Search Results

Search found 14292 results on 572 pages for 'high integrity systems'.

Page 459/572 | < Previous Page | 455 456 457 458 459 460 461 462 463 464 465 466  | Next Page >

  • Automated Syslog Error Solution Finder

    - by Dru
    Any automated syslog solution finding frameworks? I want my central syslog server to email a list of problems, their severity and suggested solutions. There have been several questions about centralising system logs and alternative log analysis systems, but I don't get the impression that any of them help with issue resolution. A little background: At work I am now literally doing the work of two people, and both jobs have expanded beyond their initial frameworks. It is not so bad as I have helpers, but they are little more than smart monkeys. While one of my predecessors [I have two, that is how I know I have the jobs of two people] set-up logwatch to email its results out, my monkeys don't have the skills necessary to identify unimportant data. This has caused all of them, and myself sadly, to setup email filters and ignore the whole thing until something goes "bang". It would be handy to have someone else tell them what is important, what is connected, and to suggest a few ways to resolve the issue (I could train then to research the solution first, ha!). My reading of the Splunk and Octopussy sites indicates that I still need to bring my own highly trained monkey to the party. Which I am several years from having.

    Read the article

  • What does opening a serial port do?

    - by reve_etrange
    What does opening the standard PC serial port do, in electrical terms (i.e. what voltages on which pins)? For example, the ancient VB6 program which controls an apparatus I am tasked with maintaining toggles .PortOpen to control some TTL. The connection only used 2 pins (bad solders fell apart), so which pins do I solder to? The only labels / documentation refer to pins 7 and 9, saying 0V and 5V parenthetically, but does .PortOpen really just put 5V between RI and RTS?. As a post script, this isn't the weirdest thing about the set up. The TTL I referred to above also connects to an instrument via a BNC to DB9 (!), with only 1 pin used. I guess there was an assumption about a common ground, since the BNC shielding isn't connected to the GND pin? The connection is to the instrument's 'foot pedal' pin, it was a way to remotely trigger the device. Update According to this page, the DTR and RTS pins can go high when the port is opened. If they were so configured, they will subsequently go low when the port is closed. If DTR and RTS are not enabled, opening the port should set both to low (and keep them low).

    Read the article

  • To what extent is size a factor in SSD performance?

    - by artif
    To what extent is the size of an SSD a factor in its performance? In my mind, correct me if I'm wrong, a bigger SSD should be, everything else being equal, faster than a smaller one. A bigger SSD would have more erase blocks and thus more leeway for the FTL (flash translation layer) to do garbage collection optimization. Also there would be more time before TRIM became necessary. I see on Wikipedia that it remarks that "The performance of the SSD can scale with the number of parallel NAND flash chips used in the device" so it seems throughput also increases significantly. Also many SSDs contain internal caches of some sort and presumably those caches are larger for correspondingly large SSDs. But supposing this effect exists, I would like a quantitative analysis. Does throughput increase linearly? How much is garbage collection impacted, if at all? Does latency stay the same? And so on. Would the performance of a 8 GB SSD be significantly different from, for example, an 80 GB SSD assuming both used high quality chips, controllers, etc? Are there any resources (webpages, research papers, presentations, books, etc) that discuss correlations between SSD performance (4 KB random write speed, latency, maximum sequential throughput, etc) and size? I realize this does not really sound like a programming question but it is relevant for what I'm working on (using flash for caching hard drive data) which does involve programming. If there is a better place to ask this question, eg a more hardware oriented site, what would that be? Something like the equivalent of stack overflow (or perhaps a forum) for in-depth questions on hardware interfaces, internals, etc would be appreciated.

