Search Results

Search found 14643 results on 586 pages for 'performance comparison'.

Page 144/586 | < Previous Page | 140 141 142 143 144 145 146 147 148 149 150 151  | Next Page >

  • Ngingx max worker_connections and access log

    - by MotoTribe
    I'm troubleshoot an issue with my site. I'm seeing in the ngingx-error.log that the max worker_connection limit has been reached when the site went down. I'm not seeing an increase of requests during that time in the ngingx-access.log. Does that mean the mysql database had a bottleneck at that time that caused the requests to queue up? Or would it not log any requests that where made after the max worker_connection limit has been reached?

    Read the article

  • WD Caviar Green Extremely Slow

    - by Steven
    I am encountering a really weird problem on my WD Caviar Green HDD. Well first of all I have 2 HDDs on my Desktop, one 160GB Seagate holding my Win7 Ultimate x64 and the problematic one, WD 1.5 Caviar Green for storage purpose. My problem is kinda weird, when I transfer files from my Seagate(C:) to my WD (D:) the speed is good (50-60MB/s). Then the problem arises when I transfer too "many" large files, the transfer speed would go straight down to kilobytes/s. Well after I cancelled the transfer and access my D:, even entering a folder requires loading for like 10 seconds. Such problem not only arises when I am transferring files to my D:, it seems like my WD can't handle much activities. For instance, last time I installed my game on D: and I would face much lag after playing for some time. When the same game is installed on C: no problem arises. Does anyone knows what is the problem? P/S: There was one temporary solution that I used to tried. After the "situation" occurs, I tried to access as many folders on D: as I can and let it load, repeating such actions and giving it some time bring the D: back to speedy transfer. However, large transfers would causes the situation to happen again. Does it have something to do with cache whatsoever?

    Read the article

  • Linux: find out what process is using all the RAM?

    - by Timur
    Before actually asking, just to be clear: yes, I know about disk cache, and no, it is not my case :) Sorry, for this preamble :) I'm using CentOS 5. Every application in the system is swapping heavily, and the system is very slow. When I do free -m, here is what I got: total used free shared buffers cached Mem: 3952 3929 22 0 1 18 -/+ buffers/cache: 3909 42 Swap: 16383 46 16337 So, I actually have only 42 Mb to use! As far as I understand, -/+ buffers/cache actually doesn't count the disk cache, so I indeed only have 42 Mb, right? I thought, I might be wrong, so I tried to switch off the disk caching and it had no effect - the picture remained the same. So, I decided to find out who is using all my RAM, and I used top for that. But, apparently, it reports that no process is using my RAM. The only process in my top is MySQL, but it is using 0.1% of RAM and 400Mb of swap. Same picture when I try to run other services or applications - all go in swap, top shows that MEM is not used (0.1% maximum for any process). top - 15:09:00 up 2:09, 2 users, load average: 0.02, 0.16, 0.11 Tasks: 112 total, 1 running, 111 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: 4046868k total, 4001368k used, 45500k free, 748k buffers Swap: 16777208k total, 68840k used, 16708368k free, 16632k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ SWAP COMMAND 3214 ntp 15 0 23412 5044 3916 S 0.0 0.1 0:00.00 17m ntpd 2319 root 5 -10 12648 4460 3184 S 0.0 0.1 0:00.00 8188 iscsid 2168 root RT 0 22120 3692 2848 S 0.0 0.1 0:00.00 17m multipathd 5113 mysql 18 0 474m 2356 856 S 0.0 0.1 0:00.11 472m mysqld 4106 root 34 19 251m 1944 1360 S 0.0 0.0 0:00.11 249m yum-updatesd 4109 root 15 0 90152 1904 1772 S 0.0 0.0 0:00.18 86m sshd 5175 root 15 0 90156 1896 1772 S 0.0 0.0 0:00.02 86m sshd Restart doesn't help, and, by they way is very slow, which I wouldn't normally expect on this machine (4 cores, 4Gb RAM, RAID1). So, with that - I'm pretty sure that this is not a disk cache, who is using the RAM, because normally it should have been reduced and let other processes to use RAM, rather then go to swap. So, finally, the question is - if someone has any ideas how to find out what process is actually using the memory so heavily?

