Search Results

Search found 19055 results on 763 pages for 'high performance'.

Page 108/763 | < Previous Page | 104 105 106 107 108 109 110 111 112 113 114 115  | Next Page >

  • Is there an objective way to measure slowness of PC/WINDOWS?

    - by ekms
    We've a lot of users that usually complain about that his PC is "slow". (we use win XP). We usually check startup programs, virus, fragmentation, disk health and common problems that causes slowness (Symantec AV drops disk to 1mb/s , or a seagate HD firmware error in certain models), but in those cases the slowness is pretty evident. In other hand, the most common is the user complaining about his pc but for us looks OK, even in 6 years old desktops. People sometimes even complains about his new quad core desktops speed!!! So, we are asking if there's a way to OBJECTIVELY check that a computer didn't dropped its performance, compared with similar ones o previous measures, specially for work use (I don't think that 3dmark benchmark o similar may help). The only thing that I found that was useful is HDTune, but it only check hard disk performance. Basically, what we want is something that enable us to say to our users "see? your PC is as slow as was three years ago! stop complaining! Is all in your head!"

    Read the article

  • Disk fragmentation when dealing with many small files

    - by Zorlack
    On a daily basis we generate about 3.4 Million small jpeg files. We also delete about 3.4 Million 90 day old images. To date, we've dealt with this content by storing the images in a hierarchical manner. The heriarchy is something like this: /Year/Month/Day/Source/ This heirarchy allows us to effectively delete days worth of content across all sources. The files are stored on a Windows 2003 server connected to a 14 disk SATA RAID6. We've started having significant performance issues when writing-to and reading-from the disks. This may be due to the performance of the hardware, but I suspect that disk fragmentation may be a culprit at well. Some people have recommended storing the data in a database, but I've been hesitant to do this. An other thought was to use some sort of container file, like a VHD or something. Does anyone have any advice for mitigating this kind of fragmentation? Additional Info: The average file size is 8-14KB Format information from fsutil: NTFS Volume Serial Number : 0x2ae2ea00e2e9d05d Version : 3.1 Number Sectors : 0x00000001e847ffff Total Clusters : 0x000000003d08ffff Free Clusters : 0x000000001c1a4df0 Total Reserved : 0x0000000000000000 Bytes Per Sector : 512 Bytes Per Cluster : 4096 Bytes Per FileRecord Segment : 1024 Clusters Per FileRecord Segment : 0 Mft Valid Data Length : 0x000000208f020000 Mft Start Lcn : 0x00000000000c0000 Mft2 Start Lcn : 0x000000001e847fff Mft Zone Start : 0x0000000002163b20 Mft Zone End : 0x0000000007ad2000

    Read the article

  • Windows XP to remote server 2008 R2 shares - awful response times

    - by nick3216
    I have a network infrastructure of Windows XP clients (a mix of XP and 64-bit XP), that are accessing a network share on a Windows 2008 R2 server. Whenever users type the address of a folder into the address bar of Windows Explorer it's as snappy at determining the contents of the current folder and presenting them to you in the address bar as if you're working on a local drive. But if you open one of the subfolders users get the animated red torch and 'Searching for items...' dialog, typically for 45 seconds. Similarly when using the open folder dialog to try and select a subfolder on this share it takes, on average, 45 seconds for the dialog to expand each node and show the subfolders of each node. Also, while the Explorer instance accsesing the network share is running slowly users notice that the performance of all other Explorer windows suffers. So while Explorer is searching for files on the network share they can't switch to another task and navigate around their local drive using Explorer because it's now as slow as a dead dog at accessing anything. Are there any settings we can change which will improve the performance accessing network shares?

