OpenSolaris derived ZFS NAS/ SAN (OmniOS, OpenIndiana, Solaris and napp-it)

I've updated. Unfortunately, I have to wait a month to see if the update works :)

Thanks for your help, _Gea.
 
Smartmontools work!
But how can i see the temperatures from here?
x3QfvA.png


Also i found this about PWN fans, supermicro boards and xcase:
http://www.youtube.com/watch?feature=player_detailpage&v=TYpQ8Xq2ryg#t=259
Seems the PWN fans were controlled from IPMI
 
Hum, weird. I get temperatures in that screen.

What controller are you using?

Matej
 
Smartmontools work!
But how can i see the temperatures from here?

I am aware of problems with smartmontools with some configs.

what you can do beside wait for a new release or try smartmontools at cli
- open a ssh connection (putty, login as root)
- enter (copy and paste with a mouse right click in putty):
perl /var/web-gui/data/napp-it/zfsos/_lib/get-disk-smart-bg.pl

This starts the script that reads smartvalues directly (normally a background job) and prints information to console. Maybee this gives some information or hints.

you can also redirect the output like
perl /var/web-gui/data/napp-it/zfsos/_lib/get-disk-smart-bg.pl > /tmp/smart.txt
and view the smart.txt ex with WinSCP from a Windows machine.
 
Hmm, my 1015 flashed to LSI reports temps. Is yours still in raid firmware by chance?

Try running directly from the command line to see if it reports:
/usr/sbin/smartctl -a -d scsi /dev/rdsk/c4t5000CCA22EC0BD39d0

For example my 9211 reports in the gui just fine, my br10i doesn't report anything in the gui, but still works fine from the CLI using the command above.
 
Using OpenSolaris and Napp-it and have successfully registered system with Active Directory. Whatever was supposed to periodically register the DNS entry did not work and the DNS entry went stale and was removed. Now I need to have that re-added.

Anyone know the process to manually force a re-registration in DNS? Something like nsupdate, except I don't think that's it.

Thanks,
D

Is it tied to the SMB service?
 
Hmm, my 1015 flashed to LSI reports temps. Is yours still in raid firmware by chance?

Try running directly from the command line to see if it reports:
/usr/sbin/smartctl -a -d scsi /dev/rdsk/c4t5000CCA22EC0BD39d0

For example my 9211 reports in the gui just fine, my br10i doesn't report anything in the gui, but still works fine from the CLI using the command above.

My M1015 were all flashed in IT, i tried the command but it returned:
Rk3hCj.png
 
what distro are you using? You don't have smart tools installed (or at least in the usual place), that's why it won't work.
 
Hello All

is someone here who have successfully played with the powermanagement at /etc/power.conf on Omnios?

It is no problem to spin down the harddrives but its not the solution for me because the system needs more than one minute to get ready. (23 HDDs)

My next attempt is to use:
Code:
# Auto-Shutdown    Idle(min)    Start/Finish(hh:mm)    Behavior
autoshutdown       30           0:00 0:00              shutdown
Looking in the man - the idle time I can define with:
diskreads 50
loadaverage 1.1
Telling the framework the changes with # /usr/sbin/pmconfig but nothing happened no shutdown when idle.

Ok, looking at powertop:
Code:
OpenSolaris PowerTOP version 1.2

C-states (idle power)   Avg     Residency                                    P-states (frequencies)
C0 (cpu running)                (2.5%)                                       1200 Mhz        98.1%
C1                      0.0ms   (0.0%)                                       1333 Mhz        0.0%
C2                      1.8ms   (3.8%)                                       1467 Mhz        0.0%
C3                      2.0ms   (93.7%)                                      1600 Mhz        0.0%
                                                                             1733 Mhz        0.0%
                                                                             1867 Mhz        0.0%
                                                                             1917 Mhz(turbo) 1.9%

Wakeups-from-idle per second: 3854.6    interval: 5.0s
no ACPI power usage estimate available