    Read the article

  • I need a driver for my Cardbus USB 2.0 card

    - by Carl
    The picture and details of the card are here: http://www.ht-link.com/en/ProductView.asp?ID=106 The Windows drivers for this don't work right. I tried them on my previous laptop, and then installed the ones off their included CD. (Note that the systems requirements includes a CD-ROM drive...) I have a new laptop, and lost the CD. The website that lists the drivers is broken - the download links don't work. It is here: http://www.ht-link.com/en/DownView.asp?ID=10 I need the very first link - The Win XP drivers for the HT-112NEC. The company does not reply to my e-mails. I've tried searching Google for other sources for the driver (I didn't bother with those sites that want me to install driver detection software or create a log-in for their site). [Here's the problem I am getting using the drivers Windows XP SP3 installs: When I plug in my USB 2.0 hard-disk adapter, a USB Mass Storage Device entry is added in the Device Manager, but there is no entry under Disk Drives, and a drive letter is not assigned, so I can't access the driver. Like I said, this card didn't work without their special drivers on my other laptop, either.]

    Read the article

  • Google Chrome is running my system out of memory

    - by jasondavis
    I am running Windows 7 x64 with 12GB of RAM I often have multiple windows and a ton of tabs open. I use the extension Session Buddy to restore all my windows and tabs once the memory gets too high. So my 12gb of ram will get up to around 93% used because of Chrome, now I can close chrome down and restore the same amount of windows and tabs and it will only use about 25% of memory, it then over time increases back up to the 90% zone after several hours. It seems that when I close tabs, instead of freeing that memory up, it doesn't so that is why the huge increase of memory usage as new tabs are opened and closed it just adds up, this sounds like a huge bug in chrome. Just for an example I just re-booted my system, I only have 1 window with 4 tabs open and in the task manager, it shows 29 chrome.exe processes I then killed all chrome processes and opened a chrome window with just 1 tab, it made 27 chrome.exe processes. Is this an issue that others have? More importantly, is there a fix? UPDATE I just read that each plugin and extension creates a chrome.exe process, I then couunted 24 extensions so that helps explain a portion of the large processes. Still not sure about memory not being freed up though!

    Read the article

  • xen + debian network after upgrade squeeze to wheeze

    - by rush
    I've got a Debian + Xen server. After a system upgrade to the stable version the network doesn't come up after boot. Every time after reboot I need to bring it up manually. The network configuration was not changed during upgrade. Here is /etc/network/interfaces: auto lo iface lo inet loopback auto eth0 iface eth0 inet static address 11.22.33.44 netmask 255.255.255.0 gateway 11.22.33.1 nameserver 8.8.8.8 After boot ip r shows no route and eth0 has no ip address. Manually ip and route setup goes fine and network starts working. Messages from dmesg about network I've found (looks like nothing interesting) [ 3.894401] ACPI: Fan [FAN3] (off) [ 3.894444] ACPI: Fan [FAN4] (off) [ 4.178348] e1000e 0000:00:19.0: eth0: (PCI Express:2.5GT/s:Width x1) 00:1e:67:14:66:c9 [ 4.178351] e1000e 0000:00:19.0: eth0: Intel(R) PRO/1000 Network Connection [ 4.178392] e1000e 0000:00:19.0: eth0: MAC: 10, PHY: 11, PBA No: 0100FF-0FF [ 4.178413] e1000e 0000:02:00.0: Disabling ASPM L0s L1 [ 4.178432] xen: registering gsi 16 triggering 0 polarity 1 -- [ 4.223667] ata5: DUMMY [ 4.223668] ata6: DUMMY [ 4.289153] e1000e 0000:02:00.0: eth1: (PCI Express:2.5GT/s:Width x1) 00:1e:67:14:66:c8 [ 4.289155] e1000e 0000:02:00.0: eth1: Intel(R) PRO/1000 Network Connection [ 4.289245] e1000e 0000:02:00.0: eth1: MAC: 3, PHY: 8, PBA No: 1000FF-0FF [ 4.506908] usb 1-1: new high-speed USB device number 2 using ehci_hcd [ 4.542920] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300) -- [ 10.362999] EXT4-fs (dm-23): mounted filesystem with ordered data mode. Opts: (null) [ 10.419103] EXT4-fs (dm-3): mounted filesystem with ordered data mode. Opts: (null) [ 10.988255] ADDRCONF(NETDEV_UP): eth1: link is not ready [ 13.175533] Event-channel device installed. [ 13.287555] XENBUS: Unable to read cpu state -- [ 13.288670] XENBUS: Unable to read cpu state [ 13.965939] Bridge firewalling registered [ 14.134048] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [ 14.283862] ADDRCONF(NETDEV_UP): peth0: link is not ready [ 14.284543] ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready [ 17.800627] e1000e: peth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx [ 17.801377] ADDRCONF(NETDEV_CHANGE): peth0: link becomes ready [ 18.307278] device peth0 entered promiscuous mode [ 24.538899] eth1: no IPv6 routers present [ 28.570902] peth0: no IPv6 routers present I've upgraded two servers and I've such behaviour on two of them. How to fix this and get network starts automatically on boot?