    Read the article

  • iptable CLUSTERIP won't work

    - by Rad Akefirad
    We have some requirements which explained here. We tried to satisfy them without any success as described. Here is the brief information: Here are requirements: 1. High Availability 2. Load Balancing Current Configuration: Server #1: one static (real) IP for each 10.17.243.11 Server #2: one static (real) IP for each 10.17.243.12 Cluster (virtual and shared among all servers) IP: 10.17.243.15 I tried to use CLUSTERIP to have the cluster IP by the following: on the server #1 iptables -I INPUT -i eth0 -d 10.17.243.15 -j CLUSTERIP --new --hashmode sourceip --clustermac 01:00:5E:00:00:20 --total-nodes 2 --local-node 1 on the server #2 iptables -I INPUT -i eth0 -d 10.17.243.15 -j CLUSTERIP --new --hashmode sourceip --clustermac 01:00:5E:00:00:20 --total-nodes 2 --local-node 2 When we try to ping 10.17.243.15 there is no reply. And the web service (tomcat on port 8080) is not accessible either. However we managed to get the packets on both servers by using TCPDUMP. Some useful information: iptable roules (iptables -L -n -v): Chain INPUT (policy ACCEPT 21775 packets, 1470K bytes) pkts bytes target prot opt in out source destination 0 0 CLUSTERIP all -- eth0 * 0.0.0.0/0 10.17.243.15 CLUSTERIP hashmode=sourceip clustermac=01:00:5E:00:00:20 total_nodes=2 local_node=1 hash_init=0 Chain FORWARD (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination Chain OUTPUT (policy ACCEPT 14078 packets, 44M bytes) pkts bytes target prot opt in out source destination Log messages: ... kernel: [ 7.329017] e1000e: eth3 NIC Link is Up 100 Mbps Full Duplex, Flow Control: None ... kernel: [ 7.329133] e1000e 0000:05:00.0: eth3: 10/100 speed: disabling TSO ... kernel: [ 7.329567] ADDRCONF(NETDEV_CHANGE): eth3: link becomes ready ... kernel: [ 71.333285] ip_tables: (C) 2000-2006 Netfilter Core Team ... kernel: [ 71.341804] nf_conntrack version 0.5.0 (16384 buckets, 65536 max) ... kernel: [ 71.343168] ipt_CLUSTERIP: ClusterIP Version 0.8 loaded successfully ... kernel: [ 108.456043] device eth0 entered promiscuous mode ... kernel: [ 112.678859] device eth0 left promiscuous mode ... kernel: [ 117.916050] device eth0 entered promiscuous mode ... kernel: [ 140.168848] device eth0 left promiscuous mode TCPDUMP while pinging: tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 65535 bytes 12:11:55.335528 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto ICMP (1), length 84) 10.17.243.1 > 10.17.243.15: ICMP echo request, id 16162, seq 2390, length 64 12:11:56.335778 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto ICMP (1), length 84) 10.17.243.1 > 10.17.243.15: ICMP echo request, id 16162, seq 2391, length 64 12:11:57.336010 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto ICMP (1), length 84) 10.17.243.1 > 10.17.243.15: ICMP echo request, id 16162, seq 2392, length 64 12:11:58.336287 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto ICMP (1), length 84) 10.17.243.1 > 10.17.243.15: ICMP echo request, id 16162, seq 2393, length 64 And there is no ping reply as I said. Does anyone know which part I missed? Thanks in advance.

    Read the article

  • Multiple data centers and HTTP traffic: DNS Round Robin is the ONLY way to assure instant fail-over?

    - by vmiazzo
    Hi, Multiple A records pointing to the same domain seem to be used almost exclusively to implement DNS Round Robin as a cheap load balancing technique. The usual warning against DNS RR is that it is not good for high availability. When 1 IP goes down clients will continue to use it for minutes. A load balancer is often suggested as a better choice. Both claims are not completely true: When the traffic is HTTP then, most of the HTML browsers are able to automatically try the next A record if the previous is down, without a new DNS look-up. Read here chapter 3.1 and here. When multiple data centers are involved then, DNS RR is the only option to distribute traffic across them. So, is it true that, with multiple data centers and HTTP traffic, the use of DNS RR is the ONLY way to assure instant fail-over when one data center goes down? Thanks, Valentino Edit: Off course each data center has a local Load Balancer with hot spare. It's OK to sacrifice session affinity for an instant fail-over. AFAIK the only way for a DNS to suggest a data center instead of another is to reply with just the IP (or IPs) associated to that data center. If the data center becomes unreachable then all those IP are also unreachables. This means that, even if smart HTML browsers are able to instantly try another A record , all the attempts will fail until the local cache entry expires and a new DNS lookup is done, fetching the new working IPs (I assume DNS automatically suggests to a new data center when one fail). So, "smart DNS" cannot assure instant fail-over. Conversely a DNS round-robin permits it. When one data center fail, the smart HTML browsers (most of them) instantly try the other cached A records jumping to another (working) data center. So, DNS round-robin doesn't assure session affinity or the lowest RTT but seems to be the only way to assure instant fail-over when the clients are "smart" HTML browsers. Edit 2: Some people suggest TCP Anycast as a definitive solution. In this paper (chapter 6) is explained that Anycast fail-over is related to BGP convergence. For this reason Anycast can employ from 15 minutes to 20 seconds to complete. 20 seconds are possible on networks where the topology was optimized for this. Probably just CDN operators can grant such fast fail-overs. Edit 3:* I did some DNS look-ups and traceroutes (maybe some expert can double check) and: The only CDN using TCP Anycast seems to be CacheFly, other operators like CDN networks and BitGravity use CacheFly. Seems that their edges cannot be used as reverse proxies. Therefore, they cannot be used to grant instant failover. Akamai and LimeLight seems to use geo-aware DNS. But! They return multiple A records. From traceroutes seems that the returned IPs are on the same data center. So, I'm puzzled on how they can offer a 100% SLA when one data center goes down.

    Read the article

  • How to force two process to run on the same CPU?

    - by kovan
    Context: I'm programming a software system that consists of multiple processes. It is programmed in C++ under Linux. and they communicate among them using Linux shared memory. Usually, in software development, is in the final stage when the performance optimization is made. Here I came to a big problem. The software has high performance requirements, but in machines with 4 or 8 CPU cores (usually with more than one CPU), it was only able to use 3 cores, thus wasting 25% of the CPU power in the first ones, and more than 60% in the second ones. After many research, and having discarded mutex and lock contention, I found out that the time was being wasted on shmdt/shmat calls (detach and attach to shared memory segments). After some more research, I found out that these CPUs, which usually are AMD Opteron and Intel Xeon, use a memory system called NUMA, which basically means that each processor has its fast, "local memory", and accessing memory from other CPUs is expensive. After doing some tests, the problem seems to be that the software is designed so that, basically, any process can pass shared memory segments to any other process, and to any thread in them. This seems to kill performance, as process are constantly accessing memory from other processes. Question: Now, the question is, is there any way to force pairs of processes to execute in the same CPU?. I don't mean to force them to execute always in the same processor, as I don't care in which one they are executed, altough that would do the job. Ideally, there would be a way to tell the kernel: If you schedule this process in one processor, you must also schedule this "brother" process (which is the process with which it communicates through shared memory) in that same processor, so that performance is not penalized.

