Automounter and volume manager based on DevKit

Hi there,
I've written a simple automounter and volume manager based on DeviceKit:
http://www.ashrentum.net/jmcejuela/prog … 0.2.tar.gz
Some may find it interesting.
Currently it only manages media drives. I hasn't support yet for CDs/DVDs.

jmcejuela wrote:
Hi there,
I've written a simple automounter and volume manager based on DeviceKit:
http://www.ashrentum.net/jmcejuela/prog … 0.2.tar.gz
Some may find it interesting.
Currently it only manages media drives. I hasn't support yet for CDs/DVDs.
Hello, good job!
I was thinking of writing a patch for skvm to use DeviceKit. How does DeviceKit compare to HAL? any obvious deficiencies or whatnot? I've not looked into this at all yet. I'll look into the code to see if I can get any ideas

Similar Messages

Solaris 10 x86 and Volume Manager fdisk identifiers

Hi,
I'm playing with Sun Volume Manager and have a question. I have installed two drives in addition to my system drive where Solaris 10 resides. I used format to add a Solaris fdisk partition of Solaris2 (0xbf). I'm aware that the old identifier used to be 0x82, same as Linux swap identifier. However, when I open up smc and go to Storage->Disks and select the two unused disks, smc complains and tells me that no Solaris Fdisk partitions exist. If I re-partition these drives with the old Solaris identifier (0x82), smc doesn't complain.
Is this a known issue with the Volume Manager gui? Because I am able to work with the disks just fine from the command line using the meta* commands no matter what identifier I use. Maybe I need a patch? Thanks for any pointers.
Jason

Had the same problem. Look here:
http://forums.sun.com/thread.jspa?threadID=5348727
Hope that helps!
Jeremy

Solaris 10 - native cluster, multipath, volume management.

Hi,
I have downloaded Solaris 10 x86 DVD. Planning to install it inside VMWare. Kindly let me know if there is any native clustering , multipathing and volume management software thats pre bundled with Solaris 10 or is it paid?
I am a storage admin and I want to get a clear understanding of how the above mentioned things work. Will install two instances of Solaris 10 and try out a couple of things.
Please let me know.

Hi Pascal,
Thank you for such a quick reply.
I am looking for a clustering software that will basically be free, yet give me a glimpse of how the clustering stuff works on Solaris. I cannot afford the paid stuff, so will have to make do with the freebies...If my understanding is correct, you are stating that the native multipathing and volume management is integrated AND free. Kindly let me know.
So if I install a few virtual disks inside VMware, can I use the native multipathing to detect them? I cant get the vmware solaris to access SAN, so basically looking for something that will make the OS "think" that its getting SAN LUNs, so that I can try multipathing.
Sorry for topping up another question.
Edited by: 992432 on Mar 7, 2013 3:40 AM

Gnome-volume-manager doesn't work, nor another automounter

Hi there,
I have installed gnome-volume-manager but it apparentely does not work. I've followed the instructions from the FAQ - Arch Linux, and I have tried also recompiling it with --enable-automounting. I have tried also having Nautilus opened at the same time but it didn't work. My WM is IceWM.
//ivman works for the cdrom but not for usb devices. And I don't like AutoFS.
Last edited by jmcejuela (2009-04-10 15:06:34)

I had this problem and it was related to ConsoleKit not being initialized properly. I was using SLIM as login manager and had to add ck-launch-session to my .xinitrc when starting the gnome session.
Last edited by iBertus (2009-04-10 17:21:33)

Training and Event Management Implementation based on competencies

Dear Friends,
My client is going ahead for Training and Event Managment Implementation. They have a basic requirement to start with and that is :
1) They have done competency mapping for all its employees and they want that the competencies of each employees(along with the skill levels) to be recorded in the system and that has to be the starting point of using Training and Event Management module.
2) They want, if the competencies can flow based on Job/ Position.
3) Some identifier to the competencies, whther it has flowed from Appraisal or any other sources in the Final Training Needs.
Kindly provide me help, as to how I will be able to achieve that and in what Infotypes the data pertaining to Training and Event Managment will be stored.
If u all can kindly share with me the User Manuals and Configuration Docs of Training and Event Management, it will be of great help.
Thank you all.

