Search Results

Search found 24623 results on 985 pages for 'linux'.

facebook twitter digg google delicious technorati stumbleupon myspace wordpress linkedin gmail igoogle windows live tumblr viadeo yahoo buzz yahoo mail yahoo bookmarks favorites email print

Page 362/985 | < Previous Page | 358 359 360 361 362 363 364 365 366 367 368 369  | Next Page >

  • Changing Corosync/Heartbeat pair's active node based on MySQL/Galera cluster state

    - by Hace
    Background I'm planning on building a High Availability "cluster" for our Zabbix instance by placing two physical servers in one server room and two in another server room. In each server room one of the physical servers will run Zabbix on RHEL and the other will run Zabbix's MySQL database, also on RHEL. I'd prefer synchronous replication for the MySQL nodes so I'm planning on using Galera in a master-slave configuration. The Zabbix instances on the two Zabbix servers would be controlled by Heartbeat/Corosync (although Red Hat Cluster Suite is also an option...) If the Zabbix server in Server Room A goes down, the one in Server Room B becomes active (and vice versa). Ditto for the MySQL servers/instances. If either of those cases happen, however, the connection between the Zabbix server and the MySQL server becomes significantly slower as ti has to travel over WAN. Question Is it possible to configure the Heartbeat/CoroSync pair to instruct the MySQL/Galera cluster to change the master node to switch to (if available) the one that's in the server room as the active Heartbeat/Corosync -node and (more challengingly) is it possible to do the same in the other direction, i.e have the Galera cluster change the active Heartbeat/CoroSync server to be in the same room as the active MySQL master server in case of a failover in over to avoid unnecessary WAN transfers between the application and its DB? Theories Most likely I can get CoroSync to run something that'd log in to one of the DB nodes to change the MySQL/Galera master but I don't know if it's really possible to do anything similar in the other direction in Galera. Is it possible to define a "service" in CoroSync/Heartbeat so that both the service and its MySQL service would migrate as one if possible. Using the DB server that's behind WAN should still be a better option to DB downtime. Am I just using too many tools to solve a problem that'd be far simpler with something else?

    Read the article

  • Google Search w/ Chrome Incognito w/ Gnome Do

    - by jrc03c
    I've installed Google Chrome as my default browser in Ubuntu, and recently installed Gnome Do and enabled the Google Search plugin. The Google Search from Gnome Do works exactly as expected but for one thing: Chrome (which is typically set to open in "incognito" mode) does not open in "incognito" mode. The shortcuts on my desktop, taskbar, and menus all have the --incognito flag attached (which works just fine), but the browser refuses to open in this mode when launched from Gnome Do. Any suggestions? Also, please note the settings for the Google Search plugin in Gnome Do: It's obvious that Gnome Do just passes the Google Search blindly to the default browser. In other words, there are no configurable settings specifically for Chrome. Any thoughts?

    Read the article

  • Have an unprivileged non-account user ssh into another box?

    - by Daniel Quinn
    I know how to get a user to ssh into another box with a key: ssh -l targetuser -i path/to/key targethost But what about non-account users like apache? As this user doesn't have a home directory to which it can write a .ssh directory, the whole thing keeps failing with: $ sudo -u apache ssh -o StrictHostKeyChecking=no -l targetuser -i path/to/key targethost Could not create directory '/var/www/.ssh'. Warning: Permanently added '<hostname>' (RSA) to the list of known hosts. Permission denied (publickey). I've tried variations using -o UserKnownHostsFile=/dev/null and setting $HOME to /dev/null and none of these have done the trick. I understand that sudo could probably fix this for me, but I'm trying to avoid having to require a manual server config since this code will be deployed on a number of different environments. Any ideas? Here's a few examples of what I've tried that don't work: $ sudo -u apache export HOME=path/to/apache/writable/dir/ ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=path/to/apache/writable/dir/.ssh/known_hosts -l deploy -i path/to/key targethost $ sudo -u apache ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=path/to/apache/writable/dir/.ssh/known_hosts -l deploy -i path/to/key targethost $ sudo -u apache ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -l deploy -i path/to/key targethost Eventually, I'll be using this solution to run rsync as the apache user.

