heard of another anti-spam product called dspam. How can I integrate dspam with the Java Messaging Server?
Thanks!Installation of JES Messaging Server by itself suggests 1.2 gig of RAM. SpamAssassin adds to that suggestion.
You have less than 1/4 of the suggested RAM. You're 'way short.
SpamAssassin will use cpu. I found that nx x86 box was far less expensive to feed, and works extremely well for Messaging and SpamAssassin. 1 gig or ram for my PC costs about $75. Adding ram to your Sparc will cost mor than a PC complete.
4 messages/min isn't much, but. . .
You're likely swapping pretty hard. Check for "I/O wait". I often use "top" for a quick look at such.
When I switched from an "ultra-5" with 384 megs ram, to an AMD 1200+ with 1 gig of ram, my system went from feeling totally buried to feeling mostly un-used. I'm currently able to handle far more than 4 message/minute, with no overload or swapping. -
DMVPN phase 3 - scalability - nhrp generates high cpu load
Hey all.
Been running into a scalability issues with DMVPN. Mainly caused (as I see it) by NHRP.
Scenario:
IOS-SLB-based DMVPN solution in a dual-cloud setup. Practically it's 2 separate solutions with spokes having 2 tunnels (one in each cloud). See attachment sketch. We're running a phase 3 hierarchy design (trying at least)
Spoke routers:
- 2500 routers in a mixture of c871, c881, c2800, c2900. Need to scale to at least twice that.
- Spoke-to-spoke is heavily used
Farm routers:
- Cisco 7201 with VAM2+. Around 1 router per 350 spokes (+1 for secondary tunnel)
Superhub:
- ASR 1004 (one for primary and one for secondary dmvpn-cloud).
We're not running any IPSEC between the farms and the superhubs. Just regular unencrypted DMVPN (mGRE).
Problem:
- NHRP is causing high CPU load on the ASRs. With around 2000 spokes up and running on DMVPN the CPU is overloaded with NHRP traffic. We're talking like 60-70% load caused by the NHRP process alone!
We're using 'ip nhrp interest' on all the spokes - and farms. We're in need of the spoke-to-spoke functionality so we allowing LAN-segments of our customers but denying everything else.
Solutions?
1. Turning off all NHRP resolutions? Basically remove any directly spoke-spoke communications (denying everything on the interest list). We can't go there since a lot of our customers are in dire need for directly spoke-spoke connectivity (due to latency). Haven't tested that it will actually give the much needed scalable solution either (we're facing around 5000 spokes in the next 2-3 years).
2. Chopping the DMVPN solution up in lesser VPN-blocks. This will administratively be a nightmare.
3. ?
Will really appreciate if anyone have a input here. It's really hard finding anything about a LARGE scale phase3 design on the web. Everything I find seems to mix stuff from small-scale phase 2 and 3 - making it rather messy cooking reciept for a small breakfast while I need a 7 course perfect dinner
When will Cisco come with an updated design guide btw?
Thanks in advance!Thank you for your quick reply.
Our ASRs (rp1) are acting as BGP RR while the farm routers are setup as RR clients.
We haven't tried connecting spokes directly to the ASRs but we have seen the same symptoms on the 7613s (sup720) and the 7200-platform.
Earlier the 7600 had the same role the ASRs have today. We were expecting that the ASRs should be doing "a better job" in terms of CPU load but we were wrong (NHRP generated around 10% more cpu load on the ASRs in comparison).
We concluded that the ASRs have a less optimized OS (coding) being rather new and all. Further we're not all happy about the stability of the platform (clear ip nhrp or taking a shutdown on the tunnel in the current situation will crash the router. 15.1(2)S1 and 15.1(3)S0a adv ip services). Haven't made a TAC case of it yet but will (has to be a bug as I see it since the 7200/7600 is handling this just fine).
Due to what I mentioned above I don't dare to debug the problem in production time and have to wait until the next scheduled maintenance window for some decent debug output (24. Oct).
We've contacted Cisco AS for assistance since it's hard to find local consultants (Norway) with enough knowledge of such scenario.
I just hope it's a config-issue and not a design issue, but we're willing to to whatever for this to scale to the thousands. -
Flash causes very high CPU load (including high Xorg load)
Hello everybody,
this is my first message. I'm an Arch user since a few months, and I'm pretty sure this is the distribution I'll stick to - I pratically love everything about it, especially its simpleness. It's the first Linux that doesn't scare me away with tons of configuration files scattered around dozens of different locations. I've already gotten quite a number of things fixed by myself without even looking at the forum or wiki.
