I am finally getting around to  using my GIGABYTE GA-EP45-UD3P m/b for an unraid server. I bought 4 1.5tb samsung ecogreen drives and ran the pre_clear script on two of them (took around 30 hours for two passes) and I am running it again on the second two (could not find all my sata power connectors the first day so I only ran the two).

 

 

But what I am wondering is can I get to the unraid start page from the file server itself? I have a monitor attached to it and wanted to look more at it there as it is in my basement right now. Everytime I search I see you need to get to it by typing //tower , but I am at the local/file server machine so how do I do that if I dont have a web browser on it. Or do I and just dont know it?

 

Thanks

 

 

You need another machine with a network connection and browser to access the emhttp browser based front end.

I suppose you could do it with lynx, but that is not installed on the console.

 

Joe L created a cooL shell based script to provide readability of the unRAID array status.

http://lime-technology.com/forum/index.php?topic=153.msg897#msg897

 

Joe Maybe you could post this on your google code page too.

Thanks for the info...

 

I have a switch and was going to hook it all up so I could check it out, but thought it would be nice to be able to do everything from the server and then run the wires when I was done. maybe I will just take a laptop down to it and then connect it up there so I can be close.

I could not connect to //tower

 

But I think it was because I booted up without a network connection? I started my other to drives on pre clearing so I just hooked up the cable and the switch and I can get out to the internet and all but not to the unraid server.

 

Or do I have to start someting else on the server before I can see the box?

 

 

You hooked up the network cables after unRAID was already running. It didn't have a chance to detect the network. Did you manually setup a ip address in unRAID or are you running a router with a DHCP server? That would make a big difference on telling you how to proceed. If you are hooked up to a router which it sounds like you are since you can reach the internet with your other machine then you shouldn't have many problems. Finish pre-clearing your drives then restart unRAID. It should acquire a ip address & you should be good to go..

You hooked up the network cables after unRAID was already running. It didn't have a chance to detect the network. Did you manually setup a ip address in unRAID or are you running a router with a DHCP server? That would make a big difference on telling you how to proceed. If you are hooked up to a router which it sounds like you are since you can reach the internet with your other machine then you shouldn't have many problems. Finish pre-clearing your drives then restart unRAID. It should acquire a ip address & you should be good to go..

 

Since I had two drives already pre cleared I shut down the server, I upgraded to 4.5 beta6 and restarted the server. As far as I know everything is good on that end. I see the lights near the lan connection lit up and on the other end I detect a connection on the switch.

 

I have it this way:

Internet -> DSL Modem -> Linksys wrt160n router -> D-link dgs 2208 switch ( then my desktop  hooked into the switch too.)

 

I did not manually set the address, the router is set to dhcp, but the router connects to the switch as does my desktop and the unraid server.

 

Also I did ifconfig  on unraid and it said up was 127.0.0.1, my router is 192.168.1.1 and is supposed to assign ports 192.168.1.100 thru 115.

 

Any help would be greatly appreciated. I am guessing it is just a configuration issue on my part, but I though everything would be auto detected.

 

 

 

Also I did ifconfig  on unraid and it said up was 127.0.0.1, my router is 192.168.1.1 and is supposed to assign ports 192.168.1.100 thru 115.

 

Any help would be greatly appreciated. I am guessing it is just a configuration issue on my part, but I though everything would be auto detected.

127.0.0.1 is a local alias for the local machine.  It exists for EVERY machine.  It does not indicate anything about a connection to the LAN.

 

Type

ifconfig eth0

 

and

 

ethtool eth0

 

They will look a lot like this:

root@Tower:~# ifconfig eth0

eth0      Link encap:Ethernet  HWaddr 00:11:11:75:fb:7e

          inet addr:192.168.2.100  Bcast:192.168.2.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:419126 errors:0 dropped:0 overruns:0 frame:0

          TX packets:125868 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:1000

          RX bytes:111334054 (106.1 MiB)  TX bytes:105651378 (100.7 MiB)

The IP address of my server is in BLUE above.

 

root@Tower:~# ethtool eth0

Settings for eth0:

        Supported ports: [ TP ]

        Supported link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Supports auto-negotiation: Yes

        Advertised link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Advertised auto-negotiation: Yes

        Speed: 1000Mb/s

        Duplex: Full

        Port: Twisted Pair

        PHYAD: 0

        Transceiver: internal

        Auto-negotiation: on

        Supports Wake-on: umbg

        Wake-on: g

        Current message level: 0x00000007 (7)

        Link detected: yes

 

I have a valid link to my router, and the speed is 1000Mb/s, also as shown in blue above in the ethtool output.

 

What do the equivalent commands show on your server?  Nowhere does it give a "range" of assigned addresses... is that something you are seeing on your router?  (I'm confused where you saw that)

 

Joe L.