    Read the article

  • CentOS - dual boot from new partition

    - by Dima
    I need to install two copies of the CentOS 5.5 (bank A and bank B) on different partitions of the same hard disk and install grub boot loader to another partition (visible from both banks). The boot loader should redirect the boot menu to bank A or bank B (according to the configuration). The new partition is mounted to /common_partition and grub is installed on it using following command: grub-install /dev/hda In the new partition I'm created the following menu.lst file: title BOOTCONTROL REDIRECT : PLEASE WAIT root (hd0,1) configfile /boot/menu.lst boot On my setup: both partitions (bank A and bank B) are primary and grub is installed on MBR. The problem is: but the new boot loader (on common_partition) did not load. What wrong on my configuration?

    Read the article

  • Subversion multi checkout post-commit hook?

    - by FLX
    The title must sound strange but I'm trying to achieve the following: SVN repo location: /home/flx/svn/flxdev SVN repo "flxdev" structure: + Project1 ++ files + Project2 + Project3 + Project4 I'm trying to set up a post-commit hook that automatically checks out on the other end when I do a commit. The post-commit doc explicitly lists the following: # POST-COMMIT HOOK # # The post-commit hook is invoked after a commit. Subversion runs # this hook by invoking a program (script, executable, binary, etc.) # named 'post-commit' (for which this file is a template) with the # following ordered arguments: # # [1] REPOS-PATH (the path to this repository) # [2] REV (the number of the revision just committed) So I made the following command to test: REPOS="$1" REV="$2" echo "Updated project $REPOS to $REV" However when I edit files in Project1 for example, this outputs "Updated project /home/flx/svn/flxdev to 1016" I'd like this to be: "Updated project Project1 to 1016" Having this variable allows me to specify to do different actions per project post-commit. How can I specify the project parameter? Thanks! Dennis

    Read the article

  • Subversion multi checkout post-commit hook?

    - by FLX
    The title must sound strange but I'm trying to achieve the following: SVN repo location: /home/flx/svn/flxdev SVN repo "flxdev" structure: + Project1 ++ files + Project2 + Project3 + Project4 I'm trying to set up a post-commit hook that automatically checks out on the other end when I do a commit. The post-commit doc explicitly lists the following: # POST-COMMIT HOOK # # The post-commit hook is invoked after a commit. Subversion runs # this hook by invoking a program (script, executable, binary, etc.) # named 'post-commit' (for which this file is a template) with the # following ordered arguments: # # [1] REPOS-PATH (the path to this repository) # [2] REV (the number of the revision just committed) So I made the following command to test: REPOS="$1" REV="$2" echo "Updated project $REPOS to $REV" However when I edit files in Project1 for example, this outputs "Updated project /home/flx/svn/flxdev to 1016" I'd like this to be: "Updated project Project1 to 1016" Having this variable allows me to specify to do different actions per project post-commit. How can I specify the project parameter? Thanks! Dennis

    Read the article

  • formatting before md device creation in RAID5

    - by kumar
    consider you are creating a raid5 device with three drives. mdadm --create /dev/md0 --leve=5 --raid-disk=3 /dev/sda1 /dev/sdb1 /dev/sdc1 After issuing this command , I can see the progress of md device creating using cat /proc/mdstat. During the progress ITSELF, can I create a file ssytem partition say ext2 on md0 device like: mkfs.ext2 /dev/md0. Actually I am able to create this and want to confirm whether doing this before 100% completion of md device creation is CORRECT?

