unRAID Server Release 5.0-rc8a Available


limetech

Recommended Posts

  • Replies 418
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

Still not happy with the parity check speed on this build.

 

Last checked on Sat Dec 1 11:26:52 2012 SGT (today), finding 0 errors.

* Duration: 10 hours, 26 minutes, 50 seconds. Average speed: 53.2 MB/sec

 

Hardware specs in my sig.

You can tweak the "tunable" parameters under 'disk settings'. I have made mine 10 times larger (i.e. add a '0' to each value) as a result parity check went 3 hours faster...

 

Thanks!  I'll try this too.  Is there any downside of increasing these values? 

 

Because I have no idea what these settings do, I wonder why the Default isn't already set larger.

Link to comment

Still not happy with the parity check speed on this build.

 

Last checked on Sat Dec 1 11:26:52 2012 SGT (today), finding 0 errors.

* Duration: 10 hours, 26 minutes, 50 seconds. Average speed: 53.2 MB/sec

 

Hardware specs in my sig.

You can tweak the "tunable" parameters under 'disk settings'. I have made mine 10 times larger (i.e. add a '0' to each value) as a result parity check went 3 hours faster...

 

Nice.

 

Last checked on Sat Dec 1 23:09:21 2012 SGT (yesterday), finding 0 errors.

* Duration: 5 hours, 47 minutes, 18 seconds. Average speed: 96.0 MB/sec

 

Remarkable difference !

Link to comment

IIRC those values were a happy median between the older drives and rigs vs memory consumption.

Now with more ram and faster rigs, plus faster drives, there is a benefit to be seen by changing those values.

 

Also the values are set with the minimum requirements in mind. Unraid being able to be ran on less than a gig of ram... With higher values that would cause issues!

Link to comment

It seems t be the version Smb rc8 is using.

 

Looks quite likely, and also easy to check:

Just get the latest samba from the Slackware repo and try with it.

 

Telnet into your box and...

/etc/rc.d/rc.samba stop
wget ftp://ftp.osuosl.org/pub/slackware/slackware-current/slackware/n/samba-*.txz
installpkg samba-*.txz
logger "### samba `smbd -V` installed. ###"
/etc/rc.d/rc.samba start

 

(Edit: fixed the mistyped link)

 

Newbie here, need some help. I too was having issues deleting file and renaming files, etc...

I logged in through putty and typed in exactly that, stopped and started the array and it seems to run much butter and I can delete files again! Thanks!

 

Here is what I don't understand. I am not sure where the file actually got downloaded to. I browsed through the "Flash" share in windows explorer and I am not seeing it there... Can you guys give me some commands on how I should copy the file to a different directory?

 

Also, from what I understand, it will not be installed if I reboot the server, which is why I have to add something to the "go" script? If my understand is correct, can you guys help me out on how I can access this script and what I have to write in it?

 

Thanks so much in advance!

Again, I'm not sure where the file got downloaded to, but I just logged in with putty, then typed root and did everything in "root@unraid~#"

Link to comment

That wget command will download samba-*.txz into whatever directory you were in when you ran it.  I assume it was meant to be run from a directory on your flash drive (/boot/config perhaps).  If you just logged in and ran the command as noted, it's probably in ~, which is the home directory of whatever user you logged in as.

 

I don't believe installpkg actually copies over the plugin to your flash drive, but I may be wrong.

Link to comment

That wget command will download samba-*.txz into whatever directory you were in when you ran it.  I assume it was meant to be run from a directory on your flash drive (/boot/config perhaps).  If you just logged in and ran the command as noted, it's probably in ~, which is the home directory of whatever user you logged in as.

 

I don't believe installpkg actually copies over the plugin to your flash drive, but I may be wrong.

 

Well I didn't run it from any directory, I did just what I said. I can't seem to navigate or get to the file and I don't know the copy command or even where to copy it to or what to do...

 

I can see the file if I type "dir" under root@unRAID:~# prompt

Link to comment

IIRC those values were a happy median between the older drives and rigs vs memory consumption.

Now with more ram and faster rigs, plus faster drives, there is a benefit to be seen by changing those values.

 

Also the values are set with the minimum requirements in mind. Unraid being able to be ran on less than a gig of ram... With higher values that would cause issues!

Can you give a idea where to set them to with 16 GB of memory ?

Like stated above, mulitply with 10 ?