    Read the article

  • Maximise network transfer speed of various applications

    - by Alex
    When using nc, scp, wget to transfer files between 2 machines on a dedicated 2Mbps link, I get speeds between 0.5 and 1 Mbps. However, when I use iperf -c 10.0.1.4 -t 20 -P 12 (for example) I can maximise the speed of the link (getting stable 2Mbps). Is there a way to make single stream transfers (such as those done by scp) to utilise all/most of the link? Some kind of tcp settings, or iptables...?

    Read the article

  • How would I recognize the "spoon-feeding problem" on a dynamic webapp server?

    - by Don Spaulding
    The "spoon-feeding problem", as it was recently explained to me, happens when connections to your application server are tied up feeding data across slow network connections to your clients. This makes sense to me and now I understand the importance of putting a highly-concurrent proxy in front of my app servers. My question is, how did the first person to recognize this problem figure it out? What *nix tools and troubleshooting techniques would help me to recognize this problem if I hadn't had it explained to me?

    Read the article

  • BackupExec 12 + RALUS - VERY slow backups

    - by LVDave
    We use Backup Exec 12 and the Remote Agent for Linux/Unix Servers (RALUS) to backup a large RHEL5 system. For various reasons we need to do a daily working set job. These working-set jobs run abysmally slow. The link between the target machine and the BE server is gigabit, and any other type of job runs 1-3GB/min. These working-set jobs start out at perhaps 40MB/min and over the course of the backup job slowly drops down so low that the BE job rate display in the "current jobs" goes blank.. Since we usually are only doing changed-files for one day, the job is usually small and finishes overnight and we don't worry abotu the slowness, but we had some issues with the backup server, and missed about 6 days of fairly heavy work on the Linux box, so this working-set job will be a doozy.. We have support with Symantec, and I've pestered them a lot about this, they've had me run RALUS in debug mode, sent them that log and a VXgather from the BE host and they had no fix/workaround.. To give an idea, I have the mentioned working-set job running for the last 3 1/2 hours and it's backed up just under 10MEGAbytes.... I'm posting this here to see if anybody in the "real world" has seen this/and/or has any ideas what might be causing these abysmally slow jobs, since Symantec seems to be clueless...

    Read the article

  • troubleshooting really slow login on a (linux) machine

    - by Peeter Joot
    Within the last couple of weeks, any attempt to login to a specific linux server has gotten really slow. Once I've logged in, things appear to run without significant delay, but some other login like activities (like starting a new screen session) are slow. The machine's been rebooted a couple of times recently and that hasn't helped. , and it doesn't appear to be $PATH search (where $PATH can sometimes include bad NFS mounts), which I've seen historically in our environment. I've also tried completely removing my .profile/.bash*/... type of init files to rule out anything bad there. I also see slow login for at least one other userid on the system. One thing I've noticed is the following message when trying to exit from a screen terminal: Utmp slot not found -> not removed and am wondering if this is related (having a vague recollection that Utmp has something to do with login). Any idea what that message means, or how to fix it, and if it would be related? Failing that, what sort of problem determination tools are available to investigate what is slowing down this login process?

    Read the article

  • SQL Management Studio external database only

    - by Robuust
    I'm trying to speed up my PC, and I figured out that a full version of SQL Management Studio 2012 is installed including localhost server. I only need to connect to remote hosts, so running a local server by default should be disabled. Is there an easy way to disable certain parts so I can speed up my PC and booting time? Thanks in advance. I really have no clue what processes I can disable without ruining everything.

    Read the article

  • Best use of new express card on Windows

    - by jckdnk111
    I just bought a 48GB SSD express card for my laptop and I am trying to decide how best to use it. I will be running some sort of virtualization (prob VirtualBox) to test / learn Windows Server administration. I am running Windows 7 Ultimate 64 bit. I have 4GB of RAM and a 7200 RPM SATA hard disk. The express card will read at 115MB/s and write at 65MB/s. So how best to use this new disk? Readyboost, relocate pagefile, store VM disks, mix / match?

    Read the article

  • Apache's htcacheclean doesn't scale: How to tame a huge Apache disk_cache?