    Read the article

  • Nginx and automatic updates

    - by Desmond Hume
    I'm on Ubuntu 12.04.1 with unattended-upgrades configured for automatic security updates, and I installed Nginx by first adding deb http://nginx.org/packages/ubuntu/ lucid nginx deb-src http://nginx.org/packages/ubuntu/ lucid nginx to /etc/apt/sources.list file, just as was suggested by the official wiki, and then by sudo apt-get update sudo apt-get install nginx which installed Nginx with all the standard modules. But now I think I could make good use of one or two of the Nginx optional modules, like the gzip precompression module or some security-related one. So far, I see two ways of adding an optional module to Nginx, one is compiling and installing from the source code and the other is described in this article. So, which of the ways should I choose so that automatic updates still run for and apply to Nginx and its optional modules? Or should I create a cron job with a command/script specific for Nginx instead of using unattended-upgrades utility? Can I choose between volume updates and security-only updates to be automatically applied to the standard and optional modules? And finally, is there a possibility to automatically update Nginx's modules on the fly (without any connections having been dropped), like the documentation suggests it's possible with sudo kill -USR2 $( cat /run/nginx.pid ) P.S. Actually I'm not certain if unattended-upgrades utility would automatically update the standard modules in the first place, not enough time has passed since Nginx was installed to say for sure.

    Read the article

  • winbind failing after a semi-random amount of time

    - by The Digital Ninja
    I have winbind set up to authenticate to our AD for samba shares. This is the third such server, and the only one having any issues. It seems after a random amount of time samba shares will just stop working. Winbind processes seem to be running but restarting them seems to fix the issue for a while. Looking at the logs have been kind of hit an miss and I don't know exactly when it fails. One interesting thing is that it seems to be pulling from another domain controller that it shoudlnt. I censored out the domain name in this example. But isnt there some way to block authentication to a domain? I'm not sure if this is a symptom or a cause of the issue. [2010/10/18 08:02:10, 0] winbindd/winbindd_cache.c:initialize_winbindd_cache(2577) initialize_winbindd_cache: clearing cache and re-creating with version number 1 [2010/10/18 09:15:54, 1] libsmb/clikrb5.c:ads_krb5_mk_req(686) ads_krb5_mk_req: krb5_get_credentials failed for [email protected] (Cannot find KDC for requested realm) [2010/10/18 09:15:54, 1] libsmb/cliconnect.c:cli_session_setup_kerberos(624) cli_session_setup_kerberos: spnego_gen_negTokenTarg failed: Cannot find KDC for requested realm [2010/10/18 09:15:54, 0] lib/util_sock.c:write_data(1139) write_data: write failure. Error = Connection reset by peer [2010/10/18 09:15:54, 0] libsmb/clientgen.c:write_socket(242) write_socket: Error writing 108 bytes to socket 18: ERRNO = Connection reset by peer [2010/10/18 09:15:54, 0] libsmb/clientgen.c:cli_send_smb(290) Error writing 108 bytes to client. -1 (Connection reset by peer)

    Read the article

  • Calculate minimum ext3 partition size for certain amount of data

    - by Daniel Beck
    These following ext3 partitions contain identical data. As we can see, the larger the partition size, the more space is required for the same files: Filesystem 1K-blocks Used Available Use% Mounted on /dev/loop11 3965777 561064 3199964 15% [...] /dev/loop19 573029 543843 29186 95% [...] Filesystem Size Used Avail Use% Mounted on /dev/loop11 3.8G 548M 3.1G 15% [...] /dev/loop19 560M 532M 29M 95% [...] Filesystem Inodes IUsed IFree IUse% Mounted on /dev/loop11 1024000 1656 1022344 1% [...] /dev/loop19 1024000 1656 1022344 1% [...] I start with a partition of fixed size that possibly wasted a lot of space and I want to create a partition that is able to hold that data but with (almost) minimal size. How can I reliably calculate that minimal partition size needed for storing a certain amount of data? The amount of data changes over time, and I need to automate these calculations.

