Difference between revisions of "OpenIndiana"
(→zpool scrub weekly) |
(→give user zfs and zpool rights) |
||
| Line 143: | Line 143: | ||
*http://www.unix.com/solaris/212235-solaris11-permission-issues-auto-scrub-zfs-pool.html | *http://www.unix.com/solaris/212235-solaris11-permission-issues-auto-scrub-zfs-pool.html | ||
| − | usermod -P "ZFS Storage Management,File System Management" username | + | usermod -P "ZFS Storage Management,ZFS File System Management" username |
=zpool scrub weekly= | =zpool scrub weekly= | ||
Latest revision as of 12:37, 29 August 2015
Contents
- 1 Share Data Via NFS
- 2 Share Data Via CIFS/Samba
- 3 SATA Port Commands
- 4 drive commands
- 5 Set Default Editor Nano
- 6 give user zfs and zpool rights
- 7 zpool scrub weekly
- 8 managing services
- 9 smart monitoring
- 10 Date and Time
- 11 Creating Static IP
- 12 Repos
- 13 Upgrading
- 14 Interesting
- 15 Notes
- 16 Mirroring rpool GPT (Hipster)
- 17 Managing Disks
- 18 Erasing GPT labels
- 19 pfexex, rbac, sudo
zfs set sharenfs=rw eh/data
To share a zfs file system named "bob" via smb/cifs execute a command like the following below. Note that the share name is arbitrary and can be completely different than the name of the file system itself, as seen in the example below.
$ zfs set sharesmb=name=myshare yourpool/shares/bob
You can set some text description (Windows Explorer shows this text) for the share using this:
$ zfs set "sharesmb=name=myshare,description=My ZFS share" yourpool/shares/bob
To just turn on cifs sharing for a zfs file system without specifying a share name execute a command like the following. Note that the auto-generated names created using this method are usually not user-friendly.
$ zfs set sharesmb=on yourpool/shares/bob
Check your work or check current status with:
$ zfs get sharesmb yourpool/shares/bob
You may need to enable "pam_smb_passwd" to make regular OpenIndiana users have smb passwords. To do so, add the following line to the end of the file "/etc/pam.conf":
other password required pam_smb_passwd.so.1 nowarn
And then change the user's password with the "passwd" command. After this, their smb password will also be set so they can connect via smb with the same username and password.
If you want to use AD to manage your users then take a look at the following http://docs.oracle.com/cd/E19963-01/html/821-1449/manageidmutm.html#enableidmusupporttask It allows you to enable IDMU so you can easily keep UIDs and SIDs in sync.
SATA Port Commands
https://osdude.wordpress.com/2010/09/20/sata-in-openindiana-oi_147/
# zpool import pool: backup id: 53109282349614151 state: ONLINE status: The pool is formatted using an older on-disk version. action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'. config: backup ONLINE c8t2d0s0 ONLINE # cfgadm -a sata Ap_Id Type Receptacle Occupant Condition sata4/0::dsk/c8t0d0 disk connected configured ok sata4/1::dsk/c8t1d0 disk connected configured ok sata4/2::dsk/c8t2d0 disk connected configured ok sata4/3::dsk/c8t3d0 disk connected configured ok sata4/4 sata-port empty unconfigured ok sata4/5 sata-port empty unconfigured ok # cfgadm -c unconfigure sata4/2 Unconfigure the device at: /devices/pci@0,0/pci1458,b002@11:2 This operation will suspend activity on the SATA device Continue (yes/no)? yes # cfgadm -c disconnect sata4/2 Disconnect the port: /devices/pci@0,0/pci1458,b002@11:2 This operation will disable activity on the SATA port Continue (yes/no)? yes # cfgadm -a sata Ap_Id Type Receptacle Occupant Condition sata4/0::dsk/c8t0d0 disk connected configured ok sata4/1::dsk/c8t1d0 disk connected configured ok sata4/2 sata-port disconnected unconfigured unknown sata4/3::dsk/c8t3d0 disk connected configured ok sata4/4 sata-port empty unconfigured ok sata4/5 sata-port empty unconfigured ok Now, hot swap the disks here, and continue: # cfgadm -c connect sata4/2 Activate the port: /devices/pci@0,0/pci1458,b002@11:2 This operation will enable activity on the SATA port Continue (yes/no)? yes # cfgadm -a sata Ap_Id Type Receptacle Occupant Condition sata4/0::dsk/c8t0d0 disk connected