Hi,
Competencies can be stored as qualifications in PD and then by activation of PD PA intergration can be seen from pa30 infotype 24.
Qualifications can be stored against a Job/Position and are called as the Requirements. They are seen as a separate Tab and to which ever position the person is linked to the corresponding qualifications of the position will appear in the requirements tab.
You can maintain the proficiency and a note along with the qualification when assigned to a person.
Also Appraisals can have qualifications in the template rather than criteria and criteria groups.
Also after training is completed during the follow up we can create an appraisal and transfer the qualifications or simply transfer the qualifications to the employee.
Regards,
Divya

IO issues on Red Hat Linux SAN storage and veritas volume manager

Oracle 11.0.1.7:
We recently moved from Solaris Sparc box to linux box and since then we are seeing IO issues. And it's quite obvious in iostat that the io goes as high as 100ms. So I am wondering if someone can advice if there are any mount options we need to give or if we need to use vxodmfs file system. Below are the details of this environment:
1. It's using veritas volume manager. (Same as solaris)
2. It's using vxfs file system (Same as solaris)
3. It's using veritas ODM libraries that helps us in IO. (Same as solaris)
4. SAN EMC storage (Same as solaris)
5. Mount options /dev/vx/dsk/orapdata1dg/oradata1vol1 on /u01/oradata type vxfs (rw,delaylog,largefiles,ioerror=mwdisable)
Edited by: user628400 on Aug 7, 2009 5:48 PM

Not sure if this really applies here but, from Metalink note 359515.1, parameters should be like follows on NAS devices:
Linux x86 #*
Mount options for Binaries
rw,bg,hard,nointr,rsize=32768,
wsize=32768,tcp, vers=3,
timeo=600, actimeo=0
Mount options for Oracle Datafiles
rw,bg,hard,nointr,rsize=32768,
wsize=32768,tcp,actimeo=0,
vers=3,timeo=600
Linux x86-64 #*
Mount options for Binaries
rw,bg,hard,nointr,rsize=32768,
wsize=32768,tcp,vers=3,
timeo=600, actimeo=0
Mount options for Oracle Datafiles
rw,bg,hard,nointr,rsize=32768,
wsize=32768,tcp,actimeo=0,
vers=3,timeo=600
Give it a try and let us know....
Edited by: JRodriguez on Aug 10, 2009 6:35 PM

WebStart Flash and Veritas Volume Manager

I would like to use WebStart Flash for backup of the system
disk. The goal is to be able to recover the system disk rapidly.
It works perfectly for systems without Veritas Volume
Manager on the system disk.
However, if Veritas Volume Manager is installed and used
for mirroring the root disk, the system is not able to boot using
WebStart Flash. This is probably because the "private region"
of the disk is not included in the flash archive.
Does anybody have a solution for this, or does any of
you successfully combine WebStart Flash and Veritas
Volume Manager?
I use Jumpstart and the install_type is configured to
flash_install.
The question was also asked in the newsgroup
comp.unix.solaris.
Rgds,
Henrik

For many reasons, today you cannot save the VxVM
private region information as an implicit part of a
flash archive. The procedure would likely be to
unencapsulate the root drive, create the flash archive,
then re-encapsulate the root drive. This is an ugly
procedure and may cause more pain than it is worth.
When a root disk is encapsulated, an entry is put
into the /etc/system file which says to use the VxVM
or SVM logical volume for the rootdev rather than
the actual device from which the system was originally
booted. When you create a flash archive, this modification
to the /etc/system is carried along. But, when you install
it on a new system which doesn't have VxVM properly
installed already (a chicken-and-the-egg problem)
then the change of the rootdev to the logical volume
will fail. The result is an unbootable system (without
using 'boot -a' and pointing to a different /etc/system
file like /dev/null).
The current recommended process is to use a prototype
system which does not have an encapsulated root
to create the flash archive.
VxVM also uses the ELM license manager which will
tie the VxVM runtime license to the hostid. This makes
moving flash archives with VxVM to other machines
impractical or difficult.
The long term solution would be to add logical volume
management support to the JumpStart infrastructure.
I'm not counting on this anytime soon :-(
-- richard