    Read the article

  • Alternative or succesor to GDBM

    - by Anon Guy
    We a have a GDBM key-value database as the backend to a load-balanced web-facing application that is in implemented in C++. The data served by the application has grown very large, so our admins have moved the GDBM files from "local" storage (on the webservers, or very close by) to a large, shared, remote, NFS-mounted filesystem. This has affected performance. Our performance tests (in a test environment) show page load times jumping from hundreds of milliseconds (for local disk) to several seconds (over NFS, local network), and sometimes getting as high as 30 seconds. I believe a large part of the problem is that the application makes lots of random reads from the GDBM files, and that these are slow over NFS, and this will be even worse in production (where the front-end and back-end have even more network hardware between them) and as our database gets even bigger. While this is not a critical application, I would like to improve performance, and have some resources available, including the application developer time and Unix admins. My main constraint is time only have the resources for a few weeks. As I see it, my options are: Improve NFS performance by tuning parameters. My instinct is we wont get much out of this, but I have been wrong before, and I don't really know very much about NFS tuning. Move to a different key-value database, such as memcachedb or Tokyo Cabinet. Replace NFS with some other protocol (iSCSI has been mentioned, but i am not familiar with it). How should I approach this problem?

    Read the article

  • Refactoring or Rewriting Monolithic PHP Spaghetti Codebase

    - by nategood
    I've inherited a really poorly designed PHP spaghetti code project. It's been gaining a good bit of traffic recently and is starting to have performance issues on top of the poor monolithic code base. Its maxing out performance on a chunky 16GB dedicated machine when it really shouldn't be. I'm planning on doing some performance tweaks right off the bat to help the performance issue, but this still won't really help the horrible code base. The team is small but expecting to grow very soon. I've read Joel's article on the troubles of doing a complete rewrite and see the concerns. But how bad does the code base have to be before you consider a rewrite? There is PHP handling logic interjected into what one would usually consider a "view". Even worse, in some places SQL statements are in these same files! The only real separation of presentation and logic are a few PHP scripts that serve as function libraries. These scripts do most of the ORM stuff... if you can even call it that. Trying to slowly refractor this seems like a nightmare. Open to your thoughts and opinions... however not interested in hearing, "Run away, Run away!".

    Read the article

  • How can a single disk in a hardware SATA RAID-10 array bring the entire array to a screeching halt?

    - by Stu Thompson
    Prelude: I'm a code-monkey that's increasingly taken on SysAdmin duties for my small company. My code is our product, and increasingly we provide the same app as SaaS. About 18 months ago I moved our servers from a premium hosting centric vendor to a barebones rack pusher in a tier IV data center. (Literally across the street.) This ment doing much more ourselves--things like networking, storage and monitoring. As part the big move, to replace our leased direct attached storage from the hosting company, I built a 9TB two-node NAS based on SuperMicro chassises, 3ware RAID cards, Ubuntu 10.04, two dozen SATA disks, DRBD and . It's all lovingly documented in three blog posts: Building up & testing a new 9TB SATA RAID10 NFSv4 NAS: Part I, Part II and Part III. We also setup a Cacit monitoring system. Recently we've been adding more and more data points, like SMART values. I could not have done all this without the awesome boffins at ServerFault. It's been a fun and educational experience. My boss is happy (we saved bucket loads of $$$), our customers are happy (storage costs are down), I'm happy (fun, fun, fun). Until yesterday. Outage & Recovery: Some time after lunch we started getting reports of sluggish performance from our application, an on-demand streaming media CMS. About the same time our Cacti monitoring system sent a blizzard of emails. One of the more telling alerts was a graph of iostat await. Performance became so degraded that Pingdom began sending "server down" notifications. The overall load was moderate, there was not traffic spike. After logging onto the application servers, NFS clients of the NAS, I confirmed that just about everything was experiencing highly intermittent and insanely long IO wait times. And once I hopped onto the primary NAS node itself, the same delays were evident when trying to navigate the problem array's file system. Time to fail over, that went well. Within 20 minuts everything was confirmed to be back up and running perfectly. Post-Mortem: After any and all system failures I perform a post-mortem to determine the cause of the failure. First thing I did was ssh back into the box and start reviewing logs. It was offline, completely. Time for a trip to the data center. Hardware reset, backup an and running. In /var/syslog I found this scary looking entry: Nov 15 06:49:44 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_00], 6 Currently unreadable (pending) sectors Nov 15 06:49:44 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_07], SMART Prefailure Attribute: 1 Raw_Read_Error_Rate changed from 171 to 170 Nov 15 06:49:45 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_10], 16 Currently unreadable (pending) sectors Nov 15 06:49:45 umbilo smartd[2827]: Device: /dev/twa0 [3ware_disk_10], 4 Offline uncorrectable sectors Nov 15 06:49:45 umbilo smartd[2827]: Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error Nov 15 06:49:45 umbilo smartd[2827]: # 1 Short offline Completed: read failure 90% 6576 3421766910 Nov 15 06:49:45 umbilo smartd[2827]: # 2 Short offline Completed: read failure 90% 6087 3421766910 Nov 15 06:49:45 umbilo smartd[2827]: # 3 Short offline Completed: read failure 10% 5901 656821791 Nov 15 06:49:45 umbilo smartd[2827]: # 4 Short offline Completed: read failure 90% 5818 651637856 Nov 15 06:49:45 umbilo smartd[2827]: So I went to check the Cacti graphs for the disks in the array. Here we see that, yes, disk 7 is slipping away just like syslog says it is. But we also see that disk 8's SMART Read Erros are fluctuating. There are no messages about disk 8 in syslog. More interesting is that the fluctuating values for disk 8 directly correlate to the high IO wait times! My interpretation is that: Disk 8 is experiencing an odd hardware fault that results in intermittent long operation times. Somehow this fault condition on the disk is locking up the entire array Maybe there is a more accurate or correct description, but the net result has been that the one disk is impacting the performance of the whole array. The Question(s) How can a single disk in a hardware SATA RAID-10 array bring the entire array to a screeching halt? Am I being naïve to think that the RAID card should have dealt with this? How can I prevent a single misbehaving disk from impacting the entire array? Am I missing something?

