Search Results

Search found 25088 results on 1004 pages for 'dsl linux'.

Page 348/1004 | < Previous Page | 344 345 346 347 348 349 350 351 352 353 354 355  | Next Page >

• How to correctly partion WD "Advanced Format" drive (4k sector size) with Disk Druid?

  - by FractalizeR

  Hello. Unaligned Advanced Format drives with sector size of 4k will have degraded performance. How do I make Disk Druid to partition drives correctly?

  Read the article

• hosts file seems to be ignored

  - by z4y4ts

  I have almost fresh Ubuntu desktop box. OS was installed two weeks ago and updated from karmic repositories. Last week I had no problems with DNS. But this week something had changed. I'm not sure what and when, and not sure whether I changed any configs. So now I have some really weird situation. According to logs name resolving should work normally. /etc/hosts 127.0.0.1 localhost test 127.0.1.1 desktop /etc/host.conf order hosts,bind multi on /etc/resolv.conf # Generated by NetworkManager search search servers obtained via DHCP nameserver 192.168.0.3 /etc/nsswitch.conf passwd: compat group: compat shadow: compat hosts: files mdns4_minimal [NOTFOUND=return] dns mdns4 networks: files protocols: db files services: db files ethers: db files rpc: db files netgroup: nis But if fact it is not. user@test ~ping test PING localhost (127.0.0.1) 56(84) bytes of data. [skip] Pinging is ok. user@test ~host test test.mydomain.com has address xx.xxx.161.201 But pure I suspect that NetworkManager might cause this misbehavior, but don't know where to start to check it. Any thoughts, suggestions?

  Read the article

• Is it possible to install all packages from an APT repository?

  - by Kristoffer Hagen

  Is it possible to install all packages from an APT repository? I know it is possible to do it manually, but then you would need to know all the package names, and I don't. Any suggestions? Thanks. Update: Well, you guys are going to kill me for this, but the reason for my madness is that I want to install all the packages from BackTrack into my Ubuntu installation. I really don't like the idea of having it in a VM and having a separate partition for it is even more out of the question. I know that the folks at BackTrack doesn't like it when people leech their repositories, but that's what you get for releasing open source software. Stupid? maybe.. A valid reason? probably not.. Do I still want it? Yes. Another edit: I have now given up on this as it seems impossible to get it to work even by manually installing packages.

  Read the article

• xorg DPMS off: keep VT from turning screen back on

  - by Metiu

  I have an embedded board with a small UPS. When AC power goes down, I need to turn off all power hungry devices in order to have a clean shutdown. First thing I do, is set DPMS to force powerdown, then go through the usual SIGTERM/SIGKILL/umount sequence. I have an Intel i915 Display adapter connected to an LVDS LCD panel. Unfortunately, when Xorg dies, Xserver or the VT code turn the LCD panel back on. I even tried working around it by directly poking the panel enable register in the Display chip, so that X doesn't know about it, but the panel goes back on when the VT comes back. Is there any "legal" way of keeping the display off? Thank you

  Read the article

• How to upgrade XBMC Live from 9.04.1 to 9.11?

  - by sunpech

  I've been unable to do a fresh install of XBMC Live 9.11 to my hard drive. Everytime it fails at the Install System step. But I am able to get XBMC Live 9.04.1 to install successfully. How do I upgrade XBMC Live 9.04.1 to 9.11? I understand that Ctrl+Shift+F2 brings up the command line, but what are the next set of commands to run?

  Read the article

• Surprising corruption and never-ending fsck after resizing a filesystem.

  - by Steve Kemp

  System in question has Debian Lenny installed, running a 2.65.27.38 kernel. System has 16Gb memory, and 8x1Tb drives running behind a 3Ware RAID card. The storage is managed via LVM. Short version: Running a KVM guest which had 1.7Tb storage allocated to it. The guest was reaching a full-disk. So we decided to resize the disk that it was running upon We're pretty familiar with LVM, and KVM, so we figured this would be a painless operation: Stop the KVM guest. Extend the size of the LVM partition: "lvextend -L+500Gb ..." Check the filesystem : "e2fsck -f /dev/mapper/..." Resize the filesystem: "resize2fs /dev/mapper/" Start the guest. The guest booted successfully, and running "df" showed the extra space, however a short time later the system decided to remount the filesystem read-only, without any explicit indication of error. Being paranoid we shut the guest down and ran the filesystem check again, given the new size of the filesystem we expected this to take a while, however it has now been running for 24 hours and there is no indication of how long it will take. Using strace I can see the fsck is "doing stuff", similarly running "vmstat 1" I can see that there are a lot of block input/output operations occurring. So now my question is threefold: Has anybody come across a similar situation? Generally we've done this kind of resize in the past with zero issues. What is the most likely cause? (3Ware card shows the RAID arrays of the backing stores as being A-OK, the host system hasn't rebooted and nothing in dmesg looks important/unusual) Ignoring brtfs + ext3 (not mature enough to trust) should we make our larger partitions in a different filesystem in the future to avoid either this corruption (whatever the cause) or reduce the fsck time? xfs seems like the obvious candidate?

