I'm experiencing slow disk speeds on my server. Since I'm not too experienced in that area, I searched the forums and found a similar thread: Very slow SATA on Centos 5.5 Based on that thread I ran hdparm on all of my disks, with the following results:
[Code]....
In Short: I want something that will show me which files and associated requested URLs are causing the highest load on our web server.
Verbose: We already monitor our Apache web server using Cacti, Nagios, and of course traditional nix commands, but I am looking for something specifically to take an Apache process ID and tie that to a users request for a specific file and URL. The great thing about something like this is that it would be able to show me the worse performing PHP files. Special bonus if it can store this data over time for reports.
I have been googling around for something like this but can't seem to find it. Bonus if it exists in Cacti or Nagios already and I'm just to blind. I started writing my own PERL script to do this, but have limited time to devote to this at work so if a solution already exists I'm game, else I'll just have to write it myself. I'm also worried about my own scripted solution in that it won't get it write because it will be leveraging the output of a specialzed apache log that records the PID and then doing a look up with ps aux looking for that PID.
I have a user complaining about performance:-
Quote: At times on our server, during more-or-less idle moments, context switching is around the 2000-3000 mark.When we run the app, we can get this up to the 500,000 CSs per second. Not only this, but at the same time, the CPUs go to >60% system (the kernel is having to do all that context switching after all), leaving only around 30% for user space. We think it is a possibility that the massive context switching is contributing significantly to our performance issues.
The server is a HP DL580 G5 with 4 x Intel Xeon E7340 @ 2.4Ghz.My issue is that I don't know if the contect switching is a problem or not.From what i've read those values are not necesserilly cause for concern.
I just installed CentOS 5.4 on a brand new DL160 G6. In order to get the data raided I decided to go with mdraid. But I see quite harsh performance hits, I have no GUI or anything and just the basic server installation, and even the console feels sluggish.
The output from getinfo.sh disk:
==================== BEGIN uname -rmi ====================
2.6.18-164.15.1.el5 x86_64 x86_64
==================== END uname -rmi ====================
[code]....
I just wanted to know if having my laptop set to ondemand, will this affect performance in any way? I realize it increases the clock speed to performance when the CPU is under load, but does the time it take to go from ondemand to performance affect speed? Will there be any noticeable difference between the two setups? I have a dual core intel at 2.2GHz when in performance. When ondemand is set with no load it downclocks to 800Mhz.
View 7 Replies View Relatedok, we had to move one of our databases due to failing hardware. This box is newer than the old but it's just dog slow. 1st I thought it was mysql but now I am realizing it's the network. As a test, I tried copying a remote file to that box and the old server (located at the same co-lo, same provider), same switch, etc.Here are the results;
Old server:
access_log.1 100% 4192KB 1.4MB/s 00:03
New Server:
access_log.1 100% 4192KB 72.3KB/s 00:58
So now I see why people were complaining about webpage load times, etc. I can't figure out why the network latency. I looked around a bit, see below for some things, I thought mtu, nic speed, etc.
mii-tool shows;
eth0: 100 Mbit, full duplex, link ok
eth1: 100 Mbit, full duplex, link ok[code].....
Are there other things I can type to test or provide more feedback somehow to get more information.
I've just removed fedora 10 and installed CentOS 5.4 both x86_56 on my Lenovo ThinkCentre 9091-CTO workstation, and observed serious performance degradation while intensive I/O operations and huge load without significant CPU consumption in user mode with about 50% in I/o wait.For example if I do the "tar xjvf some_large_tar.bz" file and run vmstat i get the output below. This is the only cpu and io intensive process running. The load gets up to 6 on C2D CPU and the machine has unacceptable responsivness.
[root@f00 ~]# vmstat 5 10
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------
r b swpd free buff cache si so bi bo in cs us sy id wa st
[code]....