    Read the article

  • Capabilities of business and SOHO routers

    - by Q8Y
    I'm currently studying for the CCNA certifications (especially for Cisco routers and configuration). I know that business routers provide more features than SOHO routers, the processing speed and RAM can be enough. Assume I need to connect a number of users through a network (accessing internet, share files, printers, ...). I have a high speed connection to the internet and I already applied QoS. How can I find out how many users such a single (SOHO) router could handle? In my case I'd attach to it multiple switches until I have the number of ports needed. Would everything work well and smoothly with 50 users? What about 300? At which point would I need a business router instead? If I implemented VLAN here, would it make any difference in the performance? When do I really need to use more than one router? (Both SOHO and business) I'm thinking that I may need them only if I want to increase the performance (instead of replacing the existing one) and if I have multiple locations, so in this situation I need to have multiple routers, right? Put differently: Is there is a need to have another router if my business all in one place?

    Read the article

  • Postfix : outgoing mail in TLS for a specific domain

    - by vercetty92
    I am trying to configure postfix to send mail in TLS (starttls in fact), but only for a specific destination. I tried with "smtp_tls_policy_maps". This is the only line in my main.cf file regarding TLS configuration, but it seems not working. Here is my main.cf file: queue_directory = /opt/csw/var/spool/postfix command_directory = /opt/csw/sbin daemon_directory = /opt/csw/libexec/postfix html_directory = /opt/csw/share/doc/postfix/html manpage_directory = /opt/csw/share/man sample_directory = /opt/csw/share/doc/postfix/samples readme_directory = /opt/csw/share/doc/postfix/README_FILES mail_spool_directory = /var/spool/mail sendmail_path = /opt/csw/sbin/sendmail newaliases_path = /opt/csw/bin/newaliases mailq_path = /opt/csw/bin/mailq mail_owner = postfix setgid_group = postdrop mydomain = ullink.net myorigin = $myhostname mydestination = $myhostname, localhost.$mydomain, localhost masquerade_domains = vercetty92.net alias_maps = dbm:/etc/opt/csw/postfix/aliases alias_database = dbm:/etc/opt/csw/postfix/aliases transport_maps = dbm:/etc/opt/csw/postfix/transport smtp_tls_policy_maps = dbm:/etc/opt/csw/postfix/tls_policy inet_interfaces = all unknown_local_recipient_reject_code = 550 relayhost = smtpd_banner = $myhostname ESMTP $mail_name debug_peer_level = 2 debugger_command = PATH=/bin:/usr/bin:/usr/local/bin:/usr/X11R6/bin xxgdb $daemon_directory/$process_name $process_id & sleep 5 And here is my "tls_policy" file: gmail.com encrypt protocols=SSLv3:TLSv1 ciphers=high I also tried gmail.com encrypt My wish is to use TLS only for the gmail domain. With this configuration, I don't see any TLS line in the source of the mail. But if I tell postfix to use TLS if possible for all destination with this line, it works: smtp_tls_security_level = may Beause I can see this line in the source of my mail: (version=TLSv1/SSLv3 cipher=OTHER); But I don't want to try to use TLS for the others domains...only for gmail... Do I miss something in my conf? (I also try whith "hash:/etc/opt/csw/postfix/tls_policy", and it's the same) Thanks a lot in advance