  Read the article

• How to create an EFI System Partition?

  - by Alex Popov

  TL; DR How do I create an EFI system partition from scratch? How do I put the EFI firmware on it onces it is created? Long version I hava Toshiba T430 laptop. I received it with Windows 7 installed (but I think originally it has shipped with Windows 8). I installed Ubuntu on it, but deleted some partitions on the disk so that I ended up wiping out the Windows and only having Ubuntu. Among the deleted partitions was the EFI System partition. I discovered that Ubuntu now boots in Legacy mode (and not UEFI). I am trying to follow this guide on converting my Ubuntu installation from Legacy to UEFI. The problem - since there is no EFI partition whenever I choose from BIOS to boot using UEFI I cannot boot. That counts not only for the harddrive, but usb and DVD as well. I think this is logical - it expects an EFI partition and since it can't find it, it cannot continue booting futher, be it from HDD or DVD. So how do I recreate the EFI partition? The guide above says: Creating an EFI partition If you are manually partitioning your disk in the Ubuntu installer, you need to make sure you have an EFI partition set up. If your disk already contains an EFI partition (eg if your computer had Windows8 preinstalled), it can be used for Ubuntu too. Do not format it. It is strongly recommended to have only 1 EFI partition per disk. An EFI partition can be created via a recent version of GParted (the Gparted version included in the 12.04 disk is OK), and must have the following attributes: Mount point: /boot/efi (remark: no need to set this mount point when using the manual partitioning, the Ubuntu installer will detect it automatically) Size: minimum 100Mib. 200MiB recommended. Type: FAT32 Other: needs a "boot" flag. I had some trouble creating this partition: I boot from a live Ubuntu DVD, open GParted, create a 200MB partition and format it to FAT32. In GParted I cannot set the mount point and thus cannot set the bootflag. I didn't set the mount point in /etc/fstab since it's a live CD and fstab looked quite differently from what I expected compared to a normal boot. Anyway, I just didn't know what values to set. I booted again via the live DVD and then chose to install Ubuntu. I then created a partition with the mentioned criteria - mount point, 200MB, FAT32, boot flag. However, I continue to have this problem and I suppose it's because on that partition there is no EFI firmware, it's just an empty partition, which is suitable to have EFI firmware. So again, how do I create an EFI partition, which has the EFI software, so that the laptop can once again boot in UEFI mode?

  Read the article

• Force logout a user

  - by Mithun

  I When I logged into the machine as root and typed who to see which users are logged in, I found somebody else too logged in as root devuser pts/0 2011-11-18 09:55 (xxx.xxx.xxx.xxx) root pts/1 2011-11-18 09:56 (xxx.xxx.xxx.xxx) testuser pts/2 2011-11-18 14:54 (xxx.xxx.xxx.xxx) root pts/3 2011-11-18 14:55 (xxx.xxx.xxx.xxx) How can I force a root user at pts/3 to logout?

  Read the article

• How to configure mod_proxy_balancer to gracefully fail under high load

  - by bramp

  We have a system which has one Apache instance in front of multiple tomcats. These tomcats then connect to various databases. We balance the load to the tomcat with mod_proxy_balancer. Currently we are receiving 100 requests a second, the load on the Apache server is quite low, but due to database heavy operations on the tomcats, the load there is roughly 25% (of what I estimate they can handle). In a few weeks there is an event happening and we estimate that our requests will jump significant, maybe by a factor of 10. I'm doing everything I can do reduce the load on our tomcats, but I know we are going to run out of capacity, so I would like to fail gracefully. By this I mean, instead of trying to deal with too many connections which all timeout, I would like Apache to somehow monitor average response time, and as soon as the response time to Tomcat is getting above some threshold, I would like a error page displayed. This means that users who are lucky still get a page rendered quickly, and those who are unlucky get a error page quickly. Instead of everyone waiting far too long for their page, and eventually everyone timing out, and the database being swamped with queries which are never used. Hopefully this makes sense, so I was looking for suggestions on how I could achieve this. thanks