How to break Veritas Volume Manager mirrors and wipe disks

I have an old Solaris 10 server that is ready to be decommissioned. I would like to go in and break the Veritas Volume Manager mirrors and still be able to go in and wipe each of the two internal disks by using dd or something similar. Below is my Veritas Vol. Mgr. layout. I would greatly appreciate it if someone can post the exact commands (both for the Veritas Vol. Mgr. mirror break and the dd) I would need to run to perform the above tasks. I want to wipe the disks clean as to remove any company proprietary information prior to releasing the system.
Note: I will be logged into a remote console session to perform these actions as the system is in a datacenter and is located in a rack which does not have a monitor.
vxdisk list*
DEVICE TYPE DISK GROUP STATUS
c0t0d0s2 sliced rootdisk rootdg online
c0t1d0s2 sliced rootmirror rootdg online
vxdg list_
NAME STATE
rootdg enabled
vxprint -htr_
Disk group: rootdg
dg rootdg default default 0 1113432795.1025.dh22020
dm rootdisk c0t0d0s2 sliced 2632 71121402 -
dm rootmirror c0t1d0s2 sliced 2888 71124291 -
v appsvol - ENABLED ACTIVE 27262976 SELECT - fsgen
pl appsvol-01 appsvol ENABLED ACTIVE 27263493 CONCAT - RW
sd rootdisk-04 appsvol-01 rootdisk 41948280 27263493 0 c0t0d0 ENA
pl appsvol-02 appsvol ENABLED ACTIVE 27263493 CONCAT - RW
sd rootmirror-04 appsvol-02 rootmirror 41948280 27263493 0 c0t1d0 ENA
v rootvol - ENABLED ACTIVE 16779312 ROUND - root
pl rootvol-01 rootvol ENABLED ACTIVE 16779312 CONCAT - RW
sd rootdisk-01 rootvol-01 rootdisk 0 16779312 0 c0t0d0 ENA
pl rootvol-02 rootvol ENABLED ACTIVE 16779312 CONCAT - RW
sd rootmirror-01 rootvol-02 rootmirror 0 16779312 0 c0t1d0 ENA
v swapvol - ENABLED ACTIVE 8389656 ROUND - swap
pl swapvol-01 swapvol ENABLED ACTIVE 8389656 CONCAT - RW
sd rootdisk-02 swapvol-01 rootdisk 16779312 8389656 0 c0t0d0 ENA
pl swapvol-02 swapvol ENABLED ACTIVE 8389656 CONCAT - RW
sd rootmirror-02 swapvol-02 rootmirror 16779312 8389656 0 c0t1d0 ENA
v var - ENABLED ACTIVE 16779312 ROUND - fsgen
pl var-01 var ENABLED ACTIVE 16779312 CONCAT - RW
sd rootdisk-03 var-01 rootdisk 25168968 16779312 0 c0t0d0 ENA
pl var-02 var ENABLED ACTIVE 16779312 CONCAT - RW
sd rootmirror-03 var-02 rootmirror 25168968 16779312 0 c0t1d0 ENA
Thanks much for any assistance on this.

I would boot off of a cdrom and then simply use the format command to wipe the disks individually.
alan
http://www.sun.com/blueprints/0600/scrub.pdf