    Read the article

  • VPS Memory Exchausted Even With Light Settings

    - by user101570
    Linux noob here. I have a 256MB VPS on Ubuntu 11.04 server and when I run "free -m" the result shows all memory being used (including the second line re: buffers/cache). I found this very strange, considering I only have 5 Apache processes running each chewing up about 20MB each. MYSQL is taking up 30MB. To my knowledge, and according to "top", I have no other memory hogs operating. Settings that may be relevant: PHP memory_limit = 32M MYSQL key_buffer = 16M Prefork MPM Maxclients = 10 So when I reviewed these settings, I naturally thought maxclients was too high, so I tried switching it to 5. Now not only does my memory still show as being 100% used, my website loads much, much slower, despite not getting any traffic aside from mine at the moment. I don't understand this. I thought a single Apache process handles all requests from a client received within the "KeepAliveTimeout" window, which I've set to 2 seconds. With my initial config. of 10 maxclients, my page load times are around .3ms, so a single process should handle that no problem, correct? So next I went to an extreme level of 1 for maxclients. My memory is still at 100% usage and my site loads painfully slow. I'm a noob at a complete loss here. According to the many tutorials I've read on basic server setup, I should be good to go. Help! Please! Edit: total used free shared buffers cached Mem: 256 256 0 0 0 0 -/+ buffers/cache: 256 0 Swap: 0 0 0

    Read the article

  • Raspberry pi slows down my entire network

    - by gnusouth
    Whenever my Raspberry Pi is connected to the network (via ethernet) the entire network is slowed to a crawl. On my main computer, ping times for google.com go from ~10ms to ~200ms and it takes forever to load web pages. Connections are also slow on the Pi, with an apt-get update showing pathetic speeds in the order of 1KB/s. Turning off the Pi completely removes the drag from the network. I've tried static and dynamic IP addresses for the Pi, but both have the same problems. I'm currently using Raspbian (downloaded today), but also had this problem with Arch Linux. I've checked the connection's duplex with dmesg | grep -i duplex, which shows that the Pi's connection is running at 100Mbps, full-duplex, as expected. My modem/router is a Billion 7404VNPX (an Australian thing); relatively high-end, albeit a bit buggy at times (it will occassionally delete all its firewall settings). It assigns IPs in the range 192.168.1.1 to 192.168.1.20 and has 192.168.1.254 as its own IP. When I assign static IPs I tend to use the 192.168.1.200 area. Does anyone have any idea as to what could be causing this weird slowdown? Or any tests I could try? Thanks

    Read the article

  • Subdomain is preventing my search results from rising as expected in page rank

    - by culov
    My problem is that I have a site which has requires a dedicated page for every city I choose to support. Early on, I decided to use subdomains rather than a directly after my domain (ie i used la.truxmap.com rather than truxmap.com/la). I realize now that this was a major mistake because Google seems to treat la.truxmap.com as a completely different site as ny.truxmap.com. So for instance, if i search "la food truck map" my site will be near the top, however, if i search "nyc food truck map" im no where in sight because ny.truxmap.com wouldnt be very high in the page rank by itself, and it doesnt have the boost that it ought to be getting from the better known la.truxmap.com So a mistake I made a year ago is now haunting my page rank. I'd like to know what the most painless way of resolving my dilemma might be. I have received so much press at la.truxmap.com that I can't just kill the site, but could I re-direct all requests at la.truxmap.com to truxmap.com/la and do the same for all cities supported without trashing my current, satisfactory page rank results I'm getting from la.truxmap.com ?? EDIT I left out some critical information. I am using Google Apps to manage my domain (that is, to add the subdomains) and Google App Engine to host my site. Thus, Google Apps provides a simple mechanism to mask truxmap.appspot.com (the app engine domain) as la.truxmap.com, but I don't see how I can mask it as truxmap.com/la. If I can get this done, then I can just 301 redirect la.truxmap.com to truxmap.com/la as suggested below. Thanks so much!

    Read the article

  • Same native and tagged vlan possible on Redhat?

