unRAID Server release 4.5-beta8 available

[limetech]

Now I'm thinking...could have bumped Samba to 3.3.9, stable for quite awhile..

Hopefilly in 4.5.9-final.

Yes I should update Samba to that release since there are some fixes for Windows 7.  We'll see...

Dare you try 3.4.x?

Or is that still too raw?

 

Had some problems compiling 3.4.x.  It's a little different also - it includes 4.0 source.  Probably will wait for unRAID 5.0 for this, but looks like there will be a -beta9 with Samba 3.3.9 for the sake of Win7 changes.

[limetech]

I will be releasing -beta8 which has, among other things, these two changes:

- set default scheduler to 'cfq'

Looks like for some reason CFQ did not make it in beta8. It still ships with NOOP as default.

 

So I did some serious testing on my server with the new beta. (copying ~20GB folders with DVD files to the server over samba).

It turns out that by just changing the scheduler to CFQ my write speeds increase by about 20%. Repeated reproduceable results.

 

(Note: My server may not be representative for the average unRAID build: It has a Ultra-Low-Voltage 1GHz Mobile processor, without L2 cache, and only 1GB RAM)

 

As for max_sectors_kb - well this just sets the upper limit on an I/O request to a block device (ie, disk drive).  The default value of '512' means that the largest request the disk driver will get is 512KB (or 1024 sectors).  The largest it "should" see is generally 128K.  I don't see how lowering this to 64 would help performance...

 

No, I did not say 64.  I am setting this to precisely 128K.

And I didn't say that it helps performance.  It helps the overall responsiveness of the server.

 

Anything over 128 is killing my tiny little CPU: copying stuff starts to time-out and abort, processes start getting stuck in disk-wait state, and the server starts freezing for long periods of time -- can't even telnet to it.

 

Again, this behavior may only be typical for my particular configuration.

 

Yours,

Purko

 

At first I did have CFQ enabled, but I saw a decrease in write performance vs. NOOP, which actually is expected since the unraid driver is carefully controlling sequence of I/O's to disk drives.  But, I think I will add a config setting on the Settings page under 'Disk settings' to let you choose the scheduler.  Also you might want to experiment with re-enabling NCQ on that same page.

[purko]

I think I will add a config setting on the Settings page under 'Disk settings' to let you choose the scheduler.

 

On second thought, that's hardly a necessity.  We can easily change it now if we want to.

 

Keep it simple.

 

Also you might want to experiment with re-enabling NCQ on that same page.

 

My NCQ is enabled, and works perfectly.

 

Thanks again for this great build.

 

 

[purko]

BTW, I always meant to ask...

 

Is there a speciffic reason why you would bother to manage when to put disks to sleep,

rather than set up their idle timers and let them take care of themselves?

 

 

[limetech]

BTW, I always meant to ask...

 

Is there a speciffic reason why you would bother to manage when to put disks to sleep,

rather than set up their idle timers and let them take care of themselves?

 

LOL  That's how it originally worked but there are so many drives out there where this feature is surprisingly buggy.

[NAS]

Tom,

 

can I request that Joes excellent cache drive script be added to beta 9. Id love to properly have it beta trialled to see if it works as well for the masses as it does for me.

 

We can always remove it beta 10 if it proves unreliable for the complete unRAID userbase... thats the beauty of betas

 

FYI beta 8 upgrade seemless and so far none of the weird samba dropouts I was seeing before.

[purko]

At first I did have CFQ enabled, but I saw a decrease in write performance vs. NOOP, which actually is expected

since the unraid driver is carefully controlling sequence of I/O's to disk drives.

(Me speaking as someone who doesn't know what he's talking about...)

Then maybe if the unraid driver would not so carefully controll the sequence of I/O's to disk drives, and would leave more of it to the scheduler,

then maybe my CPU usage would not spike so high close to 100% every time I write to the array.

 

I mean, we did have hard disks way back when the CPU was running at 8 MHz.  Right?  And we did just fine.

Even if my 1 GHz CPU is weak by today's standards, it is still a gazillion times more powerful than its 8 MHz grandpa.

 

So why should the writing operations be so processor intensive?  Something's not right.

Could it be that the unraid driver has lots of potential for optimization?