Top causes for wakeups:
46.7% (1799.9)                 sched :  <xcalls> unix`dtrace_xcall_func
12.4% (477.2)                  sched :  <xcalls> unix`speedstep_pstate_transition
 3.1% (119.8)               <kernel> :  genunix`cv_wakeup
 3.0% (115.0)                   perl :  <xcalls> unix`speedstep_pstate_transition
 2.6% (100.2)               <kernel> :  genunix`clock
 1.3% ( 50.0)               <kernel> :  SDC`sysdc_update
 0.2% (  8.2)               <kernel> :  cpudrv`cpudrv_monitor_disp
 0.2% (  7.6)               <kernel> :  ehci`ehci_handle_root_hub_status_change
 0.1% (  4.0)               <kernel> :  genunix`schedpaging
 0.1% (  2.8)                syslogd :  <xcalls> unix`speedstep_pstate_transition
 0.1% (  2.8)                  sleep :  <xcalls> unix`speedstep_pstate_transition
 0.1% (  2.8)                   nscd :  <xcalls> unix`speedstep_pstate_transition
 0.1% (  2.8)                   bash :  <xcalls> unix`speedstep_pstate_transition
 0.1% (  2.8)                   ntpd :  <xcalls> unix`speedstep_pstate_transition
 0.1% (  2.0)               <kernel> :  e1000g`e1000g_local_timer
 0.0% (  1.6)                   smbd :  <xcalls> unix`speedstep_pstate_transition
 0.0% (  1.4)        hald-addon-acpi :  <xcalls> unix`speedstep_pstate_transition
 0.0% (  1.4)               devfsadm :  <xcalls> unix`speedstep_pstate_transition
 0.0% (  1.4)            <interrupt> :  e1000g#0
 0.0% (  1.4)                fsflush :  <xcalls> unix`speedstep_pstate_transition
Looks like dtrace makes some wakeup calls.

I want that the media-server shutdown when no one in my network have a connection to the server.
Any hints??
Maybe "loadaverage" set to 2.8?? because ---> C0 (cpu running) (2.5%)
 
OmniOS_Text_r151010 + nappit 0.9e1 (the one i got from wget command)

I managed to set the fans to full speed using IPMI, but infortunately there's no "medium" speed available...
(Only 900 or 2400RPM, but i'd like to have decent cooling without hurting my ears)

Other thing, i was forced to power it off from IPMI yesterday as it didn't respond and now it's not very happy and keep returning errors, i can't even access the smartinfo page.
How do you fix this?
IxoyOa.png

U07e1q.png


I tried reinstalling omnios but it also return error!
XGFnum.png

oh, this time the reinstallation seems to work! Don't even know why but it worked.
 
Last edited:
I've decided to toss in a 1366 quad-core Xeon in the x58 and use 4x4GB ECC DDR3, probably going with those Ultrastars (going for 2 to begin with in a RaidZ1, then later will dump and recreate into a RaidZ2 or Z3 when I obtain more disks) It's a very modest setup here at the home office, I think this should be sufficient.

When reading the documentation (http://www.napp-it.org/doc/downloads/napp-it.pdf) it says NFS V4 is also supported and allows user based permissions and ACL (not used with ESXi). Does that mean that an ESXi host ignores/is incompatible with user based permissions/ACLs or that Napp-it NFS4 and ESXi are incompatible entirely and I should use NFS3 instead? I am using ESXi 5.5.0

Thanks for the advice

Are you sure your motherboard supports ECC RAM ?
 
nappit free don't allow ACL permissions??

napp-it free or pro does not restrict any OS features.
You can use ACL when you

- set them via CLI like you would do without napp-it
- use the ACL extension for free but only for base settings like everyone@
- use Windows to set ACL (but do not set deny rules from Windows)
 
Not at all, the motherboard has to have the traces and BIOS support for ECC. Typical Intel consumer boards don't have them, only workstation/server ones.
 
Not at all, the motherboard has to have the traces and BIOS support for ECC. Typical Intel consumer boards don't have them, only workstation/server ones.

