Ubuntu Servers :: Passing Mount Options To Autofs5 Via Init Script?
May 7, 2010
At work we use autofs4, and we also take advangate of the -DOSNMAE=blah and -DOSREL=blah for our automount maps. We're moving some systems to autofs5 and I can't for the life of me figure out how to pass these options properly. There's no 'localoptions' in the init script anymore as there was for autofs4. I've tried adding the flags to the OPTIONS variable in /etc/default/autofs. That adds them to the global autofs process but then nothing in the automounter mounts.
I've been dual booting 10.10 with Windows7 for about a month. Today is the first time I've encountered a serious problem.
This morning, nothing functioned properly after trying to open several programs. The computer seemed to be "frozen", although the mouse was working fine.
I decided to reboot, but then encountered an even bigger problem.
It failed to boot and got this message: no init found. try passing init= bootarg
The problem now is that it requires a Live CD session and I keep getting this: GLib-WARNING **: getpwuid_r(): failed due to unknown user id (0)
In case it matters, I didn't install 10.10 from an ISO, I just upgraded from 10.04.
Ubuntu 9.10 will not boot! System froze this morning, I restarted and it is now failing to boot. Starts loading grub and I get this message:
mount: mounting /dev/disk/by-uuid/04aa3697-7bc0-45b5-b86a-77a1e6534bd5 on /root failed: invalid argument mount: mounting /sys on /root/sys failed: no such file or directory mount: mounting /dev on /root/sys failed: no such file or directory
[code]....
I booted with 9.04 LiveCD discovered the drive could not be mounted-ran fsck -ln and it told me the drive has no valid partition table. I have had intermittent problems mounting flash drives before this, so I'm kind of worried it might be a hardware issue.Also have files on that drive I would rather not lose, so reinstalling is hopefully a last resort.
While I was using my computer a few days ago, the terminal stopped working properly, so I tried to reboot, and when it started up again it wouldn't boot and said "no init found. try passing init=bootarg"
This has happened twice before, so I really need to figure out what keeps happening, otherwise I can't continue to use linux. i reinstalled both times before. i think that this is caused by a process that prevents me from using the hard drive, because when I try to check the disk in the terminal or in gparted, it says Device or resource busy while trying to open /dev/sda1. Filesystem mounted or opened exclusively by another program?
Also, in the disk utility, in the lower right corner of the filesystem it has a spinning "loading wheel".(i'm not sure if that means anything)
I am using ubuntu 10.10, but am not sure what kernel I am using, but i tried a few different kernel options(there's three of them at start up). safe mode does not work either.
Linux box info: root@mytestbox:~# uname -a Linux mytestbox 2.6.32-30-generic-pae #59-Ubuntu SMP Tue Mar 1 23:01:33 UTC 2011 i686 GNU/Linux
Windows box info: Windows Server 2008 SP2 Enterprise I've verified via --verbose output that mount.cifs is indeed processing the passed on options.
root@mytestbox:~# mount -t cifs //10.1.1.10/Test /root/testwin --verbose -o credentials=/root/testcreds,rw,nocase,noperm,noacl,nounix,noserverin o,iocharset=utf8,file_mode=0777,dir_mode=0777
[Code]...
Yet, when I type mount all it reports is (rw,mand). The share works just fine, and I can see the masking (all files are showing as rwxrwxrwx as expected etc) but mount is not listing the options?!
Is this normal expected behavior? Is there a bug report on this? I've google'd to the best of my capabilities and could not locate any such information which is why I decided to hit the forums prior to filing a bug.
I have a problem with ssh, in that it's extremely slow when using putty to connect from Windows. A bit of googling suggested that I should use -u0 as a startup option since there's no DNS entry for this machine.
So, at the risk of sounding stupid, how do I put this options in to the /etc/init.d/ssh file? I tried adding it in the the "set" part but got an error, tried adding another "set" line and got an error and tried adding it to the first command there, but also got an error! Where does it go?
I'm trying to set up a hot spot using a squid proxy server for network logging of accessed website URLs. I really would like to pass all web traffic through a custom PHP program.Like when a computer tries to access www.google.com it first goes through the proxy server, and then the proxy server sends back a URL like URL...which is then accessed by the end-user's computer.. I'd eventually set up the php program to redirect the user to the requested URL after my application executes.Is this at all possible to do, with or without squid?
I hope anyone of you can help me build autofs5_5.0.4-3.2_amd64.deb and autofs5-ldap_5.0.3-3_amd64.deb-packages with a modified source for debian testing. The trouble is, the packages from the repositories do not work with our ldap server. I know that using the configure option DISABLE_MOUNT_LOCKING might solve it. But I (a) fail to build the modules and (b) am not sure if I put the compile option in the right place. I dont have much experience with building deb packages, maybe someone can give me a hint.