User for ADVM and ACFS Volume management

Hi I am getting confused which user to use for ADVM and ACFS management.
I want to create a new volume and ACFS.
My normal grid and db admin user is "oracle" in the group 'oinstall". For the ACFS setup, I have granted the oracle user with root privileges.
No I want to create a new volume in ADVM using "ASMCMD". No matter whether I create the volume with the root user, or the oracle user., the volume is assigned to the following user:
[root@grid1 asm]# ll /dev/asm/
total 0
brwxrwx---. 1 root oinstall 251, 194561 Mar 24 10:15 volume1-380
Is this correct?
Because when I do (as the oracle user):
sudo /sbin/acfsutil registry -a /dev/asm/volume1-380 /srv/u01/app/acfsmounts/myacfs
The privileges are:
[oracle@grid1 app]$ ll
total 16
d---------. 3 root root 4096 Mar 24 10:51 acfsmounts
[root@grid1 acfsmounts]# ll
total 4
drwxr-xr-x. 4 root root 4096 Mar 24 10:32 myacfs
Now the condition is as follow:
As oracle user:
no access at all on both nodes:
[oracle@grid1 acfsmounts]$ ls
ls: cannot open directory .: Permission denied
[oracle@grid2 acfsmounts]$ ls
ls: cannot open directory .: Permission denied
As root user:
On node1, I can do everything:
[root@grid1 myacfs]# echo "dadwad" > dawdawd.txt
--> SUCCESS
But on node2, I cant create a file:
[root@grid2 myacfs]# cat dawdawd.txt
--> SUCCESS
[root@grid2 myacfs]# echo "dawd" > newfile
-bash: newfile: Permission denied

Please refer "Basic Steps to Manage Oracle ACFS Systems"
Creating an Oracle ACFS File System
http://docs.oracle.com/cd/E11882_01/server.112/e18951/asmfs_util001.htm#OSTMG94787
I believe you can succeed by following the steps described in the documentation, so if you get any issue just post here.

[SOLVED]gnome-volume-manager / hal 0.5.10 - can't deactivate automount

http://wiki.archlinux.org/index.php/HAL
I've created /etc/hal/policy/preferences.fdi
<?xml version="1.0" encoding="UTF-8"?>
<deviceinfo version="0.2">
<device>
<match key="storage.hotpluggable" bool="false">
<match key="storage.removable" bool="false">
<merge key="storage.automount_enabled_hint" type="bool">false</merge>
</match>
</match>
</device>
</deviceinfo>
The article say "Policies" may be deprecated in hal 0.5.10. I doubt that, but i dont know what is deprecated though.
I did check the documentation out, and everything i did still seems to be valid.
http://people.freedesktop.org/~david/ha … ies-volume
Does gnome-volume-manager honor this setting in gnome 2.20?
Last edited by pelle.k (2008-02-04 20:12:51)

It would seem i solved it myself.
First of all, gnome-volume-manager need to be recompiled with a patch i found (ubuntu gutsy); http://patches.ubuntu.com/by-release/ex … hint.patch
Why this isn't included by default in gnome-volume-manager is beyond me!? You bad, bad, gnome devs!
Replace the PKGBUILD when building from ABS.
# Contributor: Link Dupont <[email protected]>
pkgname=gnome-volume-manager
pkgver=2.17.0
pkgrel=3
pkgdesc="GNOME daemon to auto-mount and manage media devices"
arch=(i686 x86_64)
license=('GPL')
url=http://www.gnome.org/
depends=('libgnomeui>=2.20.0' 'libnotify>=0.4.4' 'gnome-mount>=0.6')
makedepends=('nautilus>=2.20.0' 'perlxml' 'pkgconfig')
install=gnome-volume-manager.install
groups=('gnome-extra')
source=(http://ftp.gnome.org/pub/GNOME/sources/${pkgname}/2.17/${pkgname}-${pkgver}.tar.bz2
gvm-mount-async.patch
gvm-check-local.patch
http://patches.ubuntu.com/by-release/extracted/ubuntu/g/gnome-volume-manager/2.17.0-2ubuntu2/02_honour_automount_enabled_hint.patch)
md5sums=('104cec26e721e0bba69debd392367195'
'534b8b9633148bd975b2b9b518c41fb4'
'cb45810b2adb95d39e1272470f10a21a'
'7288a9dc57df321d6a379c9b6a7caaff')
build() {
cd ${startdir}/src/${pkgname}-${pkgver}
patch -Np0 -i ${startdir}/src/gvm-mount-async.patch || return 1
patch -Np0 -i ${startdir}/src/gvm-check-local.patch || return 1
patch -Np1 -i ${startdir}/src/02_honour_automount_enabled_hint.patch || return 1
./configure --prefix=/usr --sysconfdir=/etc \
--localstatedir=/var
make || return 1
make GCONF_DISABLE_MAKEFILE_SCHEMA_INSTALL=1 DESTDIR=${startdir}/pkg install
mkdir -p ${startdir}/pkg/usr/share/gconf/schemas
gconf-merge-schema ${startdir}/pkg/usr/share/gconf/schemas/${pkgname}.schemas ${startdir}/pkg/etc/gconf/schemas/*.schemas
rm -f ${startdir}/pkg/etc/gconf/schemas/*.schemas
Also, this is my /etc/hal/fdi/policy/preferences.fdi
<match key="storage.hotpluggable" bool="false">
<match key="storage.removable" bool="false">
<merge key="storage.automount_enabled_hint" type="bool">false</merge>
</match>
</match>
If you compare that to my first version, you'll note that it is *not* enclosed in <deviceinfo> and <device> tags, simply because that wouldn't work. Don't ask me why. i'm by no means a HAL expert...
Last edited by pelle.k (2008-02-04 20:12:00)