I've double checked, Xeon/ECC support is available for this board, the BIOS does need to support it (which I've updated to the latest). It doesn't support *all* socket 1366 Xeons, which is the catch.

http://www.evga.com/support/motherboard/legacy/
 
Hi,

Been using napp-it + OI for some time now on ESXi with Unraid and Win7 vm.
I have 1 vdev raidz1 which consists of 8x 2TB drives. The vdev is getting full and I've bought 4 new WD red 3TB drives and will swap 4 3TB drives from Unraid to the vdev, so 8x 3TB in total.
I would like to replace 1 disk at a time and resilver as I need the swapped drives back in Unraid for data. After all drives have been resilvered I would like to expand the pool from 16TB to 24 TB.
How do I proceed and what do I need to be careful of?

thanks!
 
Hi,

Been using napp-it + OI for some time now on ESXi with Unraid and Win7 vm.
I have 1 vdev raidz1 which consists of 8x 2TB drives. The vdev is getting full and I've bought 4 new WD red 3TB drives and will swap 4 3TB drives from Unraid to the vdev, so 8x 3TB in total.
I would like to replace 1 disk at a time and resilver as I need the swapped drives back in Unraid for data. After all drives have been resilvered I would like to expand the pool from 16TB to 24 TB.
How do I proceed and what do I need to be careful of?

thanks!

Drives have arrived, I've enabled autoexpand, do I just power down the server and swap the drive?
Anyone? thanks!
 
Drives have arrived, I've enabled autoexpand, do I just power down the server and swap the drive?
Anyone? thanks!

I your disks have id's like c1t0d0 and you have set autoreplace=on this should work.

If you have port independent WWN as disk id like c6t50014FE204EC5FCCd0 this will not work. In such a case you must start a disk replace (napp-it menu disk-replace) old -> new.

If your hardware is hotplug capable you can hot-plugin a new disk, wait some seconds until the disk is detected and do a disk replace. After that you can hot remove the disk. Offline, power-off or unconfigure is not really needed in this case.
 
Technically it is OmniOS and Napp-IT.

OpenSolaris is the name of an OS from Sun years ago.
Illumos was originally a fork of OpenSolaris. Current Illumos/OI/OmniOS is quite different.


about DNS
is your problem a DNS problem (i.e. does ping to any adress work?) or do you have authentication problems against the domain?

In the first case check DNS settings (napp-it menu system-network-dns) or /etc/host settings.

In second case try a smb service restart (or a rejoin domain).
 
I reinstalled last omnios+nappit on the other zfs server (with X9SRH-7TF) and it keep saying that there's no smartmontools installed.
I tried these commands, it downloaded smartmontools-5.40.tar.gz correctly but then returned "No such file or directory" on the next commands, there's an advice of creating them manually in napp-it's smartmontools help page, but how do you create them?

Uh, strange but after a while it seems to work anyway as the page is displayed.
Even more strange, i don't have smart values anymore with the last installation:
CFsGe8.png


But i had them from an older installation:
EqqQbW.png


I tried downgrade to 9d2 (and all the available update/downgrade) but i have the same panel without the smart values :confused:

About the SMB share, i created my root password, but how to connect using the root account from a windows machine? If i go to the ZFS server from the network tab on windows, i can only browse the content as a guest (and can't copy/delete stuff unless i increase the rights but for everyone)
 
Smartmontools behaviours are different on disks or controller, release or OS.
Napp-it tries different settings for each disk to get wanted informations.
But sometimes the output was not as expected or hangs

Therefor it can happen, that the smartmontools menu is not working.
In such a case you may run the smart-detection script manually from CLI like
perl /var/web-gui/data/napp-it/zfsos/_lib/illumos/get-disk-smart-bg.pl

to find and fix the problem. Other option is to use smartmontools from CLI yourself like you would do without napp-it.

The current wget installer compiles a smartmontools 6.2
If you install the old 5.4 over it this may have the result that it works with the old napp-it release.

If you up/downgrade napp-it within menu about-update, this affects only the GUI.
System settings and other tools like smartmontools are only installed/updated on a base setup via wget.


about root
You are requested for login creditentials when you connect a SMB share if
- guest is disabled on the share
- your local Windows account is not a valid OmniOS account

when you are asked for name/password, use root
 
Thanks for explanations!
Is there a way to replace the compiled smartmontools 6.2 with the older version (don't know what version it was at that time) that worked for me?

about shares, can i have at the same time:
-a guest access that can only read but not write, modify or delete
-a root full access?