  Read the article

• extract recipient addresses from mailer-daemon messages

  - by Frank Peixoto

  I need generate and email to a mailing list administrator a list of email addresses that have generated a bounceback message, to allow him to clean up the mailing list. A separate mailbox will be dedicated to receiving those messages, so I need to extract the recipients that have bounced. What's the easiest way to extract those from the mail file for that special mailbox?

  Read the article

• Rsync backup - detect new directory and backup only from that directory

  - by Pracovek

  New cpanel daily backup is creating separate directories for daily backup. This creates problem when I try to user rsync to do an offsite backup since I would like to rsync only latest data. E.g. On backup server I have directory "backup" and on server, from which we are pulling backups I get directories 2013-11-07, 2013-11-08 etc in backup directory. If I backup /backup directory on the server it will use allot more space so I would like to backup only latest directory in backup directory, eg 2013-11-08. Is there a way to detect latest directory in backup directory and pass that directory name to rsync for backup ?

  Read the article

• Why won't dhclient use the static IP I'm telling it to request?

  - by mike

  Here's my /etc/dhcp3/dhclient.conf: request subnet-mask, broadcast-address, time-offset, routers, domain-name, domain-name-servers, domain-search, host-name, netbios-name-servers, netbios-scope, interface-mtu; timeout 60; reject 192.168.1.27; alias { interface "eth0"; fixed-address 192.168.1.222; } lease { interface "eth0"; fixed-address 192.168.1.222; option subnet-mask 255.255.255.0; option broadcast-address 255.255.255.255; option routers 192.168.1.254; option domain-name-servers 192.168.1.254; } When I run "dhclient eth0", I get this: There is already a pid file /var/run/dhclient.pid with pid 6511 killed old client process, removed PID file Internet Systems Consortium DHCP Client V3.1.1 Copyright 2004-2008 Internet Systems Consortium. All rights reserved. For info, please visit http://www.isc.org/sw/dhcp/ wmaster0: unknown hardware address type 801 wmaster0: unknown hardware address type 801 Listening on LPF/eth0/00:1c:25:97:82:20 Sending on LPF/eth0/00:1c:25:97:82:20 Sending on Socket/fallback DHCPREQUEST of 192.168.1.27 on eth0 to 255.255.255.255 port 67 DHCPACK of 192.168.1.27 from 192.168.1.254 bound to 192.168.1.27 -- renewal in 1468 seconds. I used strace to make sure that dhclient really is reading that conf file. Why isn't it paying attention to my "reject 192.168.1.27" and "fixed-address 192.168.1.222" lines?

  Read the article

• Change Gnome popup menus / combo boxes mouse click behaviour

  - by pingw33n

  Whe right clicking in windows that have popup menus you can hold mouse button, wait until popup appears and release above the desired item to click it. This is different from Windows that have popup appear only on mouse release. And it leads to accident menu item clicking sometimes. Looks like the issue is there: https://bugs.launchpad.net/ubuntu/+bug/320259, https://bugzilla.gnome.org/show_bug.cgi?id=575071. Is there's any way to change popup appearance time at least?