    Read the article

  • SAN with iSCSI-Target Performance Horrendous

    - by Justin
    We have a poor man's SAN setup in a 1U Ubuntu server running iSCSI-Target with two 300GB drives in RAID-0. We then are using it for block level storage for virtual machines. The hypervisor is connected to the SAN via gigabit on a dedicated VLAN and interfaces. We only have a single virtual machine setup and doing some benchmarks. If we run hdparm -t /dev/sda1 from the virtual machine, we get 'ok' performance of 75MB/s from the virtual machine to the SAN. Then we basically compile a package with ./configure and make. Things start ok, but then all the sudden the load average on the SAN grows to 7+ and things slow down to a crawl. When we SSH into the SAN and run top, sure the load is 7+, but the CPU usage is basically nothing, also the server has 1.5GB of memory available. When we kill the compile on the virtual machine, slowly the LOAD on the SAN goes back to sub 1 figures. What in the world is causing this? How can we diagnosis this further? Here are two screenshot from the SAN during high load. 1> Output of iotop on the SAN: 2> Output of top on the SAN:

    Read the article

  • X11 performance problem after upgrading from Centos3 to Centos5 with an ATI Rage XL

    - by Marcelo Santos
    After upgrading a computer from Centos3 to Centos5 an application that does a lot of scrolling took a very high performance hit. top tells me that X is using a lot of CPU and that was not happening before. The machine has an ATI Rage XL with 8MB and X is using the ati driver as there is no proprietary ATI driver for this board on linux. The xorg.conf: Section "Device" Identifier "Videocard0" Driver "ati" EndSection Section "Screen" Identifier "Screen0" Device "Videocard0" DefaultDepth 24 SubSection "Display" Viewport 0 0 Depth 24 Modes "1024x768" "800x600" "640x480" EndSubSection EndSection Section "DRI" Group 0 Mode 0666 EndSection A similar machine that still has Centos3 installed is able to start DRI on the X server while this one is not, this is the Xorg.0.log for the Centos5 machine: drmOpenDevice: node name is /dev/dri/card0 drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: Open failed drmOpenDevice: node name is /dev/dri/card0 drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: Open failed [drm] failed to load kernel module "mach64" (II) ATI(0): [drm] drmOpen failed (EE) ATI(0): [dri] DRIScreenInit Failed (II) ATI(0): Largest offscreen areas (with overlaps): (II) ATI(0): 1024 x 1279 rectangle at 0,768 (II) ATI(0): 768 x 1280 rectangle at 0,768 (II) ATI(0): Using XFree86 Acceleration Architecture (XAA) Screen to screen bit blits Solid filled rectangles 8x8 mono pattern filled rectangles Indirect CPU to Screen color expansion Solid Lines Offscreen Pixmaps Setting up tile and stipple cache: 32 128x128 slots 10 256x256 slots (==) ATI(0): Backing store disabled (==) ATI(0): Silken mouse enabled (II) ATI(0): Direct rendering disabled (==) RandR enabled I also tried using EXA instead of XAA and setting: Option "AccelMethod" "XAA" Option "XAANoOffscreenPixmaps" "true" uname -a Linux sir5.erg.inpe.br 2.6.18-128.7.1.el5 #1 SMP Mon Aug 24 08:20:55 EDT 2009 i686 i686 i386 GNU/Linux rpm -qa | grep xorg-x11-server xorg-x11-server-utils-7.1-4.fc6 xorg-x11-server-sdk-1.1.1-48.52.el5 xorg-x11-server-Xvfb-1.1.1-48.52.el5 xorg-x11-server-Xnest-1.1.1-48.52.el5 xorg-x11-server-Xorg-1.1.1-48.52.el5 The drmOpenDevice error continues when using the suggested Option "AIGLX" "true".