configured ok sata4/1::dsk/c8t1d0 disk connected configured ok sata4/2 disk connected unconfigured unknown sata4/3::dsk/c8t3d0 disk connected configured ok sata4/4 sata-port empty unconfigured ok sata4/5 sata-port empty unconfigured ok # cfgadm -c configure sata4/2 # cfgadm -a sata Ap_Id Type Receptacle Occupant Condition sata4/0::dsk/c8t0d0 disk connected configured ok sata4/1::dsk/c8t1d0 disk connected configured ok sata4/2::dsk/c8t2d0 disk connected configured ok sata4/3::dsk/c8t3d0 disk connected configured ok sata4/4 sata-port empty unconfigured ok sata4/5 sata-port empty unconfigured ok Verify that the disk is readable: # format -Me /dev/rdsk/c8t2d0s2 Inquiry: 00 00 03 02 1f 00 00 02 41 54 41 20 20 20 20 20 ........ATA 57 44 43 20 57 44 31 30 45 41 44 53 2d 30 30 4c WDC WD10EADS-00L 31 41 30 1A0 selecting /dev/rdsk/c8t2d0s2 [disk formatted] ... format> q # zpool import pool: backup id: 53109282349614151 state: ONLINE status: The pool is formatted using an older on-disk version. action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'. config: backup ONLINE c8t2d0s0 ONLINE
drive commands
cfgadm -x remove_device c3::c1t2d0 devfsadm
Set Default Editor Nano
For sh based shells:
export EDITOR=/path/to/nano
For C shell based:
setenv EDITOR /path/to/nano
Remember that unless you save these setting to your login profile script (/home/user/.bash_profile for example), you lose the setting at log out.
give user zfs and zpool rights
usermod -P "ZFS Storage Management,ZFS File System Management" username
zpool scrub weekly
export EDITOR=/usr/bin/nano crontab -e
0 0 * * 0 pfexec /sbin/zpool scrub rpool 0 2 * * 0 pfexec /sbin/zpool scrub mainpool
managing services
- list broken services
svcs -a
- delete service
svccfg delete smartd
- status service
svcs smartd
- status more info service
svcs -l svc:/site/smartd:default
- log file service
cat /var/svc/log/site-smartd:default.log
smart monitoring
pfexec pkg set-publisher -p http://pkg.openindiana.org/sfe pfexec pkg install storage/smartmontools
Example Smart Query Command (sata)
smartctl -a /dev/rdsk/c3t2d0 -d sat,12
- if you pass "-s on" it will turn smart on
Monitor Drives
nano -w /etc/smartd.conf
#DEVICESCAN
change these drive paths to your drive names
#nas drives /dev/rdsk/c3t0d0 -d sat,12 -a -o on -S on -s (S/../.././04|L/../../7/03) /dev/rdsk/c3t1d0 -d sat,12 -a -o on -S on -s (S/../.././04|L/../../7/03) /dev/rdsk/c3t2d0 -d sat,12 -a -o on -S on -s (S/../.././04|L/../../7/03) /dev/rdsk/c3t3d0 -d sat,12 -a -o on -S on -s (S/../.././04|L/../../7/03) #os drives /dev/rdsk/c3t4d0s0 -d sat,12 -a -o on -S on -s (S/../.././04|L/../../7/03) /dev/rdsk/c3t5d0s0 -d sat,12 -a -o on -S on -s (S/../.././04|L/../../7/03)
pfexec smartd -q onecheck
sudo nano -w /var/svc/manifest/site/smartd.xml
<?xml version="1.0"?>
<!DOCTYPE service_bundle SYSTEM "/usr/share/lib/xml/dtd/service_bundle.dtd.1">
<service_bundle type="manifest" name="smartd">
<service
name="site/smartd"
type="service"
version="1">
<single_instance/>
<dependency
name="filesystem-local"
grouping="require_all"
restart_on="none"
type="service">
<service_fmri value="svc:/system/filesystem/local:default"/>
</dependency>
<exec_method
type="method"
name="start"
exec="/etc/init.d/smartd start"
timeout_seconds="60">
<method_context>
<method_credential user="root" group="root"/>
</method_context>
</exec_method>
<exec_method
type="method"
name="stop"
exec="/etc/init.d/smartd stop"
timeout_seconds="60">
</exec_method>
<instance name="default" enabled="true"/>
<stability value="Unstable"/>
<template>
<common_name>
<loctext xml:lang="C">
SMART monitoring service (smartd)
</loctext>
</common_name>
<documentation>
<manpage title="smartd" section="1M" manpath="/usr/local/share/man"/>
</documentation>
</template>
</service>
</service_bundle>
pfexec svccfg -v import /var/svc/manifest/site/smartd.xml svcs smartd
If not on:
pfexec svcadm enable smartd
Date and Time
Timezone
tzselect
- Take value learned
nano -w /etc/default/init
- Set TZ=US/Eastern
You should reboot the system as all child processes take value from that file/data.