If there is a way in autofs5 to search for a key in multiple OU's? I am migrating our existing NIS infrastructure to LDAP. Over the years the NIS environment has evolved (not necessarily for the better). I have a back end database and each NIS master extracts specific elements from the DB to create the NIS maps.
On my Linux servers I use a program map that prgramatically searches multiple NIS maps. On Solaris I have an auto.home file that will search auto_home and auto_service ( Home has all home directories associated with people (i.e., in the passwd file) auto_service contain directories that are accessed as /home/.... but are not tied to a user account.
I also have regional data and unfortunately, my back end storage is not consistant so /apps/perl in one region will resolve to regionA:/path1/path2/perl and in another region it may resolve to regionB:/path33/path2/perl. Can this be done?
I have installed FC13 on my laptop and set it up as a development server. Here is my issue when passing variable from one page to the next it gets lost. My PHP includes work DB connect string works from the include.
I run XBMC media center software which is built on a minimal Ubuntu install. I was running a version built on Karmic and I had the following line in my /etc/init.d/rc.local and it always ran without a hitch:
Code:
mount -t cifs -o file_mode=0777,dir_mode=0777 //192.168.1.20/disk7/xbmc_thumbs/Thumbnails /home/kevin/.xbmc/userdata/Thumbnails
Recently, I upgraded to a version built on Lucid and now that fails to create the mount on boot up. Here is the entire contents of the file:
Code:
#! /bin/sh ### BEGIN INIT INFO # Provides: rc.local # Required-Start: $remote_fs $syslog $all
I'm running XBMC media center which is built on Ubuntu and I'm trying to mount a network share, but I can't seem to automate it. If I manually run:
Code:
mount -t cifs -o file_mode=0777,dir_mode=0777 //192.168.1.20/disk7/xbmc_thumbs/Thumbnails /home/kevin/.xbmc/userdata/Thumbnails
The mount is created and everything works fine. But obviously the mount eill be lost on the next boot. A few months back, when I was running a previous version of XBMC and first instructed on how to do this, I was told to put that command in /etc/init.d/rc.local so the mount would automatically be created at boot. I did and it worked. The other day I upgraded to a new XBMC build (which is built on the newer Lucid Ubuntu) and while the same command creates the mount, putting it in /etc/init.d/rc.local does not create it on boot. Someone suggested the fstab was the better place to create the mount. So I inserted the following in /etc/fstab:
I had a LiveUSB of CentOS 5.5, so I decided the install it.With no installer,I just copied the files to my hard drive.his was in a multi boot with Windows 7, Ubuntu 9.10, and FreeDOS.I updated GRUB2 and it detected CentOS. I loaded my entry and it failed to mount the root filesystem.I took the initrd0 file from the LiveUSB syslinux folder and added that ramdisk to the entry. Now it finds the root filesystem (/dev/sda9 as Ext3).But it fails shortly after loading /sbin/init. It talks about an init error where it says "File not found!!!".The previous lines involved umounting old filesystems, like /dev.
I have a bunch of nfs mounts from my MythTV backend to my netbook, both running 9.10. Due to the lousy Atheros wireless driver, it typically loses connection at least every half hour. The nfs mounts are mounted soft,intr, so programs should get an error trying to talk to a server they can't contact, and the request should be interruptable. Instead, they just lock up, can't be interrupted, and the netbook can't even shut down after the network failure because it hangs trying to clean up before shutting down.
So do these nfs options actually work in Ubuntu? 'mount' shows the nfs directory mounted with the correct options, so it doesn't seem to be any kind of configuration problem... it just doesn't do what I'm telling it to do.
Have a 1TB external USB hard drive I want to use on both Windows and Linux (Mythbuntu 9.10), so I thought the easiest way would be to format it with NTFS. Installed the NTFS-3G package and I'm able to read and write to the drive fine from Linux, however I have a few questions;
1) How do I configure Linux so that when it mounts the NTFS partition it is writeable for user, group and other (bascially I want everyone to have read and write access)? Currently when the NTFS disk is mounted the permissions are restricted to the user only and I suspect I'll need to edit fstab for this, but don't have much experience here so need help with the specifics.
2) If my Linux PC is turned on with the external drive attached, the disk is not mounted until I double click on the icon on the desktop. Is there anyway I can configure Linux so that it will automatically mount the external disk when booting?
Below is what is in my fstab file at present;
Quote:
# /etc/fstab: static file system information. # # -- This file has been automaticly generated by ntfs-config -- # # <file system> <mount point> <type> <options> <dump> <pass>
I used the usual 'mkfs.xfs -l size=128m,lazy-count=1 /dev/sdX' at creation. After that, I would like to use custom mount options like: This goes instead of the "defaults" part in /etc/fstab
I receive the following error at boot: INVALID log iosize 4 [not 12-30] << No one used iosize 4... what does it mean? it is connected to the options..but which one? (At the minute I'm usig it with: noatime,nobarrier).