I am experiencing disk write performance issues and I cannot find the cause. I have LSI-9211-8i SAS 2 controller (latest firmware), Centos 5.5 latest x86_64 kernel (2.6.18-194.17.4.el5 #1 SMP with latest LSI driver v. 7.00 datet Jul 27) and Seagate Cheetah ST3600057SS drives. These drives have a std write performance (sustained) of > 200MB/s (and read as well); with Fedora core 13 (same machine), issuing a dd if=/dev/zero of=/dev/sdo bs=1024k count=16384 (16 GB direct device write), gets normally to 213 MB/s (repeated retries). On Centos 5.5 I am getting speeds around 110/113 MB/s.
iostat does not show anything specific (just 1.3 % wait, CPU 99.7 idle).
There are 14 drives: tried with several of them, same figures. Reads go around 200 MB/s.
Last I upgraded one of our cent 5 boxes which is running kvm with a mix of centos and windows VMs.After the update performance just tanked on the XP guest, but the others (including other windows VMs) are just fine.I checked with our good friend google and didn't really get any help there.We're hoping to NOT have to rebuild the XP guest, so I thought I'd check in here first.
This is CentOS release 5.5 (Final) with kernel
2.6.18-194.32.1.el5 #1 SMP (64bit)
I installed on a Intel SR2612UR Server with integrated LSI Expander and a Adaptec 5805Z controller a Raid 6 Array with 12x 2TB Drives (2 ext4 Partition: 16TB, 2TB), but I get very poor read speed depending on the blocksize that I set it vary between 130 MB/s and maximally 170 MB/s which is really poor for a Raid 6 Array of 12 drives, on a similar Dell Server I get above 650 MB/s. The Adaptec Controller has the latest Firmware installed.
View 1 Replies View RelatedI have a weird performance issue with a centos 5 running a nfs server and a rh8 client. I think the fact that it is rh8 client should be downplayed. It is just that with rh8 client the performance degradation seems more clear. See test details below OS in server is Centos 5 x86_64 kernel 2.6.18-92.1.22.el5
1Gb connection between machines File to test over NFS is a 1GB file. First of all I wanted to measure how the network alone performs while using NFS. So in the server side I run a "cat" command on the 1GB file to /dev/null. Please note that the disk read speed is about 98MBs. At this point the file system has the 1GB file cached in memory. In the client side a "cat" on the same file gives me a speed of about 113MBs. It seems then that the bottleneck in this instance is the network and it is very close to nominal speed. So the network performance is really good. (BTW I know that the server got that file from cache because a vmstat or iostat shows no disk activity.)
The second test is reading from disk with no caching involve. In the server I flushed the 1GB file from the memory. For instance by reading another 5GB file and I repeat the same thing as above in the client (a cat on the 1GB file). Now, the server has to go to disk.(vmstat or iostat shows the disk activity). However the performance, now, is about 20MBs, I was expecting something closer so 90MBs. (since the reading speed in the server in the first test showed 98MBs).
This second test was repeated for ext2, ext3, xfs with no significant differences. A similar test using a RH8 NFS server and client gets me close to 60MBs for a 1GB file not cache by the file system in the serverSince network speeds and disk read speeds are not the bottlenecks ... what or where is the limiting factor then?
I'm running 32 bit Ubuntu 10.04 with 2GB RAM. Is my processor (AMD Athlon 64 X2) running sub-optimally?Given that a larger addressing space will make no difference to me (right now) would there be a performance gain or any other advantage to switching to 64 bit Ubuntu.
View 5 Replies View RelatedIs there any program for GNU to calculate CPU's performace in marks (should support multi-core CPUs)? So it will be possible to compare performance of different CPUs.
View 4 Replies View RelatedAfter being forced to rebuild my computer (failed drive) I finally got current and installed F11 (was using F9). I was watching GKrellm last night and noticed something odd. The CPU frequency on my processor (Phenom 9600) was scaled to the slower speed and was rarely being pushed to full speed.