  Read the article

• Optimal file system type and mount options for an rsnapshot dedicated drive

  - by Nimmy Lebby

  We have an external USB 2 drive that we are using as a backup drive for our configuration. We use rsnapshot for the backups. It uses a few standard commands for managing snapshots: rm -rf: deletes expired snapshots mv: moves older snapshots down a slot cp -al: duplicates last snapshot to new slot rsync -a --delete --numeric-ids --relative: synchronizes new snapshot As you could see by the log below, the majority of the time is spent on the rm -rf and the cp -al steps: [25/Dec/2010:14:00:02] rsnapshot hourly: started [25/Dec/2010:14:00:02] echo 21012 > /var/run/rsnapshot.pid [25/Dec/2010:14:00:02] rm -rf /mnt/extdrive/snapshots/hourly.5/ [25/Dec/2010:14:15:48] mv /mnt/extdrive/snapshots/hourly.4/ /mnt/extdrive/snapshots/hourly.5/ [25/Dec/2010:14:15:48] mv /mnt/extdrive/snapshots/hourly.3/ /mnt/extdrive/snapshots/hourly.4/ [25/Dec/2010:14:15:48] mv /mnt/extdrive/snapshots/hourly.2/ /mnt/extdrive/snapshots/hourly.3/ [25/Dec/2010:14:15:48] mv /mnt/extdrive/snapshots/hourly.1/ /mnt/extdrive/snapshots/hourly.2/ [25/Dec/2010:14:15:48] cp -al /mnt/extdrive/snapshots/hourly.0 /mnt/extdrive/snapshots/hourly.1 [25/Dec/2010:14:23:32] rsync -a --delete --numeric-ids --relative /etc /mnt/extdrive/snapshots/hourly.0/sm4/ [25/Dec/2010:14:23:52] touch /mnt/extdrive/snapshots/hourly.0/ [25/Dec/2010:14:23:52] rm -f /var/run/rsnapshot.pid [25/Dec/2010:14:23:52] rsnapshot hourly: completed successfully My questions: I'm currently using ext4 for the filesystem. Maybe this is not the best choice from those available in Red Hat. Anyone have any recommendations that would speed up the process? The partition's mount options are sync,dirsync 1 2. Is there a way to optimize this since it's solely used for rsnapshot? Of course, reasoning would be greatly appreciated.

  Read the article

• Installing Midnight Commander from sources (no root privileges)

  - by ouroboros

  I tried to configure ./configure --prefix=/localfolder glib-2.26.1/ make make install but it fails at make stage. trying to configure mc-4.6.1/ and make doesn't obviously work. What are the steps I need to make in order to install midnight comander for my local user in a custom folder? Make for glib gives me these errors /usr/bin/msgfmt: found 2 fatal errors cp: cannot stat `test.mo': No such file or directory gmake[4]: *** [test.mo] Error 1 gmake[4]: Leaving directory `/remote/folder/mc/glib-2.26.1/gio/tests' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/remote/folder/mc/glib-2.26.1/gio' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/remote/folder/mc/glib-2.26.1/gio' gmake[1]: *** [all-recursive] Error 1 gmake[1]: Leaving directory `/remote/folder/mc/glib-2.26.1' gmake: *** [all] Error 2

  Read the article

• Configuration Help for Sendmail Required

  - by Vinayak Mahadevan

  Hi I need some help with respect to sendmail configuration. The basic problem is that I have some employees working from other places and they need access to their mail. So what I have done right now is whatever mails which are meant for them which are generated from within the company and collected by my internal mail server is bounced to an external mail server from where the employees access it. This is done through a email id on a different domain. This was working fine till I restricted the external mailing access for certain users using rulesets in sendmail.cf. Once I had put that in place only people who had external mailing rights could send mails to people outside the office. What I would like to know is that is there anyway where I can expose sendmail on two different ips and thereby configure everybody's email id to point to the same internal mail server using 2 different ips. one ip when inside the company and one ip outside the company. Is it possible that I have one static ip configured for both internal access and external access or is there any otherway it can be done with sendmail. Can anybody help me Sorry for the long post Regards Vinayak

  Read the article

• Postfix multiple checks

  - by xBlue

  I want to achieve the following with Postfix: Run all emails through a black list Allow any clients sending to a list of domains Allow some clients sending to any domain This is what I have: (postfix is on 10.0.8.0 and some of the senders are 10.0.8.0 and 10.0.9.0) mynetworks_style = subnet smtpd_recipient_restrictions = check_recipient_access sqlite:/etc/postfix/access-bl.query, check_client_access hash:/etc/postfix/trusted_clients, check_recipie nt_access hash:/etc/postfix/local_domains, reject_unauth_destination, permit So, right now the black list works. File /etc/postfix/trusted_clients contains who can send anywhere (3), file /etc/postfix/local_domains contains where you can send (2). Those two are fine, they return properly. My problem is getting all three working together. Not sure if it's an ordering issue. Currently sending a test from 10.0.9.17 and I get Relay access denied. If I add: mynetworks = 10.0.8.0/24 10.0.9.0/24 then anyone can send anywhere, so #2 is not working. Postfix version is 2.10 on Ubuntu 14.04. Any ideas?

