Search Results

Search found 26263 results on 1051 pages for 'linux guest'.

Page 150/1051 | < Previous Page | 146 147 148 149 150 151 152 153 154 155 156 157  | Next Page >

• Can I install linux has the host in a dell poweredge server (r710)?

  - by bksunday

  I might have a deal on dual six core poweredge server and I'm about to go test its performance but I'm wondering few things which I can't find answers for, and I can't test them before buying the machine. I don't want vmware at all so can I just wipe it and install linux instead, or is it embedded in some parts I have no access to. Will I still be able to update different firmwares (perc controllers, motherboard, etc) on this dell poweredge or does it require to have the vmware esxi installed as the host os. And optionally.. is there any foreseeable problems in doing so?

  Read the article

• How to configure linux routing/filtering to send packets out one interface, over a bridge and into another interface on the same box

  - by rj75

  I'm trying to test a ethernet bridging device. I have multiple ethernet ports on a linux box. I would like to send packets out one interface, say eth0 with IP 192.168.1.1, to another interface, say eth1 with IP 192.168.1.2, on the same subnet. I realize that normally you don't configure two interfaces on the same subnet, and if you do the kernel routes directly to each interface, rather than over the wire. How can I override this behavior, so that traffic to 192.168.1.2 goes out the 192.168.1.1 interface, and visa-versa? Thanks in advance!

  Read the article

• Where can I get a Linux distro iso based on a 2.4 kernel?

  - by Mike

  I need to get my hands on an ISO file for a Linux distribution with a 2.4 kernel. I'm looking for an ISO specifically so I can use it with my Oracle VirtualBox. Since 2.4 is so old these days, I'll explain that I'm looking for it because my company uses an ancient 2.4 uClinux distro on our ancient hardware in our devices. I'd like to run some "desktop" tests using the same kernel version as what's in the hardware. As far as I can tell I can't run uCLinux on a desktop, so next best thing, I'd like to get anything running 2.4.

  Read the article

• Is there a way to redirect certain URLs to specific web browsers in Linux?

  - by jraxxo

  I'm using Chrome as my default browser in Ubuntu 12.10. I need to use Firefox for business purposes (certain websites pertaining to my work only work with Firefox). Is there a way to force Ubuntu to use Firefox for certain types of URLs (maybe as defined by a regular expression) while maintaining Chrome as my default browser for all my other tasks? Perhaps as a shell script running in the background? I'd like this to work system-wide, covering links from Chrome itself as well as PDFs/ODTs, etc. I have searched for solutions, but I couldn't find anything besides OpenWith, a Firefox extension which adds a button to open certain links in other browsers which would again require me to open Firefox beforehand, which does not help me at all. Does anyone have any ideas? Something like Choosy for Linux?

  Read the article

• Remote desktop into a linux machine, is this possible?

  - by fire

  I have a VPS by my host running on a linux server, and they have given me SSH access. Is it possible to remote desktop into the server, like you can on Windows, so that I can physcially click on things rather than having to use SSH commands? Surely this must be running on Fedora or Ubuntu etc. so there is some type of OS. You would probably have to install something on the servers end I suppose but just want to know if its possible and what the options are. And before you say "why not ask your host" I find superuser responses are usually much quicker :-)

  Read the article

• Configure Linux server hardware buttons for soft reset or power cycle?

  - by Jakobud

  I have a small modest CentOS server at home. I run it headless because anytime I access it, it's always via SSH. Anyways, tonight it became unresponsive to the network. I could not connect to it to investigate. In this case, I have to hook up a keyboard and monitor to see the problem. I ended up just rebooting it. But after this experience, I was wondering if it's possible to configure the hardware buttons on the CPU case to perform a graceful reboot or graceful power cycle in Linux. Even though the server becomes unresponsive once in a blue moon, it would be nice to simply press a button and have it shutdown all services and gracefully reboot. Anyone know how this could be accomplished?