    - by Chris Phillips
    Hi guys and gals, I'm looking at implementing a systems using a number of tagged and a native vlan connected to a server over a a/p bonded interface. The untagged vlan is for physical machine access, the tagged vlans are connected to bridges and then to QEMU VM's inside the machine. Hopefully this plan is fine, but I'm trying to implement a crippled version of this in a dev environment due to a lack of underlying network config in this location where I just have the same single vlan delivered to the machine on a tag AND plain. I'm nto clear if this is going to work (and that I should just be confident that it will work using different vlans) as I'm seeing odd things like a vm is arping out over the vlan out to the core switch, but the arp reply is coming back on the untagged interface. Now an ARP reply is unicast right? So it's a deliberate thing to send the ARP response on the untagged interface, and not a case that a broadcast response isn't being passed on the tagged side... i.e. there's some underlying logic pushing it that way. Something about the MACs somehow? This is on a CentOS 5.5 machine, vlan's from vconfig. (I've seen reference to the Linux mac-vlan project work, but that's not available here by default.) so 1) Should having the SAME vlan tagged and untagged work? 2) Will different tagged vlans to the untagged interface work nice and easily?

    Read the article

  • Sharing music on NAS with Zune and iPod?

    - by osij2is
    After being a long time iPod owner, I'm switching to the new Zune with its subscription model. I haven't bought a Zune yet but I'm planning on doing so within the next month or so. I have approximately 40GB worth of music and my girlfriend has her iPod music library around 30GB. I've been trying to figure out how to migrate all our music off of our laptops/desktops and centralize everything on my NAS. In sharing iPod music isn't too bad. Sharing from one machine to all is fairly easy within the iTunes player. As far as storing all the music on a NAS, again, iPods aren't too bad and imagine other systems aren't difficult. But I'm really new to the Zune and I'm beginning to run into some issues. My questions are: Is it possible to store all music from our iPods and Zune subscriptions and share music between the iPod/Zune within the same file share on my NAS? I'm sure it's possible to store music on a share, but I'm not sure how iTunes and the Zune software differs. Is there 3rd party software, maybe something like DoubleTwist that can sync based from NAS to multiple desktop/laptops? I've never used DoubleTwist but it's something that I found that looks close to being what I need. I've never quite done this myself so I'm trying to find a solution that can: a) store music on a network share; b) sync between different devices (Zune/iPod) seamlessly.

    Read the article

  • Why is windows not able to create a system partition?

    - by hughes
    I'm reinstalling Windows 7 64 bit, and I encountered an issue I've never seen before. I have a legit copy of Win 64 Professional, and I've installed it probably a half dozen times on this machine in the past without a problem. Googling the error only brings me to issues with people who are upgrading to win7. The drive itself seems to not have a problem. I can mount it on other systems and I can create an NTFS partition on it on other machines. I can install Ubuntu on it without any issues. Additionally, if I try using my alternate backup hard drive, the installer gives the same error. I have run diskpart from the setup page and clean seems to report that all is well. However, I cannot get past the screen below, which says Setup was unable to create a new system partition or locate an existing system partition. This happens regardless of whether or not the disk space is already allocated. What is causing this? How do I solve or get past this?

    Read the article

  • Pushing Windows Store apps silently

    - by seagull
    As part of a requirement I have been issued, I seek the ability to push apps from the Windows Store (.appx) to remote users. The framework surrounding the data transmission is sorted; What I need to know is if there is a script that can be sent out along with a URI or the package proper to facilitate installation on the user-side of the Windows Store app. I am well aware that numerous guides exist from Microsoft on pushing out Line-of-Business (LOB) (AKA Enterprise) applications -- that is, apps that have been developed in-house and are not for the consumption of the Windows store. This is inappropriate for my requirement, however; the customer wants their clients to receive apps that appear currently on the Windows Store, and for them to be installed in a silent manner. I've seen this guide; http://blogs.technet.com/b/keithmayer/archive/2013/02/25/step-by-step-deploying-windows-8-apps-with-system-center-2012-service-pack-1.aspx, which details doing exactly this, but it is only applicable to machines that are administered by a system running Windows Server 2012 R2 with the 'System Center 2012' bundle installed. The systems I target are considerably more decentralised than this, making this guide inappropriate. I have a hunch that Microsoft have deliberately designed the Windows Store to be this way, but I figured I ought to ask around before I resign myself to the requirement. Much obliged