The range of addresses is what I set my router to assign for dhcp.

 

Well I went downstairs to the server to run the commands to see what I got this time around. But for the heck of it I went into the bios to make some changes. I disabled things I did not need and realized that I still had both onboard ethernet connectioned as set to enabled so I decided to disable one of them and for the heck of it I disabled the oneI was using and reenabled it.

 

This time around I got what you showed. I see an inet address when I enter the ifconfig eth0 command. and now I see my address as 192.168.1.113 on eth0.

 

And when I got back upstairs my computer was displaying that it saw a new connection and sure enough I can get to \\tower now.

 

I was starting to think that adding the switch to my setup was partly to blame, but all is well now so I am happy happy happy... now to explore the interface some more.

 

 

Thanks for the help and suggestions everyone..

 

 

I did have another question that someone that read this can probably answer of the top of their head.

 

Does the Parity drive need to be cleared?

 

I have all 4 of my drives spun up but only the first two made it all the way through clearing. But I see that I can select any of the 4 drives for the devices and parity.

 

for now I am running the free version of unaid so I can only use 2 data drives. I will decide which version I want to spring for this week based of how well I like everthing.

I did have another question that someone that read this can probably answer of the top of their head.

 

Does the Parity drive need to be cleared?

 

I have all 4 of my drives spun up but only the first two made it all the way through clearing. But I see that I can select any of the 4 drives for the devices and parity.

 

for now I am running the free version of unaid so I can only use 2 data drives. I will decide which version I want to spring for this week based of how well I like everthing.

No drive "needs" to be pre-cleared...  Until you assign a parity drive, unRAID will not clear disks at all, even if not pre-cleared.

 

Once you assign a parity drive unRAID WILL automatically clear any newly added data drive, unless it has the special signature my pre-clearing script puts into place.

(The signature lets it know the drive already has all zeros written to it, so it does not need to do it again.  This lets you have just a few minutes down-time instead of a hours)

 

As far as the parity drive... it is never cleared by unRAID, it is entirely written by it.  If you install it, calculate parity once, and then do a manual parity check it will not be equivalent to the pre-clear script as it is still very possible for it to have sectors pending re-allocation that were unable to be read in the parity check phase.  Typically bad sectors are found when they are unable to be read...  They are re-allocated the next time the sector is written.  (Because at that time the drive knows the correct contents of the sector and can write it to the re-allocated sector.) 

 

Unfortunately, calculating parity for the first time only writes to the parity drive.  Then, checking parity does a full "read" of the drive... But that could still result in sectors needing re-allocation, if it detects sectors that could not be properly read back. It would re-allocate them if a subsequent "write" would occur, but no subsequent "write" will occur in most situations. 

   

I recommend you pre-clear the parity drive, before assigning it as the parity drive... not to save time, but to prove it is working properly and to give it the chance to re-allocate any flaky sectors before you start trusting the protection of your data to it.

 

No matter what you do, I suggest you do a manual parity check after calculating parity for the first time.  That way you know the data was written and can be read.

 

Joe L.

Thanks a million....

 

I did just that, I assigned one of the precleared drives as parity and the other as drive 1. Once the parity goes through its thing I will do a manual check. And then once the other drives finish clearing I will assign one of them as I am limited to 2 data drives in the evaluation version. I was checking and my drive I initialy set for parity was several bytes smaller than the rest so being anal about it I went to the server and switched it so my parity would be the sda drive. lol..

 

I am hoping to bump it up to 8 data drives at first, so I am guessing I need to spring for a pro license.

 

How long does it take to mount drives into the array typically?

 

thanks again, I appreciate you taking the time on a Sunday to answer my questions...

After the parity sync is done, do I do the manual check then format that hard drive or format it first then check it?

 

 

It will only take about 30 seconds to format the drive once it is assigned to your array and shows as "Unformatted" on the management console. 

I would do the formatting first, but it really doesn't matter.

 

Joe L.

How long should a parity check take?

 

 

I started one and it says 300+ minutes to finish. I did it from the main page , but then I thought you said manual parity check, but I am not sure what to do on that or if what I did was right.

 

 

Thanks

How long should a parity check take?

 

 

I started one and it says 300+ minutes to finish. I did it from the main page , but then I thought you said manual parity check, but I am not sure what to do on that or if what I did was right.

 

 

Thanks

It depends entirely upon the number of disks, the size of the disks, and the speed of your MB bus.

 

On my older PCI bus based server, with 10 data drives, the largest being 1TB, and smallest 400Gig, it takes about 14 hours. (If I press the refresh button after a parity check is in progress for a few minutes it shows roughly 1900 minutes to complete.  That is not accurate, as it speeds up greatly once the check gets past the size of my smaller disks. )

 

Your 300 minutes indicates you have only a few disks, or fast PCIe throughput, or both.  It does not sound wrong.  If all your disks are of the same size, it might be pretty accurate.