  Read the article

• Nginx works on my linux machine but is not accessible from other computers in my local network

  - by crooveck

  In my LAN network I have a server with Scientific Linux (RedHat or Fedora based distro), I've done yum install nginx but the welcome page is not accessible from other computers in my network. When I do telnet open localhost 80 and then GET / HTTP/1.0 I get some html code from nginx, so it's running for sure. But when I want to connect remotly, doing telnet open 192.168.3.130 80 I get: Trying 192.168.3.130... telnet: Unable to connect to remote host: No route to host So I assume that there is something wrong with my network settings, maybe iptables or something else? Next step, I turned off iptables: service iptables stop and it helped, now I can connect remotely using telnet. So I think, I need to fix my iptables rules. I did some googling and found this rule -A INPUT -m state --state NEW -m tcp -p tcp --dport 80 -j ACCEPT but it still didn't allow me to connect remotely when iptables is up. Can someone please help me setting a proper iptables configuration?

  Read the article

• Total network data sent/received of a non-daemon Linux process?

  - by leden

  I'm looking for a simple and effective way of measuring total bytes received/sent from a single process upon its termination. Basically, I am looking for a tool which has the interface similar to "time" and "/usr/bin/time", e.g. measure-net-data <prog_to_run> <prog_args> Received (b): XYZ Sent (b): ABC I know that there are many tools for bandwidth/network monitoring, but as far I can tell all of them are performing the measurements it real-time, which is inappropriate not only because of overhead but also because of the inconvenience - I would need to stop the program, capture the output of the tool and then kill it. I have seen that newer versions of Linux 2.6.20+ provide /proc/<pid>/io/ which contain the information I'm looking for; however, everything under /proc/<pid> when the process terminates, so I'm again back to the same problem as with any network monitoring tool.

  Read the article

• Program to Queue Files for Copy/Move/Delete in linux?

  - by laliga

  I've search the net for Linux's answer to something like Teracopy (Windows)... but could not find anything suitable. The closest things i got are: Krusader. Mentioned in their features but indicated as 'not implemented yet'. MiniCopier. A java based app http://a.courreges.free.fr/projets/minicopier/minicopier-en.php rsync is not an option. Can someone recommend me a simple file copy tool that can queue files for copy/move/delete? Preferably if I can drag and drop from Nautilus. If something like this does not exist, can someone please tell me why? ...am I the only person that needs something like this?

  Read the article

• A "quick" vector editor (SVG) for Linux (for annotating images?)

  - by sdaau

  I often need to take a bitmap (.png) image, and draw some lines or text on top of it, and possibly export a new, thusly "annotated" image. I know I can basically do all this in inkscape - but inkscape is a complex program, and it needs almost a minute to start up properly on my PCs. So I was thinking - is there something like a "mini" vector editor for Linux, which would start up fast, and allow me to: Right-click, open an image in this editor program The program scales the active "document"/"window size" to the size of the image I can zoom in/zoom out (and possibly crop) the image I can add at least lines, boxes and text in different colors? A bonus for me would be to have the overlay graphics saved as SVG format, say with the same filename as the image - as in, "image.png.svg" being saved in the same directory where the original "image.png" is located (thus allowing opening and editing these "annotations" further, either in this editor, or possibly in inkscape). And another bonus would be the export of the annotated image to a bitmap. Anyone know about anything like this?

  Read the article

• How to save your Linux state (suspend to disk) periodically to recover from crashes?

  - by WoLpH

  One in a while my laptop crashes/dies because of a bad/empty battery, crappy wifi driver or whatever other reason. For a while I've wondered if it's possible to force Linux to periodically save the state (like vmware snapshots) to disk so you can restore from that with possibly slightly outdated work but at least with all of your apps open in the same state you've left them. I don't really see the point in having to boot everything from cratch constantly, although KDE saves your state on logout, that doesn't happen periodically (by default) either. It would make it much nicer to recover from your crashes if your ram was written to disk periodically. Anyone know if there's a system call to do this without also shutting down the machine? Even a manual button to save the entire state would be nice.