    Read the article

  • postfix test and configuration problem

    - by Woho87
    Hi Guys! I installed postfix using sudo yum install postfix postfix-mysql. I'm newbie to mail systems, but I have one AMAZON EC2 instance with a public DNS. I used the public DNS in most cases, when I configured the file main.cf. The public DNS I have is from amazon and it is a long string(ec2-123-34-234-677.....amazon.com). // I configured this on main.cf. I replaced example.com with ec2-123-.......amazon.com myhostname = mail.example.com mydomain = example.com myorigin = $mydomain mydestination = example.com, $transport_maps local_recipient_maps = $alias_maps $virtual_mailbox_maps unix:passwd.byname home_mailbox = Maildir/ How do I test postfix? I just want it to send emails for my web application. I tried to test it with >telnet localhost 25 after I typed in SSH >sudo postfix start. but I recieve the message that telnet command can not be found. I also use the Amazon linux distribution if you want to know. I use it because it is free. What have I done wrong? Are there anymore configurations required pls help!

    Read the article

  • CPU operating temperature ranges

    - by osij2is
    I have an AMD Phenom II 960T with 2 cores unlocked for a total of 6 cores. I don't overclock at all. I have a Arctic Cooling ACALP64 Heatsink/Fan installed. I'm currently running ESXi 5.0 so I have to go into the BIOS to read the CPU temperatures, which at idle seem to be in the 71-74C range. To me, this is pretty high, but I cannot find any official temperature ranges that AMD says the CPU can work well within. There seems to be a lot of questions on superuser and numerous forums around CPU temperatures but no one seems to have a clear consensus as to what the manufacturer temperature ranges are for specific CPUs. I've tried searching through AMDs site to no avail. At this point, I'd be willing to shut off the 2 extra cores if it keeps the heat down but until I get some sort of tolerance or range for temperature, I have no idea if the CPU is being damaged or not. Can anyone point to a direct source, article, FAQ from AMD that specifically states their CPUs temperature range? Or are CPU temperature ranges so varying that there's no possible baseline? Am I being too paranoid about this? To me, anything over 65C is a bit much and if I'm in the low-mid 70s range with NO VMs running, what can I expect if I have several VMs running?

    Read the article

  • How to allow users to transfer files to other users on linux

    - by Jon Bringhurst
    We have an environment of a few thousand users running applications on about 40 clusters ranging in size from 20 compute nodes to 98,000 compute nodes. Users on these systems generate massive files (sometimes 1PB) controlled by traditional unix permissions (ACLs usually aren't available or practical due to the specialized nature of the filesystem). We currently have a program called "give", which is a suid-root program that allows a user to "give" a file to another user when group permissions are insufficient. So, a user would type something like the following to give a file to another user: > give username-to-give-to filename-to-give ... The receiving user can then use a command called "take" (part of the give program) to receive the file: > take filename-to-receive The permissions of the file are then effectively transferred over to the receiving user. This program has been around for years and we'd like to revisit things from a security and functional point of view. Our current plan of action is to remove the bit rot in our current implementation of "give" and package it up as an open source app before we redeploy it into production. Does anyone have another method they use to transfer extremely large files between users when only traditional unix permissions are available?

    Read the article

  • Windows 7 logon script net use fails

    - by Bryan
    Our network PCs currently consists of Windows XP Professional on a mixed 2008/2003 domain, with exception to one machine, which is a new Windows 7 PC we have bought for testing before we deploy the operating system. But we have discovered a problem with our logon script which automatically maps network drives for our users. The logon scripts are done via User GPOs, but the script itself is just a .cmd file using net use. The permissions are perfectly fine, as the same user can log on to a Windows XP machine and get their drives mapped without problem, but this one drive mapping constantly fails. This is repeatable using the net use command, and fails every time - it actually prompts the user for a username and password when executed interactively, yet if we enter \\server\share from a run dialog, the contents of the network share appear and are accessible without any further authentication. The Windows 7 PC (just like the XP systems) are domain members and the account being used is a domain account, which does have access to the share (as stated, it works fine on XP). I fail to understand what is happening here, as other shares on the server get mapped on the Windows 7 system. More info: The effective permissions of the share in question only grant the user 'list' permission on the root directory, the share permissions are 'everyone,full control'. I've created a new share with the same permissions just to test if it was down to the 'list' permissions on the root directory, but the Windows 7 machine maps this one fine.