Well, it seems most of these statements can be found in most first forum posts, so I'll stop to bore you...
The problem: Playing flash videos or animations causes incredible processor load: Even youtube takes a whole CPU core (or even a little more), top shows around 80% for Xorg and around 20-30% for Firefox (a little less with Opera). I'm on Arch64 with XFCE, Athlon X2 4600+, NVidia 8200 chipset (which has severe driver problems anyway; I'm using the latest NVidia beta driver). I'd try the open source nvidia driver, but (it seems that) there's no support for my chipset yet.
Currently I'm using Adobe's Flash64 alpha, before that I used various versions fo the nspluginwrapper. 32-bit Flash also caused high CPU load, but it wasn't that much and it may have been related to nspluginwrapper rather than anything else; however I have got the impression that things became even worse with the switch to the x64 alpha (performance-wise; on the other hand, it at least loads faster and doesn't crash anymore). Like said before, youtube takes one core, but it mostly works ok then, but videos with higher resolutions that youtube's are practically impossible to watch (except when there is very little movement in the film). Neither Flash hardware acceleration on nor off makes any difference.
I've already read half of the web, searched in forums, wikis, etc., but didn't find anybody with the same problem. When Flash64 was released and people were reporting good improvements, I was hoping it would be the same for me...
What I'm looking for is suggestions of things to try... I'm out of ideas what to try and where to look. It just seems strange that I seem to have problems that others don't...
Thanks for reading and for any comments.cerbie wrote:And, do I know you from somewhere--a little more Windows-centric--or are there many ElkMonsters, out here?
Hehe, I'm also wondering how many Cerbies are out there...
Really funny. The internet is a small place, obviously. Us meeting in a Linux forum also explains why nobody writes to the LSML anymore. -
HELP! Computer shutting down while loading FL studio Plugins! High CPU load?!
Good day everyone. I am in dire need of some help concerning my Hp Pavilion p6-2330 . I am a producier and I use a programmed called FL studio.
I've never had this problem before but now it happens almost everytime. When I open a project with a lot of VST plug-ins or a new VST plug-in it just immeadiatly prompts windows "shutting down". I am not sure if its exclusive to FL studio or a High CPU load in general.
I've tried a lot. Replaced my PSU. Blow out the dust from my hardive throuroughly. I really doubt its a overheating problem because it only happens when opening a project with a lot of vst plugins or opening a new vst on top. and my CPU isnt hot
Is there a crash report i can look at to see the reason its shutting down? What can i do?!Hi,
The following link shows specs of your computer:
http://h20564.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c03499435
given "only happens when opening a project with a lot of vst plugins or opening a new vst on top." I think the cheapest option now is to increase RAM from 6GB to say 16GB (using information from the above link). Before goinng out to buy more RAM, you may want to insrease system page file first (similar for Windows 8.1):
http://windows.microsoft.com/en-au/windows/change-virtual-memory-size#1TC=windows-7
or
http://www.softwareok.com/?seite=faq-Windows-8&faq=133
Regards.
BH
**Click the KUDOS thumb up on the left to say 'Thanks'**
Make it easier for other people to find solutions by marking a Reply 'Accept as Solution' if it solves your problem.
Maybe you are looking for
-
I can't play any purchased movies. Sounds like it is the 6.0 update. So what is the fix? Am I just stuck not being able to use my Apple TV until...?
-
Does copylocale still work correctly in flex 4 / 4.1 ?
I am trying to create a new en_GB locale for use with the SDK. After running bin/copylocale.exe en_US en_GB I observe the following in frameworks/locale/en_GB: 1) There are only 8 swcs in en_GB vs. 13 swcs in en_US 2) The file sizes of the en_GB swcs
-
Where to configure Account Assignment Category for WBS Settlement Rule?
Hi, I would like to know where is the Account Assignment Category for WBS settlement rule (FXA, CTR, G/L, WBS) being configured. (SPRO Path) Currently when I go to our settlement rule creation screen I can only see WBS and FXA, I need to add MAT (Ma
-
How do I copy clipping masks including relative positions for multiple images?
I have a number of microscope images of the same subject that were taken under three different conditions. I'd like to use clipping masks to crop the images in the same way (the same size but also the same positions relative to the corners/sides of t
-
QueueSubmitTimesheet for other users with impersonation? Project Server 2013
Hello. In an console application using TimeSheet.asmx on Project Server 2013 I want to call QueueSubmitTimesheet for timesheets that belong to other user. Is this possible? Thanks and best regards, Timo