    Read the article

  • Cannot install CentOS 6.5 using UEFI usb boot

    - by Vaindil
    I am trying to dual-boot CentOS 6.5 on my desktop that is currently running Windows 8.1. I have two storage devices: an SSD that has my Windows installation, and an HDD that has all of my data. Both are formatted using GPT, and Windows boots using UEFI. I used the CentOS 6.5 live DVD (CentOS-6.5-x86_64-LiveDVD.iso) to create an EFI-bootable flash drive (it does boot properly in EFI mode). I receive an error, however, when CentOS is booting (error is below). I have a 6.4 boot DVD which boots as expected, but it does not boot in UEFI mode and therefore doesn't play nicely with my Windows installation (I have no way to access it, even using rEFInd or any other similar tools). What do I need to do to get the device to boot properly in UEFI mode? Kernel panic - not syncing: Attempted to kill init! Pid: 1, comm: init Not tainted 2.6.32-431.el6.x86_64 #1 Call Trace: [<ffffffff815271fa>] ? panic+0xa7/0x16f [<ffffffff81077622>] ? do_exit+0x862/0x870 [<ffffffff8118a865>] ? fput+0x25/0x30 [<ffffffff81077688>] ? do_group_exit+0x58/0xd0 [<ffffffff81077717>] ? sys_exit_group+0x17/0x20 [<ffffffff8100b072>] ? system_call_fastpath+0x16/0x1b drm_kms_helper: panic occurred, switching back to text console

    Read the article

  • MySQL Master - Master Broken

    - by Recc
    I've Inherited a Mysql master master system, I've noticed the second master (lets call it slave from now on as it's running on a 'slave' machine) stopped getting its db's updated. I saw that Master: Slave_IO_Running: Yes Slave_SQL_Running: Yes Slave: (with an error I truncated) Slave_IO_Running: Yes Slave_SQL_Running: No Last_Errno: 1062 Last_Error: Error 'Duplicate entry '3' for key 'PRIMARY'' on [...] I don't know what caused it to process considering we cant get duplicate there. What's important is to resume normal operations; Right now I've stop slave; on the Master and stop slave; on the Slave because I saw that if I change records on the Slave the changes Do Get Propagated to Master which is in active use. How do I: Force sync EVERYTHING from master to slave without affecting data on master? Then hopefully have slave pickup replication as usual? UPDATE OK I Tried deleting all tables on slave then it complained in that error section that the 'table' doesnt exist. So i made a no data dump of Master, and made sure I have only empty tables in Secondary (slave). I start slave; on slave BUT now it's complaining about bloody alter table statements for instance: Last_Errno: 1060 Last_Error: Error 'Duplicate column name [...] Query: 'ALTER TABLE [...] How to skip the fracking alter statements I just want to replicate the bloody data and be done with it, my tables have the lates changes already FFS and now its complaining about changes made after the replication seized weeks ago How do I reset the log or something? OUTSTANDING Why would this start happening? The "Secondary" is propagating to "Primary". "Primary" is not propagating to "Secondary". But any fixes I tried to do left it in the same state Yes-Yes Yes-No with same Last_Error. I think around that time the server was taken off the network, could that confuse MySQL in some way?