  Read the article

• Problems setting NTP sever with w32tm for a DC that is a Hyper-V guest

  - by R.Tonheim

  Hello ! I have tried to sett my DC to get its time from several NTP severs. I follow this answer (http://serverfault.com/questions/24298/w32time-sync-problems-for-hyper-v-guests-w32time-event-ids-38-24-29-35/24299#24299) to do it. First I disable Time Synchronization in the Hyper-V Integration Services for each guest. Then restart the Windows Time serviceon the guest. I had before this used this command: w32tm /config /manualpeerlist:"ntp.uio.no;timekeeper. uio.no;nissen.uio.no;0.no.pool.ntp.org;1.no.pool.ntp.org;2.no.pool.ntp.org" /syn cfromflags:manual /reliable:yes /update And the cmd sad: The command completed successfully. But the time was still 10 min wrong... I run w32tm again after restarted the DC without it having any effect. The w32tm /query /status still say: "Source: Local CMOS Clock" FROM MY CMD: Microsoft Windows [Version 6.0.6002] Copyright (c) 2006 Microsoft Corporation. All rights reserved. C:\Users\Administrator.MHGw32tm /query /status Leap Indicator: 0(no warning) Stratum: 1 (primary reference - syncd by radio clock) Precision: -6 (15.625ms per tick) Root Delay: 0.0000000s Root Dispersion: 10.0000000s ReferenceId: 0x4C4F434C (source name: "LOCL") Last Successful Sync Time: 05.09.2009 20:06:21 Source: Local CMOS Clock Poll Interval: 6 (64s) C:\Users\Administrator.MHGw32tm /config /manualpeerlist:"ntp.uio.no;timekeeper. uio.no;nissen.uio.no;0.no.pool.ntp.org;1.no.pool.ntp.org;2.no.pool.ntp.org" /syn cfromflags:manual /reliable:yes /update The command completed successfully. C:\Users\Administrator.MHGw32tm /query /status Leap Indicator: 0(no warning) Stratum: 1 (primary reference - syncd by radio clock) Precision: -6 (15.625ms per tick) Root Delay: 0.0000000s Root Dispersion: 10.0000000s ReferenceId: 0x4C4F434C (source name: "LOCL") Last Successful Sync Time: 05.09.2009 20:06:21 Source: Local CMOS Clock Poll Interval: 6 (64s) C:\Users\Administrator.MHG

  Read the article

• VMware Workstation update stopped autofit guest working

  - by u8sand

  VMware one day offered an update (8.0.1 build-528992 my current version) from 8.0. I accepted it, because updates usually fix problems. However not in this case... Previously it worked very well. Now, it still "works" but the one glitch I'm getting makes it too hard to deal with. This screenshot will explain my problem: As you can see, my virtual PC is not resizing correctly. (This happens with any operating system), autofit guest just doesn't work - it only results in things like this happening. Thanks to the tools it becomes very hard to NOT autofit guest. I've tried uninstalling 8.0.1 completely and installing 8.0 again but with the same results. I don't really understand what the new update has done to VMware Workstation or to my virtual machines. I do believe this isn't VMware Workstation's direct fault but from VMware Tools, which would explain why going back to 8.0 didn't work since VMware tools has its own updates. How can I fix this? The host is running Windows 7 Ultimate 64-bit.

  Read the article

• VirtualBox problems writing to shared folders (Guest Additions installed)

  - by vincent

  I am trying to setup a shared folder from the host (ubuntu 10.10) to mount on a virtualized CentOS 5.5 with Guest Additions (4.0.0) installed (Guest addition features are working ie. seamless mode etc.). I am able to successfully mount the share with: mount -t vboxsf -o rw,exec,uid=48,gid=48 sf_html /var/www/html/ (uid and guid belong to the apache user/group) the only problem is that once mounted and I try to write/create directories and files I get the following: mkdir: cannot create directory `/var/www/html/test': Protocol error I am using the proprietary version of VirtualBox version 4.0.0 r69151. Has anyone had the same problem and been able to fix it or has any idea how to potentially fix this? Another question, the reason for setting this up is this. Our production servers are on CentOS 5.5 however I am a great fan of Ubuntu and would like to develop on Ubuntu rather than CentOS. However in order to stay as close to the production environment I would like to virtualize CentOS to use a web server and use the shared folder as web root. Anyone know whether this isn't a good idea? Has anyone successfully been able to set this up? Thanks guys, your help is always much appreciated and if you need any more information please let me know.