    - by flight
    We have an Apache setup with a huge disk_cache (500.000 entries, 50 GB disk space used). The cache grows by 16 GB every day. My problem is that the cache seems to be growing nearly as fast as it's possible to remove files and directories from the cache filesystem! The cache partition is an ext3 filesystem (100GB, "-t news") on an iSCSI storage. The Apache server (which acts as a caching proxy) is a VM. The disk_cache is configured with CacheDirLevels=2 and CacheDirLength=1, and includes variants. A typical file path is "/htcache/B/x/i_iGfmmHhxJRheg8NHcQ.header.vary/A/W/oGX3MAV3q0bWl30YmA_A.header". When I try to call htcacheclean to tame the cache (non-daemon mode, "htcacheclean-t -p/htcache -l15G"), IOwait is going through the roof for several hours. Without any visible action. Only after hours, htcacheclean starts to delete files from the cache partition, which takes a couple more hours. (A similar problem was brought up in the Apache mailing list in 2009, without a solution: http://www.mail-archive.com/[email protected]/msg42683.html) The high IOwait leads to problems with the stability of the web server (the bridge to the Tomcat backend server sometimes stalls). I came up with my own prune script, which removes files and directories from random subdirectories of the cache. Only to find that the deletion rate of the script is just slightly higher than the cache growth rate. The script takes ~10 seconds to read the a subdirectory (e.g. /htcache/B/x) and frees some 5 MB of disk space. In this 10 seconds, the cache has grown by another 2 MB. As with htcacheclean, IOwait goes up to 25% when running the prune script continuously. Any idea? Is this a problem specific to the (rather slow) iSCSI storage? Should I choose a different file system for a huge disk_cache? ext2? ext4? Are there any kernel parameter optimizations for this kind of scenario? (I already tried the deadline scheduler and a smaller read_ahead_kb, without effect).

    Read the article

  • How dow I remove 1.000.000 directories?

    - by harper
    I found that in a directory more than 1.000.000 subdirectories has been created due to a bug. I want to remove all these directories, let's say in the directory WebsiteCache. My first approach was to use the command line tool: cd WebsiteCache rmdir /Q /S . This will remove all subdirectories except the directory WebsiteCache itself, since it is the current working directory. I noticed after two hours that the directoriws starting with A-H have been removed. Why does rmdir removes the directories in alphabetical order? It must take additional effort to do this ordered. What is the fastest way to delete such an amount of directories?

    Read the article

  • MicrosoftOnline Migration - Why do I have to wait several minutes before I can click "Finish?"

    - by Giffyguy
    I'm using the MicrosoftOnline Internet E-Mail Mailbox Migration Wizard. I'm moving my email from several GMail accounts to my Microsoft Exchange Online mailboxes. Every time I migrate all or part of a GMail mailbox, I have to wait about five minutes after it completes migration before the Finish button becomes available. What is going on during this time? Is it something I am doing wrong, or is the system just slow?

    Read the article

  • What are your tricks for optimizing your Subversion configuration?

    - by Scott Markwell
    For a Linux or Windows system, what tricks do you do to optimize your Subversion server? The following are my current tricks for a Linux system serving over Apache with HTTPS and backed by Active Directory using LDAP authentication. Enabling KeepAlive on Apache Disable SVNPathAuthz Increase LDAP Cache Using the FSFS storage method instead of BDB Feel free to call this into question. I don't have hard proof that FSFS out performs BDB, only lots of tribal knowledge and hearsay.

    Read the article

  • SQL Server - VMWare install - Utilize more RAM

    - by alex
    We have a SQL server machine - It’s a VMWare image (running on ESXi hardware etc..) It has windows 2008 x64 standard The SQL install is SQL 2008 standard The virtual machine has 12gb of RAM, and 4 virtual CPU The box is suffering from near 100% CPU a lot of the time I enabled the AWE- but SQL server only seems to use 3-4gb of RAM Is there a way of making it use more available ram more effectively? cache results for example..?

    Read the article

  • What does the 'Burst Rate' stat mean in HDTune?

    - by UpTheCreek
    I recently upgraded my laptop's v slow hard drive to a seagate momentus 7200. Everything is working fine, but I'm a bit confused by these benchmark results: The burst rate is significantly less than the Maximim transfer rate, and not much higher than the normal minimum (if you ignore the spikes). What's going on here? On the HDtune website it defines Burst Rate as: ...the highest speed (in megabytes per second) at which data can be transferred from the drive interface (IDE or SCSI for example) to the operating system. Which begs some questions... e.g. if this is the highest, then how did the bechmarking tool record the 103MB/sec maximum? And if this really is the true maximum, then where is the bottleneck? The laptops SATA interface is on an Intel 82801GBM southbridge controller. When I check in hardware manager, I see that it's driver is iaStor.sys from 2005. Maybe that's the issue? I'll look for a newever version, but any insights would be appreciated. Thanks

    Read the article

  • How do you disable the magnifier in Windows Vista?