    Read the article

  • Xen domain migration locking problem

    - by brodie
    I am trying to live migrate a VM (domain) between two Xen servers. I have xen locking (xend-domain-lock = yes) configured with a ocfs2 shared storage between them. This locking is working fine. If I try to start up the VM on the secondary server it refuses to start (which is correct). The problem I am having is when trying to do live migration, it seems like it is trying to remove the lock twice. The first lock it removes is for "domain test", the second is for "migrating-test" which does not exist. Should their be a lock for this "migrating-test" VM? These are the relevant options in the xen config file: (xend-relocation-server yes) (xend-relocation-port 8002) (xend-relocation-address '') (xend-relocation-hosts-allow '') (xend-domain-lock yes) (xend-domain-lock-path /var/lib/xen/lock) This is the section of the log: [2010-06-10 10:45:57 14488] DEBUG (XendDomainInfo:4054) Releasing lock for domain test [2010-06-10 10:45:57 14488] INFO (XendCheckpoint:474) SUSPEND shinfo 000c6ceb [2010-06-10 10:45:57 14488] INFO (XendCheckpoint:474) delta 21ms, dom0 95%, target 0%, sent 57Mb/s, dirtied 173Mb/s 111 pages 4: sent 111, skipped 0, delta 6ms, dom0 100%, target 0%, sent 606Mb/s, dirtied 606Mb/s 111 pages [2010-06-10 10:45:57 14488] INFO (XendCheckpoint:474) Total pages sent= 131295 (0.99x) [2010-06-10 10:45:57 14488] INFO (XendCheckpoint:474) (of which 0 were fixups) [2010-06-10 10:45:57 14488] INFO (XendCheckpoint:474) All memory is saved [2010-06-10 10:45:57 14488] INFO (XendCheckpoint:474) Save exit rc=0 [2010-06-10 10:45:57 14488] INFO (XendCheckpoint:123) Domain 22 suspended. [2010-06-10 10:45:57 14488] DEBUG (XendDomainInfo:2757) XendDomainInfo.destroy: domid=22 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:2227) Destroying device model [2010-06-10 10:45:58 14488] INFO (image:567) migrating-test device model terminated [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:2234) Releasing devices [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:2247) Removing vif/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:1137) XendDomainInfo.destroyDevice: deviceClass = vif, device = vif/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:2247) Removing vkbd/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:1137) XendDomainInfo.destroyDevice: deviceClass = vkbd, device = vkbd/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:2247) Removing console/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:1137) XendDomainInfo.destroyDevice: deviceClass = console, device = console/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:2247) Removing vbd/51712 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:1137) XendDomainInfo.destroyDevice: deviceClass = vbd, device = vbd/51712 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:2247) Removing vfb/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:1137) XendDomainInfo.destroyDevice: deviceClass = vfb, device = vfb/0 [2010-06-10 10:45:58 14488] DEBUG (XendDomainInfo:4054) Releasing lock for domain migrating-test [2010-06-10 10:45:59 14488] ERROR (XendDomainInfo:4070) Failed to remove unmanaged directory /var/lib/xen/lock/b01515ae-9173-03cb-0cb7-06f3dfbede8b.

    Read the article

  • Is it possible to use SELinux MCS permissions with Samba?

    - by Yuri
    Created a user1: adduser --shell /sbin/nologin --no-create-home user1 passwd user1 smbpasswd -a user1 smbpasswd -e user1 semanage login -a -s "unconfined_u" -r "s0-s0:c0" user1 Added a category c0 for the folder ./123 inside the Samba share chcat s0:c0 /share/123/ After that the user1 can't go into this folder: type=AVC msg=audit(1332693158.129:48): avc: denied { read } for pid=1122 comm="smbd" name="123" dev=sda1 ino=786438 scontext=system_u:system_r:smbd_t:s0 tcontext=unconfined_u:object_r:samba_share_t:s0:c0 tclass=dir But if remove the c0 category: restorecon -v /share/123/ user1 opens folder with no problem. Is I'm doing something wrong or Samba doesn't support SELinux MCS? Have installed on CentOS 6.2 are: samba3.i686 3.6.3-44.el6 @sernet-samba selinux-policy.noarch 3.7.19-126.el6_2.10 @updates selinux-policy-targeted.noarch 3.7.19-126.el6_2.10 @updates

    Read the article

  • what is ranlib?

    - by Ying
    I have been using a MacOSX system for a while, but just only recently started poking into the guts. I found a guide telling me to run 'sudo ranlib /usr/local/lib/libjpeg.a'(installing libjpeg). I have read the ranlib manual, and tried looking online on it. I simply don't understand. What resources do I need to look up to learn more, or can someone give a concise explanation on its use? Thanks in advance!