  Read the article

• cloning a kvm guest os to a vmdk file

  - by Bond

  I have a production environment where I am having 4 Guest OS running on a Ubuntu server which uses kvm. These OS are in an LVM based setup.I want these Virtual Machines to be in a vmdk format also.Where people would do experiments with these Virtual Machines so this in a vmware environment (or it can be Xen too) would be different from the kvm server.I would not have any control on that other environment so I want to give people vmdk images of these virtual machines. The production Virtual Machines will still keep running on kvm server but the VMs on which experiments would be done would be of type vmdk.(vmdk is a constraint) Here is output of lvscan ACTIVE '/dev/abcd/lvm1' [100.00 GiB] inherit ACTIVE '/dev/abcd/lvm2' [150.00 GiB] inherit ACTIVE '/dev/abcd/lvm3' [50.00 GiB] inherit ACTIVE '/dev/abcd/lvm4' [100.00 GiB] inherit I was reading man page of qemu-img and what I understand is I need to first create a qcow image file which I need to populate and then convert that to a vmdk file. Is that understanding correct? Now suppose /dev/abcd/lvm4 is the virtual machine with which I am going to start this experiment.I can shutdown the production VMs for some time to do this. So is the following way correct to go on server 1 (where kvm is running) qemu-img convert -c -f raw -O vmdk /dev/abcd/lvm4 /backup/lvm4.img or it will affect the lvm4 on kvm server 1. I do not want the VM running on original server to at all loose its any of the content but also have a vmdk file for each of the Guest OS on kvm. Before I proceed with any of the above things on the production machine I just want to make sure that I am doing the correct thing so I asking here.

  Read the article

• Cannot Connect To VMWare Guest OS Using Either RDP or VNC

  - by Humanier

  Hi, I have a PC (Windows XP SP3) with VMWare Workstation 7 installed. The VMWare hosts Windows Server 2003 Enterprise Edition R2. RealVNC (4.1.3) is installed on both OS'es. Both of them use Hamachi2. Host OS (WinXP) also runs ZoneAlarm Firewall. Hamachi network is set as trusted. My goal is to allow RDP and VNC connections to be made to the guest OS (Windows Server 2003). Both options work absolutely fine if I connect from the host OS. However I have problems when other computers from our Hamachi network try to connect the guest OS (Win2K3). 1) RDP connections. RDP window opens, shows black content and after 15-20 seconds displays following error: http://lh6.ggpht.com/_yQhsRRimgKU/TArRrtiteQI/AAAAAAAABZA/e96za-y9wzo/rdp_error.JPG 2) RealVCN connections. Users are able to connect but all they see is a black screen inside VNC window. At the same their input (keystrokes or mouse moves/clicks) are visible when looking at the console window of the Win2K3. I really appreciate any ideas on how to resolve mentioned problems. Thank you in advance.

  Read the article

• Cannot Connect To VMWare Guest OS Using Either RDP or VNC

  - by Humanier

  Hi, I have a PC (Windows XP SP3) with VMWare Workstation 7 installed. The VMWare hosts Windows Server 2003 Enterprise Edition R2. RealVNC (4.1.3) is installed on both OS'es. Both of them use Hamachi2. Host OS (WinXP) also runs ZoneAlarm Firewall. Hamachi network is set as trusted. My goal is to allow RDP and VNC connections to be made to the guest OS (Windows Server 2003). Both options work absolutely fine if I connect from the host OS. However I have problems when other computers from our Hamachi network try to connect the guest OS (Win2K3). 1) RDP connections. RDP window opens, shows black content and after 15-20 seconds displays following error: http://lh6.ggpht.com/_yQhsRRimgKU/TArRrtiteQI/AAAAAAAABZA/e96za-y9wzo/rdp_error.JPG 2) RealVCN connections. Users are able to connect but all they see is a black screen inside VNC window. At the same their input (keystrokes or mouse moves/clicks) are visible when looking at the console window of the Win2K3. I really appreciate any ideas on how to resolve mentioned problems. Thank you in advance.