Link to comment

That wget command will download samba-*.txz into whatever directory you were in when you ran it.  I assume it was meant to be run from a directory on your flash drive (/boot/config perhaps).  If you just logged in and ran the command as noted, it's probably in ~, which is the home directory of whatever user you logged in as.

 

I don't believe installpkg actually copies over the plugin to your flash drive, but I may be wrong.

 

Well I didn't run it from any directory, I did just what I said. I can't seem to navigate or get to the file and I don't know the copy command or even where to copy it to or what to do...

 

I can see the file if I type "dir" under root@unRAID:~# prompt

 

If you're logged in via telnet or SSH, you're always in a directory, so you did run it from a directory.  Based on what you said here, I was correct in assuming you downloaded it to the home directory of the root user (see the part after the colon in the prompt you posted, that's what directory you're in).  The tilde means "home directory of the currently logged-in user", so you're in the home directory of root based on root@unRAID.  I don't know what directions you were following since I haven't replaced samba, but I'm certain you don't want it in root's home directory.  I'd guess you want it in /boot/config/plugins or /boot/packages, but I'm not confident without seeing the full instructions you're following.

 

As for the commands, this is basic Linux stuff, so you might want to read up a bit on how the Linux/UNIX shell works before poking around since you can potentially cause a lot of damage if you're not sure what you're doing:

ls - directory listing

cp - copy

mv - move

rm - delete

 

Edit: I kind of figured out what you're working on.  The posted commands, as you mentioned, only work until rebooting.  So, copy the downloaded file onto flash:

cp samba-*.txz /boot/config/packages/

I chose packages because this isn't a true unRAID plugin, but someone may have a better idea of where to put it.

 

Then you need to open your go script in flash/config/go (or from within telnet at /boot/config/go), and add lines similar to the commands you were following previously:

/etc/rc.d/rc.samba stop
installpkg /boot/packages/samba-*.txz
logger "### samba `smbd -V` installed. ###"
/etc/rc.d/rc.samba start

Then save and close your go script and restart.

 

Big Note: I have not tested any of the above.  It's a good guideline for what you need to do, but there may be errors.  Be careful.  You could also copy that block of commands you were following verbetim into your go script, but it would need to download samba every time you rebooted.  If you ever tried to reboot without an internet connection, or if the file moved, you'd have errors when starting up.

Link to comment

Still not happy with the parity check speed on this build.

 

Last checked on Sat Dec 1 11:26:52 2012 SGT (today), finding 0 errors.

* Duration: 10 hours, 26 minutes, 50 seconds. Average speed: 53.2 MB/sec

 

Hardware specs in my sig.

You can tweak the "tunable" parameters under 'disk settings'. I have made mine 10 times larger (i.e. add a '0' to each value) as a result parity check went 3 hours faster...

 

Nice.

 

Last checked on Sat Dec 1 23:09:21 2012 SGT (yesterday), finding 0 errors.

* Duration: 5 hours, 47 minutes, 18 seconds. Average speed: 96.0 MB/sec

 

Remarkable difference !

 

dalben. Any chance you could inform me of how full your data disks are? I see in your sig that you have about 8TB of data drives. I was wondering if those were nearly full to reach those speeds, or if they were about half empty on each drive or what.

 

Thanks in advance.

Link to comment

Still not happy with the parity check speed on this build.

 

Last checked on Sat Dec 1 11:26:52 2012 SGT (today), finding 0 errors.

* Duration: 10 hours, 26 minutes, 50 seconds. Average speed: 53.2 MB/sec

 

Hardware specs in my sig.

You can tweak the "tunable" parameters under 'disk settings'. I have made mine 10 times larger (i.e. add a '0' to each value) as a result parity check went 3 hours faster...

 

Nice.

 

Last checked on Sat Dec 1 23:09:21 2012 SGT (yesterday), finding 0 errors.

* Duration: 5 hours, 47 minutes, 18 seconds. Average speed: 96.0 MB/sec

 

Remarkable difference !

 

dalben. Any chance you could inform me of how full your data disks are? I see in your sig that you have about 8TB of data drives. I was wondering if those were nearly full to reach those speeds, or if they were about half empty on each drive or what.

 

Thanks in advance.

 

Here you go

 

Subject:unRaid Resync Notification

Status update for unRAID  tdm

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Status: The unRaid array is resync/rebuilding parity.