    Read the article

  • Nginx Forward SSL for single site

    - by Will.brown
    I have a nginx server setup and it works fine for http however i would like to bypass the proxy for https connection. I want it so that when someone goes to my ip https:// ip1 (Nginx server) it bypasses ngix and forwards all traffic to https:// ip2(webserver) i do not need ngix to do this for any ssl website just one particular website. SO Client to https:// ip1 to https:/ /ip2 to https:// ip1 to client pc I just want the nginx to not intercept the connection and forward it on and on return forward the connection to client Im guessing i do this by nat mascarade buy not exactly sure how to do it and if i will need to tell nginx to ignore ssl aswell can someone help me please this has gone me stuck

    Read the article

  • Basic clarification about Limited FTP/sFTP users

    - by mattewre
    I would like to get some clarification about the correct way to create limited users to access to my VPS user as WEBSERVER with Nginix. I'm used to NOT install FTP and access via SFTP only. It is ok for every set up? this is what I usually do from to create a limited user called "admin" that should be able to have access via SFTP to the folder with the website data mkdir -p /var/www/mysite.com/ adduser admin adduser admin www-data chown -R root:root /var/www chmod -R 755 /var/www chmod -R 755 /var/www/mysite.com chown -R admin:www-data /var/www/mysite.com/ It seems not to be the correct way, I always have problems with permission when I upload some files (for example with Wordpress in general). I would like to create an user that does work exactly as the one that the "provides" give to their client when they buy an Hosting service (that is a FTP, I would prefer SFTP access). It is for personal user, but I think that a limited user is a lot safer to use then the "root" via SFTP.

    Read the article

  • Added user to CentOS, Updated sshd_config with AllowUsers, Login denied

    - by Gregg
    CentOS 5.3. I can SSH into the system as root just fine. Added a user and set their password. They have shell access (/bin/bash). I can su to the account from root just fine. I updated /etc/ssh/sshd_config with: AllowUsers myNewUser And restarted sshd: /etc/init.d/sshd restart When trying to ssh into the server with the new user, I get a permission denied. And yes, I've double and triple checked that I am using the correct password. Any help is appreciated.

    Read the article

  • How to know how many files a process opened efficiently?

    - by Victor Lin
    I knows I can use the command lsof -p xxxx | wc -l to know the count of opened files op a process, it works, but however, it is just too inefficient. I have some server process which have too many socket files, the wc -l method never return the result. So, what is the efficient way to know how many files opened on a process? Thanks.

    Read the article

  • How to "ignore" username and password prompt in net use

    - by Mattisdada
    I have at the moment a logon.cmd script, that I'm using to map network drives to the users profile. It looks like this: ::Onboarding net use m: /delete net use m: \\BOB\onboarding ::Bookings net use n: /delete net use n: \\BOB\bookings ::Accounts net use j: /delete net use j: \\BOB\accounts It works fine up until it gets up to a folder that the current user cannot access, it then asks for a username and password instead of erroring and continuing. Notes: This very script used to work on another Samba PDC network, but I've moved it over to another server (Still Samba PDC) and now its breaking. Is there anyway for it to ignore the username/password prompt and just continue?

    Read the article

  • How can I reset the permissions of /bin /boot /etc and /dev to orignal owner, Ubuntu?

    - by Camsoft
    I accidentally changed the ownership of the /bin, /boot, /etc and /dev recursively to nobody:nogroup using chown when I misplaced a forward slash! How can I resort the original file ownerships? I've managed to get them all to root:root but I'm not sure if all the files should be owned by root and if this will break something? Is they are option to fix file permissions like there is in OS X? Help!

    Read the article

facebook twitter digg google delicious technorati stumbleupon myspace wordpress linkedin gmail igoogle windows live tumblr viadeo yahoo buzz yahoo mail yahoo bookmarks favorites email print

< Previous Page | 358 359 360 361 362 363 364 365 366 367 368 369  | Next Page >