    Read the article

  • using i7 "gamer" cpu in a HPC cluster

    - by user1219721
    I'm running WRF weather model. That's a ram intensive, highly parallel application. I need to build a HPC cluster for that. I use 10GB infiniband interconnect. WRF doesn't depends of core count, but on memory bandwidth. That's why a core i7 3820 or 3930K performs better than high-grade xeons E5-2600 or E7 Seems like universities uses xeon E5-2670 for WRF. It costs about $1500. Spec2006 fp_rates WRF bench shows $580 i7 3930K performs the same with 1600MHz RAM. What's interesting is that i7 can handle up to 2400MHz ram, doing a great performance increase for WRF. Then it really outperforms the xeon. Power comsumption is a bit higher, but still less than 20€ a year. Even including additional part I'll need (PSU, infiniband, case), the i7 way is still 700 €/cpu cheaper than Xeon. So, is it ok to use "gamer" hardware in a HPC cluster ? or should I do it pro with xeon ? (This is not a critical application. I can handle downtime. I think I don't need ECC?)

    Read the article

  • Gigabit network limited to 25MB/s by CPU. How to make it faster?

    - by netvope
    I have a Acer Aspire R1600-U910H with a nForce gigabit network adapter. The maximum TCP throughput of it is about 25MB/s, and apparently it is limited by the single core Intel Atom 230; when the maximum throughput is reached, the CPU usage is about 50%-60%, which corresponds to full utilization considering this is a Hyper-threading enabled CPU. The same problem occurs on both Windows XP and on Ubuntu 8.04. On Windows, I have installed the latest nForce chipset driver, disabled power saving features, and enabled checksum offload. On Linux, the default driver has checksum offload enabled. There is no Linux driver available on Nvidia's website. ethtool -k eth0 shows that checksum offload is enabled: Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp segmentation offload: on udp fragmentation offload: off generic segmentation offload: off The following is the output of powertop when the network is idle: Wakeups-from-idle per second : 61.9 interval: 10.0s no ACPI power usage estimate available Top causes for wakeups: 90.9% (101.3) <interrupt> : eth0 4.5% ( 5.0) iftop : schedule_timeout (process_timeout) 1.8% ( 2.0) <kernel core> : clocksource_register (clocksource_watchdog) 0.9% ( 1.0) dhcdbd : schedule_timeout (process_timeout) 0.5% ( 0.6) <kernel core> : neigh_table_init_no_netlink (neigh_periodic_timer) And when the maximum throughput of about 25MB/s is reached: Wakeups-from-idle per second : 11175.5 interval: 10.0s no ACPI power usage estimate available Top causes for wakeups: 99.9% (22097.4) <interrupt> : eth0 0.0% ( 5.0) iftop : schedule_timeout (process_timeout) 0.0% ( 2.0) <kernel core> : clocksource_register (clocksource_watchdog) 0.0% ( 1.0) dhcdbd : schedule_timeout (process_timeout) 0.0% ( 0.6) <kernel core> : neigh_table_init_no_netlink (neigh_periodic_timer) Notice the 20000 interrupts per second. Could this be the cause for the high CPU usage and low throughput? If so, how can I improve the situation? The other computers in the network can usually transfer at 50+MB/s without problems. And a minor question: How can I find out what is the driver in use for eth0?

    Read the article

  • email attachments [closed]

    - by Alan Doolan
    My company currently use software on a local machine that will take an email from the email server, extract the attachment, rename it and then add it to a folder on a webserver using ftp. This works well but they are currently asking if it can be done 'in the cloud' or what they really mean, not local. Is there any thing that would do this on the server itself? I should clarify a bit. The attachements are various reports that are being sent to different email addresses (mostly google corporate and free accounts). We need the reports to be on a folder on a webserver so that internal pages can take the information in the reports (csv) and use it on the webpages or adds them to a separate database. The key part being that the files need to be in the particular folders. Though it does work to have a computer running software that will take the files, renames them to the required name and uploads them to the folder it relies too heavily on one computer working all the time. This is not something we can depend on at this point. I'll be honest, I'm a web developer and not strong with server systems past my particular standard requirements so this is beyond me. though yes, I am aware that my boss is not 100% sure what 'cloud' means but likes the word.