    - by PhantomDrummer
    Every so often while I'm working, if I accidentally jolt the mouse, the magnifier starts. I'm fairly sure the cause is that some comination of mouse keys is supposed to start the magnifier, and I occasionally hit the right keys accidentally. However, I've never been able to reproduce the behaviour deliberately so I don't know which combination. This is very annoying since switching the magnifier off is non-trivial (control panel, or run dialog, which are hard to use when the magnifier keeps - ummm - windowing and magnifying the bit of screen you're about to click on :-) ). So I'm wondering if there's any way to disable the magnifier completely, so that it doesn't start when you do whatever it is you'd normally do to the mouse to start it? Anyone know?

    Read the article

  • Losing SQL connections

    - by john pavelka
    sql servr 2005 - Standard; one dedicated sql server (VM); windows server 2003; Small databases; About once a week we lose all sql connections. It seems to fix itself after about 5-10 minutes. System.Web.HttpUnhandledException: Exception of type 'System.Web.HttpUnhandledException' was thrown. --- System.Data.SqlClient.SqlException: Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding. We don't have a fully qualified DBA; it's kind of a joint effort here. Can somebody give me some general ideas for troubleshooting the network side and the application side? We already ran a few tuning profiles and ran through Database Tuning Advisor to apply indexing recommendations. It would sure be nice if there was a way to take a snapshot of what was running on sql server when these 100% cpu spikes occured, but sometimes we're not around. Is it common to throttle CPU for certain processes? Can this be done with Windows server 2003? For example, if security apps were making cpu spike to 100%, is there a way to limit their cpu usage? Any advice is appreciated. thanks,

    Read the article

  • Why does Oracle SQL Developer take so long to open?

    - by oscilatingcretin
    I think anyone who's used Oracle SQL Developer will agree that it's painfully slow on the load. My research has lead me to a solution that seems to have helped a little, and that's telling OSQLD not to check for updates on startup. However, it still takes several minutes to open. What could OSQLD possibly be doing during load time? Is there any way get it to open right away? Edit: Adding potentially relevant system specs: CPU: Intel i5-2520M 2.5 ghz Windows 7 32-bit RAM: 4 gb

    Read the article

  • cset shield --kthread on: should I use this?

    - by lori
    I'm reading up on cpu shielding using Alex Tsariounov's cset utility here: https://rt.wiki.kernel.org/index.php/Cpuset_Management_Utility/tutorial In the tutorial I'm finding the wording around migrating kernel threads from having access to all cpus to running only in a certain cpuset a bit ambiguous The tutorial says the following: Some kernel threads can be moved into the unshielded system cpuset as well. These are the threads that are not bound to specific CPUs. If a kernel thread is bound to a specific CPU, then it is generally not a good idea to move that thread to the system set because at worst it may hang the system and at best it will slow the system down significantly. These threads are usually the IRQ threads on a real time Linux kernel, for example, and you may want to not move these kernel threads into system. If you leave them in the root cpuset, then they will have access to all CPUs. The tutorial then goes on to say: However, if your application demands an even "quieter" shield, then you can move all movable kernel threads into the unshielded system set with the following command. [zuul:cpuset-trunk]# cset shield -k on cset: --> activating kthread shielding cset: kthread shield activated, moving 70 tasks into system cpuset... [==================================================]% cset: done I am confused by this final sentence. By using the word however, it seems to suggest that you typically should not move the movable kernel threads into the unshielded system set. Is this the case, or is it safe to move kernel threads which can be moved into a cpuset, thereby preventing them from running on some cpus?