ntp client
cp /etc/inet/ntp.client /etc/inet/ntp.conf nano -w /etc/inet/ntp.conf
- Add some servers or whatever:
server 0.pool.ntp.org server 1.pool.ntp.org server 2.pool.ntp.org
- enable and restart ntpd
svcadm enable svc:/network/ntp:default svcadm restart svc:/network/ntp:default
- check to see if it is running
svcs -a | grep ntp
- query ntp server
ntpq -p
Creating Static IP
# nwamcfg nwamcfg> create ncp Nickebo nwamcfg:ncp:Nickebo> create ncu phys e1000g0 Created ncu 'e1000g0'. Walking properties ... activation-mode (manual) [manual|prioritized]> prioritized enabled (true) [true|false]> priority-group> 0 priority-mode [exclusive|shared|all]> shared link-mac-addr> link-autopush> link-mtu> nwamcfg:ncp:Nickebo:ncu:e1000g0> end Committed changes nwamcfg:ncp:Nickebo> create ncu ip e1000g0 Created ncu 'e1000g0'. Walking properties ... enabled (true) [true|false]> ip-version (ipv4,ipv6) [ipv4|ipv6]> ipv4-addrsrc (dhcp) [dhcp|static]> static ipv4-addr> 172.16.0.21 ipv4-default-route> 172.16.0.1 ipv6-addrsrc (dhcp,autoconf) [dhcp|autoconf|static]> dhcp,autoconf,static ipv6-addr> 2001:470:df83::21 ipv6-default-route> 2001:470:df83::1 nwamcfg:ncp:Nickebo:ncu:e1000g0> end Committed changes nwamcfg:ncp:Nickebo> end
# nwamadm enable -p ncp Nickebo Enabling ncp 'Nickebo'
- Also:
Revert to DHCP
nwamadm enable -p ncp Automatic
Repos
http://wiki.openindiana.org/oi/Spec+Files+Extra+Repository
The SFE package repository is an "extras" package repository, supplementing the core OI distribution with additional software from the Spec Files Extra project.
It contains a wide variety of software, including:
- Various languages and compilers (gcc 4.6.2, Steel Bank Common Lisp, the Glasgow Haskell Compiler, Python 3
- Various daemons and database programs (PostgreSQL, dovecot, postfix)
- Qt 4.7 and some applications which use it (the Scribus desktop publishing system, the LyX LaTeX front end, the Arora Web browser)
- Multimedia applications (Music Player Daemon and several clients, FFmpeg, mplayer2, vlc, MKVToolnix)
Notes
Commands
http://wiki.openindiana.org/oi/pkg+Cheat+Sheet
Upgrading
If you are currently running OpenIndiana oi_147 or oi_148, first, verify you have the correct publishers set:
pfexec pkg publisher
PUBLISHER TYPE STATUS URI openindiana.org (preferred) origin online http://pkg.openindiana.org/dev/
If your openindiana.org publisher is set to a different URI, for example http://pkg.openindiana.org/dev-il/ (our release candidate branch), you can correct this with:
pfexec pkg set-publisher -O http://pkg.openindiana.org/dev/ openindiana.org
opensolaris.org publisher If you currently have the opensolaris.org publisher set, we would highly recommend unsetting it by running:
pfexec pkg unset-publisher opensolaris.org
Make sure pkg is up to date:
pfexec pkg install package/pkg
You can now identify what will be upgraded by issuing the command:
pfexec pkg image-update -nv
The -n flag specifies to perform no action (i.e. perform a trial run), and the -v flag is the verbose option, to provide additional output.