    Read the article

  • Developer hardware autonomy in a managed desktop environment [closed]

    - by Troy Hunt
    I’m looking for some feedback on how developer PCs are managed within environments that have a strict managed desktop policy (normally large corporations). For example, many corporate environments control the installation of software and the deployment of patches and virus updates through a centralised channel. This usually means also dictating the OS version and architecture (32 bit versus 64 bit) which will likely also mean standardised hardware configurations. I’m particularly interested in feedback from developers who work in this sort of environment but have a high degree of autonomy over their machines. This might mean choosing your own hardware vendor, OS type and version and perhaps how the machines are built and maintained. I have several specific questions: How do you satisfy the needs of security, governance etc whilst maintaining your autonomy? For example, how do you address concerns about keeping virus definitions and OS patches up to date? Do you have a process for gaining exemption from standard desktop builds and if so, what do you need to demonstrate in order to get this? How have you justified this need to the decision makers? Essentially, what is the benefit to your role as a developer by having this degree of autonomy? Thanks very much everyone. Update: There's a great post from Jean-Paul Boodhoo which addresses the developer tool component of the quesiton here: http://blog.jpboodhoo.com/TheFallacyOfTheStandardizedDeveloperMachineimage.aspx

    Read the article

  • Reverse proxy for mailserver (SMTP + HTTP for web client)

    - by ba
    I'm looking at doing some reverse proxy work for a mail server with corresponding web client. Both servers are running on the same machine, this is not a server with a high load. :) The solution I've discussed with friends is having the mail server/web client on our internal network. Then to put a reverse proxy on the DMZ to service both SMTP and web client HTTP-traffic to the mail server on the internal network. From what I understand this is the recommended secure solution? So far I've thought for the SMTP-proxy part of using postfix which will receive mail, do some spamhause and similar anti-spam measures and if it all checks out, send the mail to the mail server on the inside. The mail server on the inside will send all outgoing mail to the proxy which will then send it out on the Internet. For the web client I'm not sure exactly which software I should be running on the proxy machine, I've been thinking about using Squid -- but that's basically based on the fact that I know squid is a http proxy. The web client data will be sent out over SSL. Reading around some here on Serverfault I've seen other people using Apache with mod_proxy+mod_security for similar situations. Am I thinking correctly for this solution? What software would you guys use and with which modules? Thanks in advance for the help! :)

    Read the article

  • WINDOWS: Your computer hangs. You can windows + R (run dialog) but performance is so halted taskMGR

    - by John Sullivan
    The question is, what process are available to try to recover from total system instability before pulling the plug when we can do nothing but programs or batches in the path from the run dialog (windows + r key), and performance is so dead that taskMGR / procEXP / other programs with visual guis are not usable? I am not a windows expert, but ideally someone out there has written a program that does more or less stuff like this: Immediately set (or perhaps I can set from the run prompt) its priority to extremely high, evaluate performance bottlenecks. E.g. is CPU 100%? If so identify offending program(s) or problems. Attempt / log fixes, then provide crude feedback asking the user if his performance has stabilized enough to abort, wait a few seconds, if no feedback continue, etc. etc. Eventually try to do any "system cleanup" if the program decides it cannot recover and perhaps finally provide a series of beeps to the user, or what have you, to say "OK, I give up, time to pull the plug". Ideally create a log, when able. These kinds of horrible hangs are a situation where surely trying something, anything, is better than nothing -- as long as that something is intelligent -- when the alternative is ripping out the power coord. Again, I am not a windows expert, so perhaps there is a much more elegant "hands on" approach I am not aware of.

    Read the article

< Previous Page | 455 456 457 458 459 460 461 462 463 464 465 466  | Next Page >