 

 

Server Name: tdm

Server IP:

Date: Sat Dec  1 01:47:03 SGT 2012

Filesystem          1K-blocks      Used Available Use% Mounted on

/dev/sdf1              3949536    195328  3754208  5% /boot

/dev/md1            1953454928 1477114888 476340040  76% /mnt/disk1

/dev/md2            1953454928 1167314016 786140912  60% /mnt/disk2

/dev/md3            1953454928 1133934616 819520312  59% /mnt/disk3

/dev/sde1            488371640  3226108 485145532  1% /mnt/cache

shfs                5860364784 3778363520 2082001264  65% /mnt/user

Link to comment

Anybody else experiencing web interface timeouts with this rc? Is this a known problem with this release? Happens on both of my unRAID systems.

 

The web interface becomes unreachable though shares are functional. It seems to happen after/during a parity-check/rebuild.

 

I went to update a few plugins today and my web management page was timed out as well.

 

After a reboot, it still won't come up, and now my array won't mount :(

 

 

Edit: I should clarify.  It's not a timeout, I get a "Page sent no data" error in my browser.

Link to comment

Thanks dalben.

 

btw does anyone know the maximum amount of memory that version 5 supports and what kind of increase i would see in parity calculations, parity check, rebuild, etc were i to upgrade my memory or processor?

unrAID's kernel is compiled with PAE, (  http://en.wikipedia.org/wiki/Physical_Address_Extension ) therefore the max memory is 64GB. (with a 4GB per process limit)

 

You will barely see any difference in a memory upgrade, and unless the processor is a really low-powered one, you'll not see much difference in parity calcs.  Those are disk I/O bound operations, not memory or CPU bound.

 

 

Link to comment

I did update the tunable parameters to 10x and that speeded up the parity calculation.

However it gave me issues with a rebuilt when replacing a drive, as there seemed no low memory enough eny more.

I now changed the values to 5x the original and have run a parity check with average speed of 93,5 Mbyte/sec.

But the speed increase can also be because i removed the last 1 TB hdd, and i could see the parity check speed change as it went on. 0-1000 GB, 1000 - 1500, 1500-2000, 2000-3000.

This matches the generation of hdd's and you see the increased speed with newer generations.

Link to comment

It would be nice if the fine tuned attributes could be set for the different tasks.  i.e. parity check seems to accept the 10x values on my rig (8Gb ram) but it looks like they should be dropped for a parity rebuild.

 

I wonder if a plugin could do this or whether the technology means we're stuck with one value with manual changes (if we remember)

Link to comment

I just noticed that when doing a Shutdown, I see a message on my server screen that says ...something about not saving the Syslog file, then I see ...Consider installing Automatically zip Syslog...and then it goes by so fast I can't see exactly what is causing this message.

 

Is there a add-on or something that should be in my GO file to tell it to zip the current Syslog?

 

I'm also curious to see what others have in their GO file.  Mine is very small, just starts the webgui and installs the new Samba, that's it.

 

Link to comment

Recently I was playing around with Simple Features, looking through  the logs using the log viewer in that product.  I found that I was only running Samba 3.6.7 even though I had downloaded the 3.6.8 package, copied it to a directory (/boot/custom) on my flashdrive and added the following line to my go file.

 

installpkg /boot/custom/samba-3.6.8-i486-1.txz

 

From my first reading of this thread, I thought that this was all I had to do.  But in the back of my mind was the thought that samba always had to be stopped for any changes to be applied. 

 

I did a bit more searching through the thread and pulled some information from several different posts.  I modified my go file to read as follows:

 

# custom install of updated samba package
/etc/rc.d/rc.samba stop
installpkg /boot/custom/samba-3.6.8-i486-1.txz
/etc/rc.d/rc.samba start

 

After restarting (rebooting) unRAID, the samba logs now indicated that version 3.6.8 was running. 

 

I am definitely not a Linux expert but I had some experience with Unix (twenty-five plus years ago) and with setting up a samba server on a Linux machine (about six years ago) so I did have some inkling about how things work.  I also gleaned some information for other posters in this thread and should really credit them but I am afraid that I will miss someone so I will not attempt to do so. 

 

Perhaps I should go into greater detail about why this approach works or perhaps there is another way to accomplish the same thing.  But, with my limited knowledge of Linux, I decided it would be better to let  someone do that!  ;)

I think Switchblade stumbled on a way to streamline this. If you install samba before starting emhttp then there is no need to stop it first because it hasn't been started yet, and no need to start it because emhttp does it for you. See this post.

 

# custom install of updated samba package
installpkg /boot/custom/samba-3.6.8-i486-1.txz

# Start the Management Utility
/usr/local/sbin/emhttp &

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.