I always do big upgrades in screen so:
screen
If you are satisfied with the actions to be taken, you can perform the upgrade by typing:
pfexec pkg image-update -v
You may want to consider also passing the "--be-name" flag to give the new boot environment created a more memorable name, such as oi_151a.
References
- http://wiki.openindiana.org/oi/oi_151a+Release+Notes
- http://wiki.openindiana.org/oi/Upgrading+OpenIndiana
Interesting
- http://wiki.openindiana.org/oi/Spec+Files+Extra+Repository
- http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide#Using_ZFS_Snapshots
- http://www.dansketcher.com/solaris-sata-support-for-onboard-chipsets/
Notes
Mirroring rpool GPT (Hipster)
Note: This should work but there is a bug: https://www.illumos.org/issues/6160#change-15117
So while this will work in the future because of the script that is ran on the add event, right now you still have to manually install the bootloader to the disk you add.
So:
- New disk is c2t50014EE65AC68D37d0
zdb
- Find the full path that has the s0 on it then:
installgrub /boot/grub/stage1 /boot/grub/stage2 /dev/rdsk/c2t50014EE65AC68D37d0s0
I had to change the dsk to rdsk. (Difference: http://www.unix.com/filesystems-disks-and-memory/6224-difference-dsk-rdsk.html)
15:08 < tsoome> ah, just zpool attach rpool and use disk name ending with d0
ZFS handles mirroring rpools automagically (see: /etc/sysevent/config/SUNW,EC_zfs,ESC_ZFS_bootfs_vdev_attach,sysevent.conf) It will install grub onto the new attached disk.
Mirroring rpool
root@host:~# zpool status
pool: rpool
state: ONLINE
scan: none requested
config:
NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
c2t50014EE65AC710BCd0 ONLINE 0 0 0
errors: No known data errors
root@host:~# format
Searching for disks...done
AVAILABLE DISK SELECTIONS:
0. c2t50014EE65AC68D37d0 <ATA-WDCWD10JFCX-68N-0A82 cyl 60798 alt 2 hd 255 sec 126>
/scsi_vhci/disk@g50014ee65ac68d37
1. c2t50014EE65AC710BCd0 <ATA-WDC WD10JFCX-68N-0A82-931.51GB>
/scsi_vhci/disk@g50014ee65ac710bc
Specify disk (enter its number): ^C
root@host:~# zpool attach rpool c2t50014EE65AC710BCd0 c2t50014EE65AC68D37d0
Make sure to wait until resilver is done before rebooting.
Be sure to configure your hardware to boot from both devices in case of failure.
Managing Disks
- Find the disks
format
List the partitions
partition
then
cfgadm way
cfgadm -al cfgadm -s "select=type(scsi-sas)"
Erasing GPT labels
Erase first 32 sectors Erase last 32 sectors
GPT partitioning has reserved partition, no s2 (as there can not be overlapping partitions) and s0 starts from 256 reserved is there because backup GPT label is at the end of the disk, that prevents it to be overwritten.
The illumos (solaris) implementation limits the partition count there to maintain interface compatibility with vtoc
dd bs=512 if=/dev/zero of=/dev/sda count=2048 seek=$((`blockdev --getsz /dev/sda` - 2048))
This would do the last part of the disk......2mb. I modded mine for 8mb for the heck of it.
pfexex, rbac, sudo
sudo works on OI but pfexec is the way I guess
http://www.c0t0d0s0.org/archives/4844-Less-known-Solaris-features-pfexec.html
Basically turns a user to root:
usermod -P'Primary Administrator' username
List profiles of current user
profiles
Or another user
profiles username
Remove all profiles from user
pfexec usermod -P "" webdawg
More info:
[-P profile [, profile]]
-P profile
One or more comma-separated rights profiles defined in
prof_attr(4). This replaces any existing profile setting in
user_attr(4). If an empty profile list is specified, the existing
setting is removed.