    Read the article

  • Low FPS in some games, but hardware not fully used

    - by Mario De Schaepmeester
    I just did a little funny experiment in the game/sim "Train Simulator 2013". I normally have good FPS in it (around 30) at full settings. What I did was make a really, really long train so that the calculations the sim needed to make were enormous (the sim is quite realistic, it takes all things into account like speed/acceleration, G-forces, comfort levels, possible wheel slip and many more, and most of those things on each carriage seperately). This resulted in only 14FPS as reported by the game, but it felt more like 8FPS or so. I have a Logitech G15 keyboard which has an LCD, and it allows me to monitor CPU/RAM and video card load on it. The strange thing is, all CPU cores were busy, but the total load was only about 60% maximum at all times. The video card was only on 30% load (possibly an important note, the memory was full, which is however not unusual for the game in question). The RAM had plenty of room and there weren't many operations as it didn't grow or shrink much. I just have the feeling that the game would run smoother if it used more of my hardware power. Why is it not doing so? I had the same in another game, The Elder Scrolls: Morrowind when using more than 100 mods (that all use scripting) and a few high res texture mods, + a full-on graphics improvement program. The engine is very old (2003), and so I thought this might be the cause (not being optimised for multithreading). I had thought of possible causes, like: The operating system doesn't let the games use all the resources. It doesn't make use of multi-threading appropriately. To eliminate the former, I tried a CPU stress tool and that got 100% CPU juice as I let it run, so the OS is not the problem. I gave its thread the "higher" priority though. My actual question In both games, I did things the engine was not really built to do or support. Can those games' framerate be limited cause of their own engine not being able to cope? What is the real reason and more importantly, can I help it? And in any case, could something actually be wrong with my hardware? It's all reasonably new, a couple of months, and I (almost) never experience any other trouble. Modern and much more demanding games work absolutely fine. Specs CPU: AMD Phenom II 965 X4 @ 3.4gHz RAM: 8GB of DDR3 RAM Video: MSI GTX560 (nVidia chip) with 1GB of GDDR5 memory OS: Windows 7 Ultimate 64 bit Nothing overclocked.

    Read the article

  • How many megabytes per second may I expect from a gigabit USB 2.0 network card under linux?

    - by Nakedible
    I'm interested in the actual real-word throughput attainable with an external 1000BaseT USB 2.0 network card under Linux. I have been able to attain 90 megabytes per second on a PCI-E interface, but the USB 2.0 bus has a theoretical limit of 480Mbit/s, and in practice less than 40 megabytes per second. Is the actual throughput attainable with such a card under linux 40, 30, 20, or even as low as 10 megabytes per second, eg. no better than a normal 100BaseT network card?

    Read the article

  • High CPU Steal percentage on Amazon EC2 Instance

    - by Aditya Patawari
    I am experiencing high CPU steal percentage in a Amazon EC2 large instance. I know it means that my virtual CPU is waiting on the real CPU of the machine for time. My question is that what can I do to reduce this percentage and get maximum out of the CPU? Steal percentage is consistently at 20%. System load crosses 10 when this happens. I have checked memory and network and I am sure that they are not the bottleneck. Is that normal for such environment? Also are there any system level optimization techniques for reducing steal percentage form the virtual instance? avg-cpu: %user %nice %system %iowait %steal %idle 52.38 0.00 8.23 0.00 21.21 18.18

    Read the article

  • Why is piping dd through gzip so much faster than a direct copy?

    - by Foo Bar
    I wanted to backup a path from a computer in my network to another computer in the same network over a 100MBit/s line. For this I did dd if=/local/path of=/remote/path/in/local/network/backup.img which gave me a very low network transfer speed of something about 50 to 100 kB/s, which would have taken forever. So I stopped it and decided to try gzipping it on the fly to make it much smaller so that the amount to transfer is less. So I did dd if=/local/folder | gzip > /remote/path/in/local/network/backup.img.gz But now I get something like 1 MB/s network transfer speed, so a factor of 10 to 20 faster. After noticing this, I tested this on several paths and files and it was always the same. Why does piping dd through gzip also increase the transfer rates by a large factor instead of only reducing the bytelength of the stream by a large factor? I'd expected even a small decrease in transfer rates instead, due to the higher CPU consumption while compressing, but now I get a double plus. Not that I'm not happy, but just wondering. ;)

    Read the article

  • Load testing nginx inside AWS

    - by andy
    I'm trying to load test nginx running on AWS. I need to try to optimise it to handle 1Gbps of inbound traffic. Currently I've got it to peak at 85Mbit/s by running nginx on an m1.large with 4 other machines hitting it by using ab with -i (for head requests) -k (keepalives) -r (ignore failed requests) -n 500000 -c 20000. I'm struggling to generate more than 85 Mbit/s traffic from 4 machines, yet when I do scp a large file I get nearly 0.25Gbit/s of traffic going over the network. Are there any tools or approaches that I could use to load test nginx that might generate more load? I'm only interested in inbound traffic, so perhaps a DoS tool could help if it chucks away responses? I'm hitting a very small (40 byte) static asset, and have peaked at handling 50K concurrent connections and getting 25k reqs/s when just using a single load generator machine.

    Read the article

< Previous Page | 140 141 142 143 144 145 146 147 148 149 150 151  | Next Page >