  Read the article

• opennms postgres connection slow

  - by krisdigitx

  i am running the opennms application server on a physical server and the database on an ESXi VM. Recently the opennms webconsole has been very slow to load as such i deleted most of the events from the database table, now both servers have no load at all, and the psql connection from the application server to the database server is also very fast, but somehow opennms webconsole is still slow. this is the strace from the opennms process id: 18629 futex(0x2aaac77d8a84, FUTEX_WAIT_PRIVATE, 453, NULL <unfinished ...> 3015 futex(0x2aaabc4a2ee4, FUTEX_WAIT_PRIVATE, 323, NULL <unfinished ...> 10863 futex(0x2aaabbebaa94, FUTEX_WAIT_PRIVATE, 395, NULL <unfinished ...> 25260 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 10859 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 10982 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 3011 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 25260 futex(0x2aaae098fc28, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> 10982 futex(0x2aaac0eaf928, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> 3011 futex(0x2aaab0cb1728, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> 10859 futex(0x2aaac062c328, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> 25260 <... futex resumed> ) = 0 10982 <... futex resumed> ) = 0 3011 <... futex resumed> ) = 0 10859 <... futex resumed> ) = 0 25260 futex(0x2aaabc38b6b4, FUTEX_WAIT_PRIVATE, 443, NULL <unfinished ...> 10982 futex(0x2aaabc5d7b94, FUTEX_WAIT_PRIVATE, 99, NULL <unfinished ...> 3011 futex(0x2aaac7c55334, FUTEX_WAIT_PRIVATE, 183, NULL <unfinished ...> 10859 futex(0x2aaabbb8c9d4, FUTEX_WAIT_PRIVATE, 347, NULL <unfinished ...> 10846 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 10846 futex(0x2aaae9022428, FUTEX_WAKE_PRIVATE, 1) = 0 10846 futex(0x2aaabe0030b4, FUTEX_WAIT_PRIVATE, 251, NULL <unfinished ...> 20281 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 14100 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 2925 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 10843 <... futex resumed> ) = -1 ETIMEDOUT (Connection timed out) 20281 futex(0x2aaac7e93628, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> 14100 futex(0x2aaac04e8c28, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> 2925 futex(0x2aaaec085528, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> 10843 futex(0x2aaab20b0528, FUTEX_WAKE_PRIVATE, 1 <unfinished ...> and shows lots of connection timeout??? i think its the connection between the java application and database which is causing issues. any ideas how to troubleshoot this???

  Read the article

• Is there any way to recover files in /usr/local directory on Ubuntu?

  - by Ilya

  We are running Ubuntu server on VPS. Some files were removed accidentally by placing unnecessary space this command: rm -r /usr/local <directory to be deleted> I know, that in most cases this directory is used by packages to place some part of their content. Is there any where to recover deleted files and directories? I suppose, that theoretically it should be possible. Some software can look through the list of installed packages, check presence of files in file system and recover or reinstall corrupted packages if their file are missing in /usr/local.

  Read the article

• sudoer scheme for another web developer that retains my future control of a virtual server?

  - by Tchalvak

  Background: Virtual Private Server I have a virtual private server that I'm looking to host multiple websites on, and provide access to another web developer. I don't care about putting too many constraints on him, though I wouldn't mind isolating the site that he'll be developing from other sites on the server that I will develop. The problem: retain control Mainly what I want is to make sure that I retain control over the server in the future. I want to reserve the ability to create/promote/demote and other administrative functions that don't deal with web software. If I make him an admin, he can sudo su - and become root and remove root control from me, for example. I need him not to be able to: take away other admin permissions change the root password have control over other security/administrative functions I would like him to still be able to: install software (through apt-get) restart apache access mysql configure mysql/apache reboot edit web development configuration type files in /etc/ Other Standard Setups would be happily considered I've never really set up a good sudoers file, so simple example setups would be very useful, even if they're only somewhat similar to the settings that I'm hoping for above. Edit: I have not yet finalized permissions, standard, useful sudo setups are certainly an option, the lists above are more what I'm hoping I can do, I don't know that that setup can be done.