I verified that cpuspeed was in ondemand mode and ran some processor intense tasks (huge image loads, large file cats, large yum installs, etc...). Only when I was able to sustain near 100% load would the system throttle to the full clock speed. It would then drop back to the slower speed very quickly even though the process was still running. As a rough benchmark, I ran these tasks again in performance mode to see what impact this was having. Most of these tasks were taking as much as 40-50% longer to complete in ondemand mode.
Digging further, I found that the default up_threshold in F11 is set to 95%!! This is verified by cating /sys/devices/system/cpu/cpu0/cpufreq/ondemand/up_threshold. This means that the system would not throttle to full speed unless a 95% load was sustained for multiple samples (36 miliseconds each by default).
I overrode these settings in /etc/sysconfig/cpuspeed. I change up_threshold to 60 and down_threshold to 30. I am at work right now, so I can't benchmark this change until tonight. I guess I could have just set it to performance mode and left it, but I'd rather save the power when the machine is idle.
Does anyone have any thoughts on why 95% is the default and if there is any problem/benefit to changing this to a much lower value? It seems that anyone with a CPU running SpeedStep or CoolnQuiet would suffer the same severe performance impacts I saw with the default values.
I av just started out with fedora (fedora 15 64bit) and everything is dandy except i get less than an hour of battery life on my laptop. I used to get around 2 and half hours on ubuntu natty. could someone help me figure out whats going on?All hardware has been working well and i havent installed any proprietary drivers except the WLAN broadcom driver.
View 1 Replies View RelatedFirst of all, apologies if this is not the correct place to post this. I recently installed openSUSE 11.2 (Gnome, 32bit) on my newish PC. It is an Acer Aspire X3200 (AMD Phenom X4 Quad-Core, 1.8 GHz, 4096 MB, NVIDIA GeForce 8200 Integrated, can provide detailed specifications if needed). I've to say I am very disappointed with the results so far! Sorry to say this but I'm having a whole load of problems that I never had when I had Vista or with current Win 7. Here are the issues:
1. Slow network performance. I've disabled ipv6 but it is still the same. Not only when I use Firefox, it is also slow when I use console based programs. 2. Instability - Gnome system monitor crashed 6 (or more) out of 10 times I use it. System Monitor users more CPU resources than anything else! Also when I started the PC this morning, I only had the desktop wallpaper. No icons, no taskbar, no mouse movements or menus, nothing. I had to force shut down it after a while. I'm sure we all are familiar with this on XP but I've never seen something like that on Linux!
3. Annoying noise coming every time I move mouse (mainly when scrolling) and even if I'm not doing anything. I suspect it is coming from my HDD indicator as it always blink even when nothing is running. This can't be a hardware issue as I've never had this on Vista or Win 7 (it is virtually silent on Vista and XP but not on SuSE) This noise is driving me totally mad.
4. Slow performance in general. No matter what I do, even typing this on this text window, it seems everything is running so slow. I never had this problem when I was running 10.x on my 6 year old PC. CPU is always running around 20%-30%, when I don't run any other applications but just Firefox (I have Foxclocks, Delicious and FEBE as add-ons and I don't have any problem using them on Vista or Win 7). I am using Compiz, not sure if this has got anything to do with that.
I'm havng some seriously bad performance why i try to do any work on ubuntu.. big issue..i think it's in ubuntu 10.04 and need's some serious work!here's my list.
[Code]...
anyone experiecing the same issues? runnin ubuntu 10.04 64bit fully updated on dual 2.4 9700m 720gigs 4gig ram... i think it's fast enough for ubuntu or simple browsing..
I am currently using the 32 bit version of Ubuntu on my 64 bit laptop. My question is, does using a 32 bit version on said hardware reduce its performance (i.e will battery life be reduced, system be under more load, etc.)?