When using usb disks, there seems to be a difference in how they are mounted, based on the filesystem type on the disk.Vfat disks are mounted read/write for users, while extN filesystems are not.While I can fix that for individual devices, I would like to find a general solution, so that any usb storage device with any filesystem is mounted read/write for users.
I have a headless Ubuntu computer which is primarily use as a server but which has Ubuntu desktop edition installed. I use X forwarding through SSH occasionally to use GUI programs so I am not looking to remove them. However, I would like to disable any GUI elements that would be started automatically when the computer is booted. Is the graphical login screen the only thing that would be run? or is nautilus, gnome-session etc. started as well? How would I go about removing the necessary entries from the init scripts?
I just installed ubuntu 10.04 on a pc... the server edition. i went to restart samba and it is not in the init.d directory. is there a different folder for server plateforms? cause im use to my 9.10 desktop edition.
also one more question, when i "ls" the /ect/ directory of course i cant see the whole thing because my monitor only shows the bottom of the file, is there a way to list it like the "more" commands, or possibly scroll up?
I want to change the default mount options for removable devices, especially vfat devices, to have shortname=lower instead of the default shortname=mixed mount option. I googled around, and found references to /system/storage/default_options/vfat/mount_options gnome configuration option, but I don't seem to have this option set, actually, I don't seem to have a /system/storage tree in gconf-editor at all. searching in gconf-editor doesn't really yield any results.
I made a mistake in Gnome Disk utility tool. Instead of changing the options for a USB key, I changed the options of the root partition.
In Initial state, automatic options were desactivated. I just activated them. And after desactivacting them, I realized my mistake and switched back to "non automatic options".
By doing this, I suppose that defaults values were used since now, the system starts in command line mode and no more in graphic interface mode.
When I try the "startx" command, I get a "read-only" error.
With the command "sudo mount -o rw,remount /" the graphic interface is started.
Below is the configuration of the partition under the gnome disk utility tool :
"Mount at startup" is checked "Show in user inteface" and "Require additional authorization to mount" are unchecked
Mount options : nosuid,nodev,nofail,x-gvfs-show Mount point : /mnt/eb11d4d6-75db-XXXx-XXXx Identify as : /dev/disk/by-uuid/eb11d4d6-75db-XXXx-XXXx Filesystem type : auto
I do not want to change anything that could leat to a critical error. So what do you think I should do ?
I was wondering how to check the current mount options of my ext4 filesystems in Fedora 12? The only thing I can see in fstab is "defaults" but how can I know what is behind those defaults.mount command only shows "/dev/sda6 on / type ext4 (rw) but nothing else.I'd like to know a few more things like the data mode used like "ordered" or "journal".
I have a 2 TB hard drive in an external USB caddy that I use for backups. The drive gets automatically mounted when connected to F13, which is great, and the default mount options it uses are:Code:rw,nosuid,nodev,uhelper=udisksHowever, I want to add an extra option or two but I don't know where I can do this. Does anyone know where I can add mount options onto the default set?
I have two ubuntu servers. I have upgraded them to 10.04. Now I have serious problems at startup: iscsitarget and heartbeat scripts do not start all the times I reboot.I have installed monit to try to start iscsitarget and heartbeat and guess what? It does not start too at boot.In one of the two server sometimes it also starts with ethernet cards swapped.
Has anyone got a working apache upstart script? I'm running 10.04, and want the nice supervision stuff from upstart to run my apache instance. I've googled (especially for things like replacement-initscripts) but not hit anything concrete.
I wrote a custom init script for a java application. The script resides in /etc/init.d and follows the convention for the init.d scripts. However, the script runs the java process as root. I need these processes to run as a non-root user. Is there a way besides using sudo or su in the script (seems inelegant), to run the script as a specific non-root user? Perhaps using a .conf file in /etc/init or a daemon wrapper? What would the Ubuntu way of doing this be?
Before i set up the raid, but with this exact partitioning, the system booted perfectly. When i installed mdadm and created the raid1 mirroring on sda6 and sdb1, the init got screwed up, and all i get is a shell on initramfs, from where i can inspect that sda is binded on md, and cat /proc/mdstat tells me that i have an inactive sda[4].I can't mount the root partition (sda2), because it's busy (i suspect dmraid to lock it), which is, i guess, why init cannot be found.
I wonder if my error is to setup a raid array using a logical partition contained in an extended partition (but i hardly see why it would not work - but the sda bind and the sda[4] in mdstat seems to tell me that it does not), or it's just the initrd that is improperly configured. The other things that bothers me, is that changing the partition type of the raid partitions (fd to 0 - Empty), to disable raid autodetection, resulted in the same behavior on boot. Which might lead me again to think about configuration file problem instead of improper setup.The live cd doesn't not seem to recognize raid, so i can't inspect problems any further, but i could inspect system configuration, but i don't really know where to start.