Oracle and volume- and disk array-based replication?

Dear mighty all,
here is a situation:
Location A - active Oracle database + volume replication software + geocluster software
Location B - standby location, i.e. standby database + volume replication software + geocluster
Oracle is 10g
The size of the database is 500 GB+
The daily updates are roughly 2-3%.
The comm channel between locations can have any reasonable bandwidth, however the jitter and latency cannot be guaranteed (i.e. no synchronous replication).
This is a matter of protecting against a disaster (such as a complete destruction of Location A).
The task is to make sure that Location B has the up-to-date data (RPO is 8 hours) and Oracle database can be started automatcially without any manual intervention. (I.e. cluster administrator pushes the button and the rest is done without human intervention).
Here is what I am thinking of:
1) volume replication software usually observes the atomicity on the disk block level. I.e. it is impossible that the data block being updated on the volume in Location A is updated in location B incompletely.
2) When the database in Location A completes a transaction it writes data to the disk thus initiating the series of disk block updates being sent to Location B.
3) From the database point of view, the transaction is completed, database files are updated.
4) However there is a slim chance of Location A being destroyed while:
* the volume replication software has started pushing through to Location B the updated disk blocks representing the completed transaction
* but it did not finish transmitting the whole lot of disk updates (boom! the asteroid fell on Location A)
* and volume replicator in Location B cannot roll back the disk updates since volume replicator does not know of the database transaction
5) Volume replicator atomicity is disk data block while Oracle's atomictity is transaction (i.e. there cannot be a half-transaction applied)
6) Thus, when we start Oracle in Location B, we will get inconsistent database, since the last transaction was transferred to Location B incompletely
7) Sure that there is a way to fix this using various DBA tools, but that defies the whole purpose of automatic start
So, my question is:
1) Is volume- or disk array-based replication not a good idea in this case at all?
2) or is there a way to nonetheless ensure the consistency of the database (say start it every time in the recovery mode)?
3) Or should I dump the idea of replicating Oracle with volume replication and switch to DataGuard? (The cluster software can handle DataGuard).
thanks a million for your answers, ye mighty all!

Hi,
this is not a "typical" question what is easy to discuss in a forum.
Just some hints:
1. using volume replication won't help you against logical or physical corruption
2. RPO of 8 hours ? Sounds a lot regarding a change rate of 10-15 GB a day.
3. 100% Automation is normally not a good idea in a database disaster scenario
( a false disaster detection can end in a real disaster!) I've already seen too much databases which had been totally corrupted by a false switch.

Mpxio and Veritas Volume Manager

Is there a good procedure on how to implement mpxio with existing Veritas Volume Manager volumes? I tried last night and on the reconfigure reboot the server could not find the VM disks. I had to back out mpxio and reboot again.
Thanks,
Gary Blunk

That should work. I'd make sure you have the right version of VxVM as mentioned in the Traffic Manager install/config guid.
It could be a bunch of things but if you nstalled VxVM, then installed Traffic Manager, and the devices weren't found......that's a symptom of the wrong VxVM version.
The install/config guide has the required versions.

Solaris volume manager and RPC

I'm trying to run as few services as possible? Does anybody know if there are dependencies between Solaris Volume Manager (formerly DiskSuite) and rpc? I know that Solaris Volume Manager uses something called rpc.metamhd, but I'm not sure if this requires the full blown rpc.
thanks...