View 4 Replies View RelatedDoes slackware run slower on amd cpus compared to intel? I just read some articles on compilers optimized for intel cpus via the 'GenuineIntel' check. CPUs' not bearing intel cpuid will be run in non-optimized mode. It's part of Intel & AMD anti-trust settlement.
Just asking if someone could elaborate on the matter.
Sources :
Intel Forced to Remove "Cripple AMD" Function from Compiler?
I install KDM cause I couldn't run fglrx on GDM since i install the drivers of my GPU and the KDM ... I am suffering from poor system performance youtube videos mostly HD run very laggy and overall performance decreased ...
I run lxde ....
I have a Dell Inspiron 1525 laptop with a X3100 on board. The problem is the performance doesn't really fit my expectations. Compiz works fairly OK, but TuxRacer runs at 25 FPS, Glest runs just terribly slow and there is quite noticeable tearing in videos (MPlayer and Totem). Not to mention 400-600 FPS in glxgears.So far I have only tampered with Xorg (see below, changes marked red).
Code:
Section "ServerLayout"
Identifier "X.org Configured"
[code]....
Recently I noticed there are on sale, from several manufacturers, some interesting and inexpensive laptops equipped with AMD's new Ontario C50 processor which for one thing seems to use little power (like the Atom), and the other and more interesting is the fact that it has the graphics embedded.
I am curious if someone here owns one of said laptops and can comment about the performance/compatibility of this Ontario chip, because some of these laptops are such good deals that I am seriously considering at least recommending them. Obviously if I purchase one of those I'd install Fedora on in. Two things puzzle me: is performance very far apart from that of a netbook, and second, how do the graphics perform? Do they use the Catalyst driver as well?
i wanted to check the cpu performance monitor also, can i use readyboost in opensuse? I wanted to try that out as well
View 7 Replies View RelatedI'd like to find out if my system can take a certain load, number of PHP requests per unit of time. How would one go about this?
I'm about to make a change on Digital Camera Buying Guide, Photography Articles and Reviews | NeoCamera.com that would make all pages pass through PHP with personalized information which will prevent me from setting the cache-validity any significant amount of time (currently it is 1 hour). To handle the current load it needs to process 500 PHP requests per minute but I would like to make sure my system can handle twice the load. It is a Phenom II X4 with 8GB RAM running OpenSUSE 11.2.
How can I check my video performance?
View 4 Replies View Relatedanybody knows any tool to measure cpu performance on opensuse box, graphically would be great.
View 5 Replies View RelatedWhat is an smart setup of an ocz-ssd 120GB, so that mathematica 8 speeds up?
OS: OpenSuse 11.3 64Bit (Dualboot with XP 64)
RAM: 8 GB
AMD X4 3.4 GHrz
HHD: 2 TB
SSD: 120 GB
App: Mathematica 8 and Databases
GPU: 36 Shader 512 MB
What partitions are useful?
SWAP: 60GB
TMP: 20GB
APP: 20GB
DB-Indices: 20GB
Which file-format performs the best? Do I need any OCZ specific driver for Opensuse 11.3?
How can use the Swap and tmp Partion also to spee up XP?
I have a large HD (1 TBytes) on my Server and I need to install 11 SP1. I want to use this server for software development (Web apps with Oracle DB and Oracle Forms). Can You suggest how to partition my HD for best use and performance ?
View 2 Replies View RelatedI'm looking for a tool for Debian Linux that can measure the synchronisation performance between two desktops.
I'm using RADClock instead of the NTP client to synchronise my desktops.
Modern systems seem to be EFI based, and while there's a specific version of lilo that works with it, i can't seem to find any mention of a efi varient of grub (edit - on modern debian derivative systems there's grub-efi. Most references to it seem to be about various bugs in/for it tho) . In general, how well is EFI supported on linux?In addition, apparently windows 7 uses EFI to enable faster boot - 12 seconds or so with some systems. While a properly tweaked linux system could probably come close, are there any enhancements to the linux booting procedure that take advantage of EFI?
View 1 Replies View Related