Because i'm not asked for password i just can't see the dataset from windows if guest is disabled.
 
Thanks for explanations!
Is there a way to replace the compiled smartmontools 6.2 with the older version (don't know what version it was at that time) that worked for me?

about shares, can i have at the same time:
-a guest access that can only read but not write, modify or delete
-a root full access?

Because i'm not asked for password i just can't see the dataset from windows if guest is disabled.

You can either enable guest on shares (no login required) or disable (login always required) The rest is a question of permissions at file or share level


about smartmontools: I found a problem with OmniOS151010 and smartmontools.
I have updated the wget installer and uploaded napp-it 0.9f1 2nd preview:

- wget installer installs smartmontools on OmniOS 15010 (new gcc 4.8 compiler required)
- support for Solaris 11.2
- support for OmniOS 151010 (problem with IO::Tty and Expect fixed)
- Pool-create supports Pool 28v5 (cross platform capable ex Solaris -> OmniOS)
- new menu disks-smartinfo with foreground/background refresh

You may rerun the wget installer and update then online to 0.9f1
 
IT WORKS! :D
BjlJxm.png


CqBl2H.png


About vdev, i wanted to make a pool of 2 vdev each having 12*hdds in raidz2, i only need to make one vdev in "create pool" menu then extend it and make a second and the two z2 vdev will be stripped together automatically?

About L2ARC, i noticed that they need to be added one by one, but if i add two L2ARC ssd for my two vdev based pool, are they going to be shared for each vdev? (one l2arc for one vdev), or put in some sort of big JBOD for the whole pool?

About shares, i finally got the way to make it!
http://youtu.be/KPrgdFsE104
 
About vdev, i wanted to make a pool of 2 vdev each having 12*hdds in raidz2, i only need to make one vdev in "create pool" menu then extend it and make a second and the two z2 vdev will be stripped together automatically?

About L2ARC, i noticed that they need to be added one by one, but if i add two L2ARC ssd for my two vdev based pool, are they going to be shared for each vdev? (one l2arc for one vdev), or put in some sort of big JBOD for the whole pool?

With ZFS you create a pool from a first vdev (i.e. a Raid-Z). You can then add more vdevs to the pool where ZFS stripes them in a Raid-0 manner automatically to increase pool capacity and pool performance together.

L2ARC is not per vdev but per pool. If you add more L2ARC drives they share the load with an increased cache size.

You should think about L2ARC caches only if you cannot add more RAM or if you reached 128GB RAM (there are some stability problems above on current ZFS implementations, Nexenta seems to adress this in their newest NexentaStor) as RAM (ARC) is much fatser than SSD (L2ARC) caches. You can use arcstat to check cache efficiency.
 
Congrats! I discovered napp-it at ServeTheHome and then I came here and saw this huge topic :O

Running beautifully as VM storage.
 
How do I configure the "value_host|" host string in /var/web-gui/_log/napp-it.cfg to accept an IPv6 address?
 
Open menu About-Settings where you can enter an allowed management ip-adress or a string how ip addresses must begin like 192.168.
 
I put in 2001:db8:100:1: and I get the following error. This indicates that this control doesn't understand IPv6. It is truncating at the first colon and not matching on the address.

NOT ALLOWED !!!
Access to this appliance is restricted. IP must contain 2001:db8:100:1: or 127.0.0.1
You can set this value via menu setup or in file /var/web-gui/_log/napp-it.cfg

You used adress 2001 at 16.05.2014, 23:35 00 s
 
This indicates that this control doesn't understand IPv6.

correct, currently ip4 only.
(the parsing is in admin.pl at the beginning)

ip6 needs to be done here and in network settings

update: 17.May
I have removed a check for valid ip4 adresses in admin.pl.
You may reload the 0.9 f1 preview from today.

I have not checked myself as my net is currently ip4 only.
 
Last edited:
Back
Top