    Read the article

  • Bad Performance when SQL Server hits 99% Memory Usage

    - by user15863
    I've got a server that reports 8 GB of ram used up at 99%. When restart Sql Server, it drops down to about 5% usage, but gradually builds back up to 99% over about 2 hours. When I look at the sqlserver process, its reported as only using 100k ram, and generally never goes up or below that number by very much. In fact, if I add up all the processes in my TaskManager, it's barely scratching the surface of my total available (yet TaskManager still shows 99% memory usage with "All processes shown"). It appears that Sql Server has a huge memory leak going on but it's not reporting it. The server has ran fine for nearly two years, with this only starting to manifest itself in the last 3-4 weeks. Anyone seen this or have any insight into the problem? EDIT When the server hits 99%, performance goes down hill. All queries to the server, apps, etc. come to a crawl. Restarting the service makes things zippy again, until 2 hours has passed and the server hits 99% once again.

    Read the article

  • Performance of Virtual machines on very low end machines

    - by TheLQ
    I am managing a few cheap servers as my user base isn't large enough to get much more powerful servers. I also don't have the money lying around to invest in a server to prepare for the larger user base. So I'm stuck with the old hardware I have. I am toying with the idea of virtualizing all the current OS's with most likely VMware vSphere Hypervisor (AKA ESXi) Xen (ESXi has too strict of an HCL, and my hardware is too old). Big reasons for doing so: Ability to upgrade and scale hardware rapidly - This is most likely what I'll be doing as I distribute services, get a bigger server, centralize (electricity bills are horrible), distribute, get a bigger server, etc... Manually doing this by reinstalling the entire OS would be a big pain Safety from me - I've made many rookie mistakes, like doing lots of risky work on a vital production server. With a VM I can just backup the state, work on my machine, test, and revert if necessary. No worries, and no OS reinstallation Safety from other factors - As I scale servers might go down, and a backup VM can instantly be started. Various other reasons. However the limiting factor here is hardware. And I mean very depressing hardware. The current server's run off of a Pentium 3 and 4, and have 512 MB and 768 MB RAM respectively (RAM can be upgraded soon however). Is the Virtualization layer small enough to run itself and a Linux OS effectively? Will performance be acceptable (50% CPU overhead for every operation isn't acceptable)? Does it leave enough RAM for the Linux OS? Is this even feasible?

    Read the article

  • Hints on diagnosing performance issue in OpenBSD firewall

    - by Tom
    My OpenBSD 4.6 pf firewall has started having really bad performance in the past few weeks. I've isolated the firewall (as opposed to the WAN connection, switch, cable, etc.) as the problem, but need a hint on how to further diagnose or fix the problem. The facts: Normal setup is: DSL Modem - FW Ext. NIC - FW Int. NIC - Switch - Laptop Normal setup described above gives only 25 Kbps! Plugging the laptop straight from the DSL modem gives a 1 MBps connection (full speed, as advertised). Therefore, the DSL connection seems to be OK. Plugging the laptop directly into the firewall's internal NIC (bypassing the switch) also gives only 25 Kbps. Therefore, the switch does not seem to be a problem. I've replaced the ethernet cables, but it didn't help. Here's the weird thing. Reloading the ruleset (/sbin/pfctl -Fa -f /etc/pf.conf) causes the laptop's connection to go up to 1 Mbps (i.e. full speed) for a few minutes before it gradually degrades back down to 25Kbps again. Any ideas on what's wrong or how I could further diagnose the problem?

    Read the article

  • Nginx + uWSGI + Django performance stuck on 100rq/s

    - by dancio
    I have configured Nginx with uWSGI and Django on CentOS 6 x64 (3.06GHz i3 540, 4GB), which should easily handle 2500 rq/s but when I run ab test ( ab -n 1000 -c 100 ) performance stops at 92 - 100 rq/s. Nginx: user nginx; worker_processes 2; events { worker_connections 2048; use epoll; } uWSGI: Emperor /usr/sbin/uwsgi --master --no-orphans --pythonpath /var/python --emperor /var/python/*/uwsgi.ini [uwsgi] socket = 127.0.0.2:3031 master = true processes = 5 env = DJANGO_SETTINGS_MODULE=x.settings env = HTTPS=on module = django.core.handlers.wsgi:WSGIHandler() disable-logging = true catch-exceptions = false post-buffering = 8192 harakiri = 30 harakiri-verbose = true vacuum = true listen = 500 optimize = 2 sysclt changes: # Increase TCP max buffer size setable using setsockopt() net.ipv4.tcp_rmem = 4096 87380 8388608 net.ipv4.tcp_wmem = 4096 87380 8388608 net.core.rmem_max = 8388608 net.core.wmem_max = 8388608 net.core.netdev_max_backlog = 5000 net.ipv4.tcp_max_syn_backlog = 5000 net.ipv4.tcp_window_scaling = 1 net.core.somaxconn = 2048 # Avoid a smurf attack net.ipv4.icmp_echo_ignore_broadcasts = 1 # Optimization for port usefor LBs # Increase system file descriptor limit fs.file-max = 65535 I did sysctl -p to enable changes. Idle server info: top - 13:34:58 up 102 days, 18:35, 1 user, load average: 0.00, 0.00, 0.00 Tasks: 118 total, 1 running, 117 sleeping, 0 stopped, 0 zombie Cpu(s): 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 3983068k total, 2125088k used, 1857980k free, 262528k buffers Swap: 2104504k total, 0k used, 2104504k free, 606996k cached free -m total used free shared buffers cached Mem: 3889 2075 1814 0 256 592 -/+ buffers/cache: 1226 2663 Swap: 2055 0 2055 **During the test:** top - 13:45:21 up 102 days, 18:46, 1 user, load average: 3.73, 1.51, 0.58 Tasks: 122 total, 8 running, 114 sleeping, 0 stopped, 0 zombie Cpu(s): 93.5%us, 5.2%sy, 0.0%ni, 0.2%id, 0.0%wa, 0.1%hi, 1.1%si, 0.0%st Mem: 3983068k total, 2127564k used, 1855504k free, 262580k buffers Swap: 2104504k total, 0k used, 2104504k free, 608760k cached free -m total used free shared buffers cached Mem: 3889 2125 1763 0 256 595 -/+ buffers/cache: 1274 2615 Swap: 2055 0 2055 iotop 30141 be/4 nginx 0.00 B/s 7.78 K/s 0.00 % 0.00 % nginx: wo~er process Where is the bottleneck ? Or what am I doing wrong ?