 

Purko

 

 

[SSD]

Just to report that this update still has not increased my write speed. 4.37 GB file took 5:19 to copy. (14 mb average) I guess it's back 4.4.2 until next time. Thanks all the same.

 

DMAddict -

 

Sorry you have not observed an increase in write speed, but it would be very helpful if you would explain your configuration and steps that you went through to test this new beta.  Tom already indicated that certain drive firmware impacted the driver update he made.  So knowing what type of parity drive you are using, its firmware level, what type data drive you were copying to, its firmware level would be helpful in determining likely culprits.  Did you see a variation in copying to different data drives?  What settings did you tweak (if any)?

 

Perhaps with more information, we could make suggestions to enable you to see the improvements others are seeing.  Or perhaps it would lead to other users with similarities to your config narrowing down the issue such that it can be addressed.

 

Be a part of the community and do your part as a beta tester.

 

Thanks!!!

[Joe L.]

I just wanted to report that I loaded b8 and have also seen a remarkable improvement in the write speed. I was getting ~10 to ~12 MB/sec and now I'm getting ~26 MB/sec. The rate appears to be more consistent as well. I have noticed one thing, the "Refresh" button on the http://unraid:8080/unraid_main (unMenu) page doesn't work for me. If I got to the original page, http://unraid/main.htm, it works fine. I did a fresh reboot and started and stopped unMenu with the same results. I have attached my syslog after a fresh reboot, so maybe there is something in there.

The "Refresh" button in the unRAID main page, accessed through the unMENU plug-in will never work in Firefox.  It is a Firefox bug/difference in how they work with an embedded inline frame and how Tom is invoking the refresh button through Javascript.  It will work fine in Windows-Explorer.

 

In the interim, you can use the browser's own refresh button.  It will do the same.   This non-working refresh button has been the same in all versions of unMENU.  I looked into it when I first wrote that plug-in.  Can't fix it, since I can't fix firefox's handling of inline frames and the javascript method Tom is using.

 

Enjoy the new release otherwise.

 

Joe L.

[BRiT]

Even with green drives, there's a remarkable increase in write speeds. System specs: Core 2 Duo P8600 2.4Ghz and 4G DDR2, 3 2TB WD GreenPower drives, 1 1TB WD Black system/cache drive, Slackware-Current (13.0 with patches).

 

When using CFQ and NCQ enabled:

dd if=/cache/.downloads/extracted/Movies/test.mkv of=/mnt/disk1/Backups/test.mkv bs=200K
41674+1 records in
41674+1 records out
8534870199 bytes (8.5 GB) copied, 300.217 s, 28.4 MB/s

 

When using NOOP and NCQ enabled:

dd if=/cache/.downloads/extracted/Movies/test.mkv of=/mnt/disk1/Backups/test.mkv bs=200K
41674+1 records in
41674+1 records out
8534870199 bytes (8.5 GB) copied, 285.021 s, 29.9 MB/s

 

A full Parity check with CFQ and NCQ is the same speed as before:

Nov  7 02:48:37 Reaver kernel: mdcmd (54): check
Nov  7 02:48:37 Reaver kernel: md: recovery thread woken up ...
Nov  7 02:48:37 Reaver kernel: md: recovery thread checking parity...
Nov  7 02:48:37 Reaver kernel: md: using 1152k window, over a total of 1953514552 blocks.
Nov  7 10:02:04 Reaver kernel: md: sync done. time=26007sec rate=75114K/sec
Nov  7 10:02:04 Reaver kernel: md: recovery thread sync completion status: 0

[peter_sm]

I have a great speed to move files from my windows PC to the unRAID, without using a cache disc, so I don,t need this , BUT soon there are Christmas and it would be great if its possible to use this disc as a data disc instead when using PLUS version 

 

 

[boof]

Are any debugging options set to be on in this release? my syslog seems a bit chattier than normal.

 

I can confirm the speed increases also. Increased to an average of ~32 megs a second writes to user shares with spikes up to 45.

 

This is with an array predominantly of 5400 rpm drives (including parity).