What does your management console show for the estimated speed?  Most newer PCIe based MB and SATA-2 disk controllers get rates between 50 and 70MB/s.

 

What you did is correct.

It was for just the one drive since I am on the free version until I order later this week.

 

The drives (parity and the 1 data) are both 1.5TB samsung ecogreen f2eg drives 5400 rpms drives.

 

My motherboard is the GIGABYTE GA-EP45-UD3P.

 

the management console showed like 305 minutes or so to start.  Could I have a configuration issue that is slowing me down?

 

 

It was for just the one drive since I am on the free version until I order later this week.

 

The drives (parity and the 1 data) are both 1.5TB samsung ecogreen f2eg drives 5400 rpms drives.

 

My motherboard is the GIGABYTE GA-EP45-UD3P.

 

the management console showed like 305 minutes or so to start.  Could I have a configuration issue that is slowing me down?

 

 

Your time does not seem unusual for 1.5TB drives, but you did not answer my question.

 

What "Estimated speed" do you show?

 

 

What "Estimated speed" do you show?

 

My estimated speed shows this, Is this a normal range?

 

Total size: 1,465,138,552 KB

Current position: 38,566,168 (2.6%)

Estimated speed: 66,906 KB/sec

Estimated finish: 355.3 minutes

Sync errors: 0

 

 

 

What "Estimated speed" do you show?

 

My estimated speed shows this, Is this a normal range?

 

Total size: 1,465,138,552 KB

Current position: 38,566,168 (2.6%)

Estimated speed: 66,906 KB/sec

Estimated finish: 355.3 minutes

Sync errors: 0

The answer is on the Benchmarks page in the wiki here

Thanks..

 

looks like nearly 70Meg not bad compared to others on the benchmark page. It was early on so dont know what it wound up being later as I left for work. I did see some forced NCQ to disabled I wonder if that will improve speeds? I have AHCI enabled.

 

Could I be limited by the fact that I am using samsung 5400 rpm green drives since they are not as fast?

 

Thanks again... I am getting addicted to this, now I just need to read and see how to install unMenu and order 4 more drives and get the pro version of unraid.

 

 

Looks like you are starting to enjoy yourself.. 

 

Just remember as you add drives your average speed will go down...

 

Welcome to the unRAID group.....   

 

Edit: I don't think the 5400 rpm drives are slowing you down that much. You are mostly restricted by I/O. As you add drives the drives will actually be waiting on the controller as this is the bottleneck...

looks like nearly 70Meg not bad compared to others on the benchmark page. It was early on so dont know what it wound up being later as I left for work.

 

The syslog (located at /var/log/syslog or /var/log/syslog.1) should contain the actual averaged number of the parity check.

 

Hopefully the benchmark page contains the numbers from the syslog and not the initial values. They, initial and final, can vary quite a bit. I'll explain and share my experience.

 

In my situation, the initial parity check numbers are the peak values ~100MB/s. The overall average parity check speed is ~75MB/s. The first drive tracks are the fastest, hence the higher values initially. I'm running a small unRAID 4.5beta6 system with 3 of the Green WD 2TB drives (5400rpm). They're on the built-in SATA ports using AHCI with NCQ disabled. There is about 1.3TB of data on the array. It's installed on top of a Slackware current 2009-08-14 image.

 

An aside: Some may dislike the WD green drives and say they're slow. They're plenty fast for me as their primary usage is reading, not writing. They're also low(er) power. I'm seeing write network transfer speeds around 13 - 14 MB/s to the User Shares. That seems to be around the same performance range of those with non-green/7200 rpm drives.

 

Initial numbers from web-page (today's parity check):

Total size: 1,953,514,552 KB
Current position: 	30,151,180 (1.5%)
Estimated speed: 99,998 KB/sec
Estimated finish: 320.5 minutes
Sync errors: 0 	

 

Web page numbers refreshed (today's parity check):

Total size: 1,953,514,552 KB
Current position: 126,700,952 (6.4%)
Estimated speed: 100,157 KB/sec
Estimated finish: 303.9 minutes
Sync errors: 0 	

 

Final numbers from syslog:

Aug 15 16:43:13 Reaver kernel: mdcmd (18): check
Aug 15 16:43:13 Reaver kernel: md: recovery thread woken up ...
Aug 15 16:43:13 Reaver kernel: md: recovery thread checking parity...
Aug 15 16:43:13 Reaver kernel: md: using 1152k window, over a total of 1953514552 blocks.
Aug 15 23:56:02 Reaver kernel: md: sync done. time=25959sec rate=75253K/sec
Aug 15 23:56:02 Reaver kernel: md: recovery thread sync completion status: 0