    Read the article

  • X11 performance problem after upgrading from Centos3 to Centos5 with an ATI Rage XL

    - by Marcelo Santos
    After upgrading a computer from Centos3 to Centos5 an application that does a lot of scrolling took a very high performance hit. top tells me that X is using a lot of CPU and that was not happening before. The machine has an ATI Rage XL with 8MB and X is using the ati driver as there is no proprietary ATI driver for this board on linux. The xorg.conf: Section "Device" Identifier "Videocard0" Driver "ati" EndSection Section "Screen" Identifier "Screen0" Device "Videocard0" DefaultDepth 24 SubSection "Display" Viewport 0 0 Depth 24 Modes "1024x768" "800x600" "640x480" EndSubSection EndSection Section "DRI" Group 0 Mode 0666 EndSection A similar machine that still has Centos3 installed is able to start DRI on the X server while this one is not, this is the Xorg.0.log for the Centos5 machine: drmOpenDevice: node name is /dev/dri/card0 drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: Open failed drmOpenDevice: node name is /dev/dri/card0 drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: open result is -1, (No such device or address) drmOpenDevice: Open failed [drm] failed to load kernel module "mach64" (II) ATI(0): [drm] drmOpen failed (EE) ATI(0): [dri] DRIScreenInit Failed (II) ATI(0): Largest offscreen areas (with overlaps): (II) ATI(0): 1024 x 1279 rectangle at 0,768 (II) ATI(0): 768 x 1280 rectangle at 0,768 (II) ATI(0): Using XFree86 Acceleration Architecture (XAA) Screen to screen bit blits Solid filled rectangles 8x8 mono pattern filled rectangles Indirect CPU to Screen color expansion Solid Lines Offscreen Pixmaps Setting up tile and stipple cache: 32 128x128 slots 10 256x256 slots (==) ATI(0): Backing store disabled (==) ATI(0): Silken mouse enabled (II) ATI(0): Direct rendering disabled (==) RandR enabled I also tried using EXA instead of XAA and setting: Option "AccelMethod" "XAA" Option "XAANoOffscreenPixmaps" "true" uname -a Linux sir5.erg.inpe.br 2.6.18-128.7.1.el5 #1 SMP Mon Aug 24 08:20:55 EDT 2009 i686 i686 i386 GNU/Linux rpm -qa | grep xorg-x11-server xorg-x11-server-utils-7.1-4.fc6 xorg-x11-server-sdk-1.1.1-48.52.el5 xorg-x11-server-Xvfb-1.1.1-48.52.el5 xorg-x11-server-Xnest-1.1.1-48.52.el5 xorg-x11-server-Xorg-1.1.1-48.52.el5 The drmOpenDevice error continues when using the suggested Option "AIGLX" "true".

    Read the article

  • apache performance timing out

    - by Mike
    Im running a webserver where I'm hosting about 6-7 websites. Most of these websites get their content from MySQL which is hosted on the same server. Traffic average per day is about 500-600 unique visitors, about 150K hits per week. But for some reason sometimes websites send a timeout, OR sometimes websites dont load all images. I know that I should perhaps separate static content from dynamic content, but for now I think that's not a possibility. I would appreciate any suggestions on how could I improve the performance of apache, so it doesn't keep timing out. Server is running on Sempron LE 1300; 2.3GHz,512K Cache 2GB RAM 10Mbps/1Mbps Services: MySQL, ProFTPD, Apache. Private + Shared = RAM used Program ---------------------------------------------------- 1.2 MiB + 54.0 KiB = 1.2 MiB proftpd 4.1 MiB + 23.0 KiB = 4.1 MiB munin-node 20.8 MiB + 120.5 KiB = 20.9 MiB mysqld 47.3 MiB + 9.9 MiB = 57.3 MiB apache2 (22) top: Mem: 2075356k total, 1826196k used, 249160k free, Timeout 35 KeepAlive On MaxKeepAliveRequests 300 KeepAliveTimeout 5 <IfModule mpm_prefork_module> StartServers 10 MinSpareServers 20 MaxSpareServers 20 MaxClients 60 MaxRequestsPerChild 1000 </IfModule> <IfModule mpm_worker_module> StartServers 2 MaxClients 150 MinSpareThreads 25 MaxSpareThreads 75 ThreadsPerChild 25 MaxRequestsPerChild 0 </IfModule>