[SSD]

I have a great speed to move files from my windows PC to the unRAID, without using a cache disc, so I don,t need this , BUT soon there are Christmas and it would be great if its possible to use this disc as a data disc instead when using PLUS version 

 

The cache disk does not increase read speed.

[BLKMGK]

Nice increase in speed here too - enough so that I have gone ahead and put this beta on my more precious data server. The data transfer appears smoother with less dips and is overall moving along very well!

 

Tom, with these new configuration capabilities would it be worthwhile for some of us to upgrade RAM? I have always had a decent amount in my machines but RAM is fairly cheap these days - any advantage to say 4gig with appropriate config changes?

 

I for one haven't looked much at the various add-ons other than unMenu and I'm proabbly far behind in revs on that. I have been tempted by bubbaRAID and have pondered doing a full Slack install a few times. An easier way to integrate plug-ins would be welcomed, especially if I could have a WEB controlled Torrent client without tons of hoop jumping. <shrug> In the end though for me it's storage and working very well, this update has made me very happy with the speed improvements!

[wholly]

Anyone else having issues with realtek 8169's running gigabit in this (or maybe a previous) beta release?

 

I started out using an 8139 (100mbit) and had no troubles so bought a 8169.  Connected to the same test hub (100mbit half duplex) I have no troubles and it's rock solid.  But if I pull the cable from my cheap hub (coming from my dell gigabit switch) and plug it straight into the 8169, I'm good for about for 4G of transfer and then - poof.  It's not there.  The server thinks it's connected, the link and activity light are active but no response from remote connections.  Disconnecting and reconnecting resets everything.  Going back to the 100mbit link it returns to stability.

 

I'll see if I can't trap something good in the syslog.  But I noted that Beta7 swapped the 8169 driver so I thought I'd ask.

[lewcass]

Anyone else having issues with realtek 8169's running gigabit in this (or maybe a previous) beta release?

 

I had trouble with my on-board Realtek NIC (it's PCI Express 8168) and gigabit. Traced it to a shared interrupt with my USB flash drive. This was with beta 6. Haven't yet tested with 7 or 8.

[wholly]

Finally got some time to try to stream some video off the server and found that the 8169 was losing link every 5-7 minutes even on the 100mbit.  Syslog was very helpful in diagnosing this issue.  Dropped my 8139 back in and smooth as silk for 20 minutes and no lost link.  So...  I'm not regretting my purchase of an Intel NIC from newegg about a two hours ago.

 

The great news is that the 8169 was an 8 dollar NIC.

[purko]

..I don,t need [cache disc].. BUT soon there are Christmas and it would be great if its possible to use this disc as a data disc instead when using PLUS version 

 

+1   

 

 

[vern]

The "Refresh" button in the unRAID main page, accessed through the unMENU plug-in will never work in Firefox.  It is a Firefox bug/difference in how they work with an embedded inline frame and how Tom is invoking the refresh button through Javascript.  It will work fine in Windows-Explorer.

 

In the interim, you can use the browser's own refresh button.  It will do the same.   This non-working refresh button has been the same in all versions of unMENU.  I looked into it when I first wrote that plug-in.  Can't fix it, since I can't fix firefox's handling of inline frames and the javascript method Tom is using.

 

Enjoy the new release otherwise.

 

Joe L.

 

Thanx for the reply. It didn't even occur to me at the time that I was using FF at that moment. Good info though.

 

Thanx!

[speeding_ant]

Loving the new features!! Just wondering... I've added a 160Gb Cache drive and enabled its use on my shares. Unfortunately when I'm copying data over it isn't copying to the cache drive. What am I doing wrong? Do I need to set the minimum size field to less? It is set to 2000000 by default.

 

Cheers

 

Andrew

[cwr]

you need to enable the cache drive for your user shares

 

shares -> <share name> -> Use cache drive: YES

[speeding_ant]

I've added a 160Gb Cache drive and enabled its use on my shares.

 

Yep, done that already 

[NLS]

How do you know it is not using the cache drive?

Other from checking the writes in unRAID main page, everything else looks the same.

 

[damien]

Same thing for me, write speed is now 30 mo/s instead  of 10mo/s

 

thanks

[michael123]

What are recommended settings for 4GB RAM system?

[ I also have 'cache_dirs' script running. ]