  Read the article

• How to recover broken MJPEG AVI file after a camera crash

  - by Jam K

  I was recording a video when my KODAK EASYSHARE M531 Digital Camera fell to the ground and turned off. When i turned it on nothing happened to the camera but the video file is somehow broken. Obviously the camera did not have time to convert and close it properly. $file myvid.avi myvid.AVI: data where it should be: another_vid.AVI: RIFF (little-endian) data, AVI, 640 x 480, ~30 fps, video: Motion JPEG, audio: uLaw (stereo, 12000 Hz)` I would like to fix my video which I believe needs to be converted to AVI. Any idea? VLC does not play the file and returns: could not determine type of stream

  Read the article

• Uninstall Git completely on Ubuntu?

  - by Millisami

  I installed Git on Ubuntu Lucid (latest) manually as following. cd ~/tmp wget http://kernel.org/pub/software/scm/git/git-1.7.0.6.tar.gz tar -xzvf git-1.7.0.6.tar.gz cd git-1.7.0.6.tar.gz ./configure sudo make sudo make install Now, how can I completely uninstall it?

  Read the article

• How to copy directories using debugfs?

  - by STM

  The debugfs manpage gives the impression that the command 'rdump . .' will recursively copy all files found on the specified filesystem from the debugfs cwd to the native filesystem's cwd. Instead I seem to receive a syntax error, and no copy is initiated? These are the commands I run: cd /path/to/transfer/destination debugfs /dev/sda1 -R rdump . . My task is to copy the entire contents of a clean yet unmountable USB storage device to its host machine's HD. The host machine does not support the inode size used by the USB device's filesystem (256) and its software is not upgradeable, so my intention was to use debugfs to transfer the files. If anyone has any other suggestions for this task I'd be grateful.

  Read the article

• Strange port forwarding problem

  - by rAyt

  I've got a strange port forwarding problem. The port forwarding to my internal webserver (10.0.0.10 on Port 80) works without a problem but the port forwarding to a windows server (10.0.0.15) on port 3389 doesn't work. The port 3389 is open. Any ideas? thanks! #!/bin/sh IPTABLES="/sbin/iptables" $IPTABLES --flush $IPTABLES --table nat --flush $IPTABLES --delete-chain $IPTABLES --table nat --append POSTROUTING --out-interface eth0 -j MASQUERADE $IPTABLES -t nat -A PREROUTING -p tcp -i eth0 -d 188.40.XXX.XXX --dport 3389 -j DNAT --to 10.0.0.15:3389 $IPTABLES -t nat -A PREROUTING -p tcp -i eth0 -d 188.40.XXX.XXX --dport 80 -j DNAT --to 10.0.0.10:80 $IPTABLES -t nat -A PREROUTING -p tcp -i eth0 -d 188.40.XXX.XXX --dport 222 -j DNAT --to 10.0.0.10:22 $IPTABLES --append FORWARD --in-interface eth1 -j ACCEPT

  Read the article

• Is it reasonable that a random disk seek & read costs ~16ms?

  - by fzhang

  I am frustrated about the latency of random reading from a non-ssd disk. Based on results from following test program, it speeds ~16 ms for a random read of just 512 bytes without help of os cache. I tried changing 512 to larger values, such as 25k, and the latency did not increase as much. I guess it is because the disk seek dominates the time. I understand that random reading is inherently slow, but just want to be sure that ~16ms is reasonable, even for non-ssd disk. #include <sys/stat.h> #include <sys/time.h> #include <sys/types.h> #include <sys/unistd.h> #include <fcntl.h> #include <limits.h> #include <stdio.h> #include <string.h> int main(int argc, char** argv) { int fd = open(argv[1], O_RDONLY); if (fd < 0) { fprintf(stderr, "Failed open %s\n", argv[1]); return -1; } const size_t count = 512; const off_t offset = 25990611 / 2; char buffer[count] = { '\0' }; struct timeval start_time; gettimeofday(&start_time, NULL); off_t ret = lseek(fd, offset, SEEK_SET); if (ret != offset) { perror("lseek error"); close(fd); return -1; } ret = read(fd, buffer, count); if (ret != count) { fprintf(stderr, "Failed reading all: %ld\n", ret); close(fd); return -1; } struct timeval end_time; gettimeofday(&end_time, NULL); printf("tv_sec: %ld, tv_usec: %ld\n", end_time.tv_sec - start_time.tv_sec, end_time.tv_usec - start_time.tv_usec); close(fd); return 0; }

  Read the article

< Previous Page | 344 345 346 347 348 349 350 351 352 353 354 355  | Next Page >