  Read the article

• Xen Windows Guest spawn doesn't spawn a vnc display

  - by Henrik P. Hessel

  I'm using this HVM File to create a new guest kernel = "/usr/lib/xen-3.2-1/boot/hvmloader" builder='hvm' memory = 4096 # Should be at least 2KB per MB of domain memory, plus a few MB per vcpu. shadow_memory = 64 name = "hessel-windows2008" vif = [ 'ip=188.40.xx.xx,mac=00:16:3E:C1:8F:CE' ] acpi = 1 apic = 1 disk = [ 'file:/home/xen/disks/hessel/win2008/win2008.img,hda,w', 'file:/home/xen/isopool/win2008_32.iso,hdc:cdrom,r' ] device_model = '/usr/lib/xen/bin/qemu-dm' #----------------------------------------------------------------------------- # boot on floppy (a), hard disk (c) or CD-ROM (d) # default: hard disk, cd-rom, floppy boot="dc" sdl=0 vnc=1 vncdisplay=1 vnclisten="0.0.0.0" vncconsole=1 vncpasswd='howtoforge' stdvga=0 serial='pty' usbdevice='tablet' The guest is created without an error. But no vnc display is created. Any ideas, how to fix that? prometheus:~# netstat -ant Active Internet connections (servers and established) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 0 0 127.0.0.1:615 0.0.0.0:* LISTEN tcp 0 0 0.0.0.0:111 0.0.0.0:* LISTEN tcp 0 0 0.0.0.0:8080 0.0.0.0:* LISTEN tcp 0 0 0.0.0.0:53 0.0.0.0:* LISTEN tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN tcp 0 0 127.0.0.1:6010 0.0.0.0:* LISTEN tcp 0 232 188.40.xx.xx:8080 195.36.75.26:54032 ESTABLISHED tcp 0 0 188.40.xx.xx:8080 195.36.75.26:53085 ESTABLISHED tcp6 0 0 :::8080 :::* LISTEN tcp6 0 0 :::53 :::* LISTEN tcp6 0 0 :::22 :::* LISTEN tcp6 0 0 ::1:6010 :::* LISTEN

  Read the article

• Using VMware Guest OS to enable Host OS to ssh to remote network

  - by Reuben L.

  Basically I have an issue because my host OS is 64-bit Linux Mint (Ubuntu derived) and it doesn't seem to be compatible with the Juniper Network Connect that is used by the network at my workplace. Thus, I am unable to ssh from terminal to the network. I can't make changes to the workplace network either so that leaves me with looking for solutions on my end. The main reason for me to access the network from home is to check on my running processes or to issue more commands to a few workstations. Putty is the desperate choice I usually make but it means I have to reboot to Windows and also have limited control. I've tried several other methods and they have all failed. Recently, I setup a VM with Windows 7 as the guest OS. Now half my problems are fixed as I don't have to physically reboot the system - I just have to engage Juniper Network Connect on the VM. However, I would still like to use my Linux terminal to ssh to the network. It sounds plausible that I could somehow manipulate ports to connect to the remote network from the host OS tunneled through the guest OS, but I really have no clue how to do so... Can anyone help?

  Read the article

• Cannot Connect To VMWare Guest OS Using Either RDP or VNC

  - by Humanier

  I have a PC (Windows XP SP3) with VMWare Workstation 7 installed. The VMWare hosts Windows Server 2003 Enterprise Edition R2. RealVNC (4.1.3) is installed on both OS'es. Both of them use Hamachi2. Host OS (WinXP) also runs ZoneAlarm Firewall. Hamachi network is set as trusted. My goal is to allow RDP and VNC connections to be made to the guest OS (Windows Server 2003). Both options work absolutely fine if I connect from the host OS. However I have problems when other computers from our Hamachi network try to connect the guest OS (Win2K3). RDP connections. RDP window opens, shows black content and after 15-20 seconds displays following error: RealVCN connections. Users are able to connect but all they see is a black screen inside VNC window. At the same their input (keystrokes or mouse moves/clicks) are visible when looking at the console window of the Win2K3. I really appreciate any ideas on how to resolve the mentioned problems.