Duplicate slicing: prtvtoc /dev/rdsk/c1t0d0s7|fmthard -s - /dev/rdsk/c1t1d0s7
Lather, rinse, repeat for each file system:
metainit the bootable side of the mirror: metainit -f d71 1 1 c1t0d0s7.
metainit the other side: metainit d72 1 1 c1t1d0s7.
Attach one side of the mirror: metainit d70 -m d71
When done: lockfs -fa;init 6
Lather, rinse, repeat for each file system:
After reboot attach the other side of the mirror: metattach d70 d72
bash-3.00# metastat -p
d70 -m d71 d72 1
d71 1 1 c1t0d0s7
d72 1 1 c1t1d0s7
d50 -m d51 d52 1
d51 1 1 c1t0d0s5
d52 1 1 c1t1d0s5
d10 -m d11 d12 1
d11 1 1 c1t0d0s1
d12 1 1 c1t1d0s1
d0 -m d1 d2 1
d1 1 1 c1t0d0s0
d2 1 1 c1t1d0s0
d60 -m d61 d62 1
d61 1 1 c1t0d0s6
d62 1 1 c1t1d0s6
bash-3.00# cat /etc/vfstab
#device         device          mount           FS      fsck    mount   mount
#to mount       to fsck         point           type    pass    at boot options
fd      -       /dev/fd fd      -       no      -
/proc   -       /proc   proc    -       no      -
/dev/md/dsk/d10 -       -       swap    -       no      -
/dev/md/dsk/d0 /dev/md/rdsk/d0 /       ufs     1       no      -
/dev/md/dsk/d70 /dev/md/rdsk/d70        /usr    ufs     1       no      -
/dev/md/dsk/d50 /dev/md/rdsk/d50        /var    ufs     1       no      -
/dev/md/dsk/d60 /dev/md/rdsk/d60        /opt    ufs     2       yes     -
/devices        -       /devices        devfs   -       no      -
ctfs    -       /system/contract        ctfs    -       no      -
objfs   -       /system/object objfs   -       no      -
swap    -       /tmp    tmpfs   -       yes     -

SVM and Vertis volume management Simple Question

Pls guys can you interrelate the terminologies in Vertis volume management with that of SVM ? wht are this in SVM ?
Vertis
disk group
volume group
logical volumes
to SVM terminologies
soft partition , etc....

I'm sure there is a better answer, if you had sound level measuring equipment. I'm saying human ears are different (in sensitivity and size), different earbuds play at different levels for the same setting on the iPod, and even different iPod models play at different levels for the same setting. Another consideration is distance of the sound source from the ear drum. In-ear earbuds are the closest, so the actual sound they produce is the lowest for a given level of loudness, and the loudness probably varies upon how far IN a particular user inserts them. Regular earbuds are a bit more distant, so they produce a bit more sound. Over-the-ear headphones are farther still, so they produce even more sound. And living room speakers are across the room, so they need to produce many times the sound, for the same level of loudness. Therefore, just measuring the actual sound produced will not give the answer either.
So you're not going to get an answer like +9 out of 14 on your iPod is safe but 10 is not+.
An equally important factor is probably exposure time. If I'm listening to my iPod for two hour a day, that will have less impact on my hearing than listening for ten hours a day. If I went to a dance club once a month, I wouldn't be worried about my hearing. If I worked at a dance club, I would be wearing military grade ear plugs.

Firewire and gnome-volume-manager

I have a firewire scanner, when hotplugged, nothing happens in gnome. I have 'run when scanner is plugged' checked, but I get nothing.
Does gnome-volume-manager work with firewire devices? or is it just my scanner? (Canon 2540 photo)

I don't know about scanners, but my firefire external HD is recognised perfectly by gnome-volume-manager. So it does listen for firewire devices. Maybe you'll need to add something to udev to make gnome-volume-manager know it's a scanner?
I have a scsi scanner, so I won't try hotplugging it

Automounter and volume manager based on DevKit

Similar Messages

Maybe you are looking for