    Read the article

  • VMWare converter performance

    - by bellocarico
    Hello, I have a question about my test lab. It's more to understand the concept more than apply this into production: I have an ESXi with few VMs linux/windows configured and I'd like to use VMWare converter to create backups. To speedup the process I decided to create a Windows VM on the same ESXi host where I've installed Windows 7 and VMWare Converter. The Host has a gigabit card but it's currently connected to a 100Mb FD port. Windows 7 sees a 1gb card connected. When I do the backup using VMWare converter I specify the host IP as source and destination, so I thought the copy could be faster then use my laptop across the network. Well, to cut a long sotry short: I get dreadful performance (4Mb/sec). I'm a buit confused on this because despite the fact that the host is running 100Mb communication between VMs and hosts shouldn't (correct me if I'm wrong) have any limitation instead. I did tweak windows 7 to optimise network performane but I got just a little improvement. i still need 4 hours to back up a 50Gb (thin) VM. Additionally I wanted to ask: Would jumbo frame help in this? I know that jumbo frame have to be supported end to end, and the network switch where the host is currently connected doesn't support this, but I was wondering: 1) Does ESXi host support jumbo frames at all? 2) Can I enable it somehow? 3) If I do so, I guess bulk transfert between VMs and host would improve, but would this affect the communication going through the real switch as this doesn't do jumbo? Thanks for reading

    Read the article

  • iSCSI performance questions

    - by RyanLambert
    Hi everyone, apologies for the long-winded post in advance... Attempting to troubleshoot some iSCSI sluggishness on a brand new vSphere deployment (still in test). Layout is as such: 3 VSphere hosts, each with 2x 10GB NICs plugged into a pair of Nexus 5020s with a 10gig back-to-back between them. NICs are port-channeled in an active/active redundant fashion (using vPC-mac pinning for those of you familiar with N1KV) Both NICs carry service console, vmotion, iSCSI, and guest traffic. iSCSI is on a single subnet/single VLAN that is not routed through our IP network (strictly layer2) Had this been a 1gig deployment, we probably would have split the iSCSI traffic off onto separate NICs, but the price/port gets rather ridiculous when you start throwing 4+ NICs to a server in a 10gigabit infrastructure, and I'm not really convinced it's necessary. Open to dialogue/tech facts re: this, though. At this point even a single VM guest will boot slowly to iSCSI storage (EMC CX4 on the same Nexus 5020 10gig switches), and restores of VMs from iSCSI take about twice as long as we'd expect them to. Our server folks mentioned that if we split the iSCSI off onto its own NIC, performance seems significantly better. From a network perspective, I've run through the variables I can think of (port configuration errors, MTU problems, congestion etc.) and I'm coming up dry. There really is no other traffic on these hosts other than the very specific test being performed at the time. Important thing to note is that guest traffic works just fine... it seems storage is the only thing affected by whatever gremlin exists. Concluding that we're not 'overutilizing' the network infrastructure since we're doing hardly anything, I'm just looking for some helpful tips/ideas we can use to resolve this... preferably without hurling extra 10gig NICs that are going to sit around 10% utilization while we've got 70+% left on our others.

    Read the article

  • Bad Mumble control channel performance in KVM guest

    - by aef
    I'm running a Mumble server (Murmur) on a Debian Wheezy Beta 4 KVM guest which runs on a Debian Wheezy Beta 4 KVM hypervisor. The guest machines are attached to a bridge device on the hypervisor system through Virtio network interfaces. The Hypervisor is attached to a 100Mbit/s uplink and does IP-routing between the guest machines and the remaining Internet. In this setup we're experiencing a clearly recognizable lag between double-clicking a channel in the client and the channel joining action happening. This happens with a lot of different clients between 1.2.3 and 1.2.4 on Linux and Windows systems. Voice quality and latency seems to be completely unaffected by this. Most of the times the client's information dialog states a 16ms latency for both the voice and control channel. The deviation for the control channels mostly is a lot higher than the one of the voice channels. In some situations the control channel is displayed with a 100ms ping and about 1000 deviation. It seems the TCP performance is a problem here. We had no problems on an earlier setup which was in principle quite like the new one. We used Debian Lenny based Xen hypervisor and a soft-virtualised guest machine instead and an earlier version of the Mumble 1.2.3 series. The current murmurd --version says: 1.2.3-349-g315b5f5-2.1

    Read the article

  • Improving performance by using an additional static file server

    - by Max
    Hello there, I´m planning for a large website that includes many static assets (js, css, images and thumbnails) in the generated pages. That website will use TYPO3 as CMS (is is a customer requirement). I guess I could seriously improve performance / page load times by using a two server setup. One server where the main application (PHP) runs and another one where the static files sit being served by a trimmed down version of apache or something like lighthttpd. Including e. g. js or css files from the file server is of course no big deal. Just use an absolute url http://static.example.com/js/main.js and be done with it. But: that website will have pages with MANY thumbnails of e. g. product images on it. So I see two problems when the main application tries to create a thumbnail of some image: the original image like products/some.jpg is uploaded on the static file server and therefore not on the same server as the PHP application which tries to create the thumbnail. TYPO3 writes created thumbnails to a temp directory which is expected to be on the same server. Therefore, hundreds of thumbnails will be written and served from that temp directory which is on the same server as the main application - the static file server is in that case basically useless, all thumbnails will be requested from the server of the main application. So, my question is: how to overcome this shortcomings? Is it possible to "symlink" some directories to another server? So, for example, if PHP tries to open the original products image for thumbnail creation with imagecreate("products/some.jpg") the products folder actually "points" to the products folder on the static image server? I know something like this can be done with .htaccess but is it possible on file system level?