  Read the article

• Ubuntu 12.10 64bit host reboots when trying to install any guest system using VirtualBox

  - by gts123

  I am having a really nasty problem with VirtualBox as everytime I try to install any guest OS(using ISO file as CD for installation media), the installation starts normally but as soon as it is about to start either installing to virtual hard drive or loading(e.g. as LiveFS) it causes the host system to reboot abruptly. Config is as below: Host system: Ubuntu 12.10 64bit - Intel® Core i7-2640M CPU @ 2.80GHz × 4 Virtualbox version: 4.1.18_Ubuntu r78361 Guest OS systems tried: 32bit version of FreeBSD 9, Debian 6, Tails 0.14 VM setup Tried to have the minimal setup necessary just in case it would avoid for each system to make sure I'd avoid conflicts, but to no avail. I've tried different values and combinations of the below but the problem still persists: Shared Clipboard: Disabled Show in fullscreen/seamless: Disabled Remember runtime changes: DIsabled Base Memory: 2048 MB Chipset: PIIX3 IO APIC: Disabled EFI: Disabled Absolute Pointing device: Disabled Processor(s): 1 CPU PAE/NX: Disabled VT-x/AMD-V: Disabled Video Memory: 12 MB 3d/2d acceleration: Disabled Storage IDE COntroller: PIIX3 (same as chipset instead of PIIX4) Use host I/O cache: No Audio: disabled Network adapter: NAT USB controller: disabled No shared folder Also another sideeffect of the reboot is that it appears that it does not log any information in the error log files; not making things any easier. Please help.

  Read the article

• Cannot Connect To VMWare Guest OS Using NEither RDP Nor

  - by Humanier

  Hi, I have a PC (Windows XP SP3) with VMWare Workstation 7 installed. The VMWare hosts Windows Server 2003 Enterprise Edition R2. RealVNC (4.1.3) is installed on both OS'es. Both of them use Hamachi2. Host OS (WinXP) also runs ZoneAlarm Firewall. Hamachi network is set as trusted. My goal is to allow RDP and VNC connections to be made to the guest OS (Windows Server 2003). Both options work absolutely fine if I connect from the host OS. However I have problems when other computers from our Hamachi network try to connect the guest OS (Win2K3). 1) RDP connections. RDP window opens, shows black content and after 15-20 seconds displays following error: http://lh6.ggpht.com/_yQhsRRimgKU/TArRrtiteQI/AAAAAAAABZA/e96za-y9wzo/rdp_error.JPG 2) RealVCN connections. Users are able to connect but all they see is a black screen inside VNC window. At the same their input (keystrokes or mouse moves/clicks) are visible when looking at the console window of the Win2K3. I really appreciate any ideas on how to resolve mentioned problems. Thank you in advance.

  Read the article

• KVM Guest Reboot Loop

  - by javano

  I have been pulled into a situation where a KVM server (CentOS 6.2) lost power and upon reboot one of the guests hasn't started up again (XP SP3). I have SSH'ed in and someone must have changed something relating to the hyper visors prior to the power loss, but not rebooted all the guests. This particular guest wouldn't start because it was configured to use /usr/bin/qemu-system-x86_64 which isn't there now (assuming it was before?). I changed it to use /usr/libexec/qemu-kvm as this is what all the other guests on this server seem to be using, and its booting up. Using virt-manager on my local machine I can connect to the display of the XP machine and it gets as far as this screen; http://support.gateway.com/emachines/issues/2-1131285152-01.gif The problem I face now, is that which ever option I choose the machine just reboots. So it's and endless loop. I thought that perhaps a file system error maybe present due to the unclean shutdown. There is an XP SP3 ISO mounted under the guest, which I booted from in an attempt to access the recovery tools, but I don't have the Administrator password! I am out of ideas, and it's turning out to be quite the conundrum. Should I use a 3rd party live CD to test the FS for errors? How else can I trouble shoot these restarts?