    Read the article

  • Incredble low disk performance on HP DL385 G7

    - by 3molo
    Hi, As a test of the Opteron processor family, I bought a HP DL385 G7 6128 with HP Smart Array P410i Controller - no memory. The machine has 20GB ram 2x146GB 15k rpm SAS + 2x250GB SATA2, both in Raid 1 configurations. I run Vmware ESXi 4.1. Problem: Even with one virtual machine only, tried Linux 2.6/Windows server 2008/Windows 7, the VMs' feel really sluggish. With windows 7, the vmware converter installation even timed out. Tried both SATA and SAS disks and SATA disks are nearly unsusable, while SAS disks feels extremely slow.I can't see a lot of disk activity in the infrastructure client, but I haven't been looking for causes or even tried diagnostics because I have a feeling that it's either because of the cheap raid controller - or simply because of the lack of memory for it. Despite the problems, I continued and installed a virtual machine that serves a key function, so it's not easy to take it down and run diagnostics. Would very much like to know what you guys have to say of it, is it more likely to be a problem with the controller/disks or is it low performance because of budget components? Thanks in advance,

    Read the article

  • Poor write performance on Debian server running NFS with 22TB exported JFS filesystem

    - by user143546
    I am currently running a debian server that is exporting a large JFS filesystem (22TB) over NFS (nfs-kernel-server.) When attempting to write to the NFS share, the performance is very poor. The 22TB disk is sitting on a NAS mounted using iSCSI. It will bust for a moment near expected line speed, and then sit idle for several seconds. Very little traffic measured in the low kb/sec. The wait peeks on write. When reading from the NFS mount, the system operates at expected speeds (11MB/sec). The issue does not occur when using SFTP, rsync, or local coping (non-nfs). The issue persists between stable and testing releases. On the same machine I have a 14TB ext4 filesystem using the exact same export configuration that does not share the issue. This share is not in regular use and thus not consuming resources. NFS Server: cat /etc/exports /data2 10.1.20.86(rw,no_subtree_check,async,all_squash) cat /sys/block/sdb/queue/scheduler noop [deadline] cfq cat /etc/default/nfs-kernel-server RPCNFSDCOUNT=8 RPCNFSDPRIORITY=0 RPCMOUNTDOPTS=--manage-gids NEED_SVCGSSD= RPCSVCGSSDOPTS= NFS Client: cat /etc/fstab 10.1.20.100:/data2 /root/incoming nfs rw,noatime,soft,intr,noacl 0 2 cat /sys/block/sdb/queue/scheduler noop [deadline] cfq cat /proc/mounts 10.1.20.100:/data2/ /root/incoming nfs4 rw,noatime,vers=4,rsize=262144,wsize=262144,namlen=255,soft,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=10.1.20.86,minorversion=0,addr=10.1.20.100 0 0 This problem has me pretty stumped. Any help would be greatly welcomed. Thanks.

    Read the article

  • "What happens?" server performance monitor

    - by AlexAtNet
    Hello! After reviewing some thread about server monitoring software I end up with a simple question: Which of the server monitoring tools should I use for automatic detection of "abnormal" situations with recommendations on how to fix them? I look for software that checks the system performance after installation and calculate some average load values (memory, CPU, etc). And when something happens (CPU load is increased to 20%) then it tries to detect a reason for this. If it is apache, it should check for access logs. If mysql, it should check mysql logs and tell me what happens. It this is because some user decodes a lot of images, I'd like to know which command is executed, when and user name. The same for disk usage, memory, number of processes, threads and so on. Ideally, this software should periodically checks the system and report problems: errors in PHP error log, outdated packages, security vulnerabilities. In other word I'm looking a software that will keep my simple Debian/Apache/PHP/MySQL server without forcing me to monitor the charts every day. I hope that such program exists. Thanks, Alex

    Read the article

  • Software raid 0 with six disks performance

    - by user134880
    I have some problems with disk performance. I have 6 x WD 500Gb RE4 disks. Each disk gives 135Mb/sec throughput. All measurements are made with hdparm with options "-tT" (I know that it is just synthetic test, but I need some start point to make measurements). I have controller with Sil3124 x 4 ports PCI Express 1x So... RAID0 on controller with 2 disks gives 200Mb/s - ok, pcie limit. RAID0 on motherboard with 2 disks gives 270Mb/s - niceeee :) RAID0 on contorller with 4 disks gives 200Mb/s - ok, pcie limit. RAID0 on controller with 4 disks + 1 disks on motherboard = 340Mb/s ... :( RAID0 on controller with 4 disks + 2 disks on motherboard = 300Mb/s .... why? Any ideas? Maybe need more cpu power? Now there is Pentium D Dual core 2.8Ghz, 4Gb RAM. It is dedicated box for storage.. no other activity.

    Read the article

< Previous Page | 104 105 106 107 108 109 110 111 112 113 114 115  | Next Page >