  Read the article

• Assign individual NIC to KVM guest

  - by Bin S

  I have a server with 6 NICs installed and is running Ubuntu 12.04LTS. I want to setup 4 guest VMs using kvm. Now I want to assign 2 NICs for the host(1 Public IP and 1 private IP), and 1 NIC each to 4 guest VM(all private IP). How do I do this? /etc/network/interfaces I am having trouble with my configuration file shown below: # The loopback network interface auto lo iface lo inet loopback # The primary network interface auto eth0 iface eth0 inet static address 192.168.1.109 netmask 255.255.255.0 gateway 192.168.1.5 auto eth1 iface eth1 inet static address 192.168.1.117 netmask 255.255.255.0 auto eth2 iface eth2 inet manual auto br0 iface br0 inet static address 192.168.1.118 netmask 255.255.255.0 bridge_ports eth2 bridge_fd 9 bridge_hello 2 bridge_maxage 12 bridge_stp off auto eth3 iface eth3 inet manual auto br1 iface br1 inet static address 192.168.1.119 netmask 255.255.255.0 bridge_ports eth3 bridge_fd 9 bridge_hello 2 bridge_maxage 12 bridge_stp off auto eth4 iface eth4 inet manual auto br2 iface br2 inet static address 192.168.1.123 netmask 255.255.255.0 bridge_ports eth4 bridge_fd 9 bridge_hello 2 bridge_maxage 12 bridge_stp off auto eth5 iface eth5 inet manual auto br3 iface br3 inet static address 192.168.1.124 netmask 255.255.255.0 bridge_ports eth5 bridge_fd 9 bridge_hello 2 bridge_maxage 12 bridge_stp off

  Read the article

• Linux not buffering block I/O when the device is not "in use" (i.e. mounted)

  - by Radek Hladík

  I am installing new server and I've found an interesting issue. The server is running Fedora 19 (3.11.7-200.fc19.x86_64 kernel) and is supposed to host a few KVM/Qemu virtual servers (mail server, file server, etc..). The HW is Intel(R) Xeon(R) CPU 5160 @ 3.00GHz with 16GB RAM. One of the most important features will be Samba server and we have decided to make it as virtual machine with almost direct access to the disks. So the real HDD is cached on SSD (via bcache) then raided with md and the final device is exported into the virtual machine via virtio. The virtual machine is again Fedora 19 with the same kernel. One important topic to find out is whether the virtualization layer will not introduce high overload into disk I/Os. So far I've been able to get up to 180MB/s in VM and up to 220MB/s on real HW (on the SSD disk). I am still not sure why the overhead is so big but it is more than the network can handle so I do not care so much. The interesting thing is that I've found that the disk reads are not buffered in the VM unless I create and mount FS on the disk or I use the disks somehow. Simply put: Lets do dd to read disk for the first time (the /dev/vdd is an old Raptor disk 70MB/s is its real speed): [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 36.8038 s, 71.2 MB/s Buffers: 14444 kB Rereading the data shows that they are cached somewhere but not in buffers of the VM. Also the speed increased to "only" 500MB/s. The VM has 4GB of RAM (more that the test file) [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 5.16016 s, 508 MB/s Buffers: 14444 kB [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 5.05727 s, 518 MB/s Buffers: 14444 kB Now lets mount the FS on /dev/vdd and try the dd again: [root@localhost ~]# mount /dev/vdd /mnt/tmp [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 4.68578 s, 559 MB/s Buffers: 2574592 kB [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 1.50504 s, 1.7 GB/s Buffers: 2574592 kB While the first read was the same, all 2.6GB got buffered and the next read was at 1.7GB/s. And when I unmount the device: [root@localhost ~]# umount /mnt/tmp [root@localhost ~]# cat /proc/meminfo | grep Buffers Buffers: 14452 kB [root@localhost ~]# dd if=/dev/vdd of=/dev/null bs=256k count=10000 ; cat /proc/meminfo | grep Buffers 2621440000 bytes (2.6 GB) copied, 5.10499 s, 514 MB/s Buffers: 14468 kB The bcache was disabled while testing and the results are same on faster (newer) HDDs and on SSD (except for the initial read speed of course). To sum it up. When I read from the device via dd first time, it gets read from the disk. Next time I reread it gets cached in the host but not in the guest (thats actually the same issue, more on that later). When I mount the filesystem but try to read the device directly it gets cached in VM (via buffers). As soon as I stop "using" it, buffers are discarded and the device is not cached anymore in the VM. When I looked into buffers value on the host I realized that the situation is the same. The block I/O gets buffered only when the disk is in use, in this case it means "exported to a VM". On host, after all the measurement done: 3165552 buffers On the host, after the VM shutdown: 119176 buffers I know it is not important as the disks will be mounted all the time but I am curious and I would like to know why it is working like this.

  Read the article

< Previous Page | 146 147 148 149 150 151 152 153 154 155 156 157  | Next Page >