unRAID Server Release 5.0-rc1 Available


Recommended Posts

  • Replies 331
  • Created
  • Last Reply

Top Posters In This Topic

May  4 03:58:28 Tower1 kernel: sas: command 0xf74c7180, task 0xde1ed680, timed out: BLK_EH_NOT_HANDLED

 

Nothing nearby in syslog. This just spit out on the console while I happened to be nearby. The system was quiet. The last disk spun down 25minutes before. Mover started 42 minutes after the error.

 

Are others still seeing these things? My syslog isn't really fit for consumption but if these aren't supposed to still be happening I'll do the drill.

 

This is 5.0-rc2-test, upgraded from rc1 and before that b14. There are definitely a few non-standard things running.

Board is a Supermicro MBD-X9SCL-F-O with i3-2100.

Parity plus 5 data disks are on motherboard ports.

4 data disks plus cache are on AOC-SASLP-MV8.

Link to comment

*** Taken Down ***

 

This is a "release candidate".  When/if no major, glaring bugs are found, it will be renamed to "final".

 

 

I have tried to look though the thread to figure out why it was taken down - can anyone tell me the reason?

Link to comment

But that was replaced with the version RC2-test, which is linked to in Limtech's post below and which may soon be replaced with RC2 or RC3 or similar.

 

Here is the link for "unRAID OS 5.0-rc2-test".  If you have been following along in this thread, and you have an LSI controller, or you have a Realtek NIC, please give it a try and let me know what issues you find.  I'm expecting everything to be great.  I'm seeing 90+MB/sec read via Samba (win7), and around 47MB/sec write to cache drive - not happy with the write speed, it should be better going to cache.  Also parity sync/check speed has no slow down on any of my test systems.  Note that I have NOT addressed any issue with webGui appearing to "hang" - if you let it sit long enough, it seems to come back.  I realize this is not ideal and will address in rc3.  For now I want to know if LSI chipset issues are solved and Realtek NIC's still work  ;)  If so, I'll release as "official" -rc2, and then work on the webGui issues.

Link to comment

But that was replaced with the version RC2-test, which is linked to in Limtech's post below and which may soon be replaced with RC2 or RC3 or similar.

 

Here is the link for "unRAID OS 5.0-rc2-test".  If you have been following along in this thread, and you have an LSI controller, or you have a Realtek NIC, please give it a try and let me know what issues you find.  I'm expecting everything to be great.  I'm seeing 90+MB/sec read via Samba (win7), and around 47MB/sec write to cache drive - not happy with the write speed, it should be better going to cache.  Also parity sync/check speed has no slow down on any of my test systems.  Note that I have NOT addressed any issue with webGui appearing to "hang" - if you let it sit long enough, it seems to come back.  I realize this is not ideal and will address in rc3.  For now I want to know if LSI chipset issues are solved and Realtek NIC's still work  ;)  If so, I'll release as "official" -rc2, and then work on the webGui issues.

 

 

Nice, currently downloading and thinking about if I should use this version or the stable 4.7 - please go to appropriate thread and respond so we don't clutter this one. Thanks :)

Link to comment

May  4 03:58:28 Tower1 kernel: sas: command 0xf74c7180, task 0xde1ed680, timed out: BLK_EH_NOT_HANDLED

 

Nothing nearby in syslog. This just spit out on the console while I happened to be nearby. The system was quiet. The last disk spun down 25minutes before. Mover started 42 minutes after the error.

 

Are others still seeing these things? My syslog isn't really fit for consumption but if these aren't supposed to still be happening I'll do the drill.

 

This is 5.0-rc2-test, upgraded from rc1 and before that b14. There are definitely a few non-standard things running.

Board is a Supermicro MBD-X9SCL-F-O with i3-2100.

Parity plus 5 data disks are on motherboard ports.

4 data disks plus cache are on AOC-SASLP-MV8.

 

If you still have the system log I'd like to see it.  I've seen a few of these messages get generated and I have have tracked one down to a time out set too short.  To fix this, I changed in linux/include/linux/blkdev.h

  BLK_MIN_SG_TIMEOUT from 7 seconds to 30 seconds

 

This change is in -rc2 which I'll post today.

Link to comment

Great work. I'm really pleased on the progress here and seeing an RC2. I would have to agree with the above posts suggesting that the focus should really be on getting a stable version 5 with MINIMAL changes, only bugs/errors and actual fixes for known issues that are relevant rather than "nice to haves" and "additionals". If LSI is fixed, then as above the focus should be on:

 

1: NFS reliability and performance/speed

Mentioned above that NFS has been "broken" since a late 5 beta? Presumably it crashes completely? NFS is a must have feature surely given that the performance is so much better than SMB?

 

2: Write speeds

Tom you mentioned that you are still not happy with write speeds and I think this might have been only when writing to arrays with cache drive?

 

3: LSI Support

This appears to be fixed from the testing completed above so far by people running RC2 Test.

Link to comment

May  4 03:58:28 Tower1 kernel: sas: command 0xf74c7180, task 0xde1ed680, timed out: BLK_EH_NOT_HANDLED

 

If you still have the system log I'd like to see it.  I've seen a few of these messages get generated and I have have tracked one down to a time out set too short.  To fix this, I changed in linux/include/linux/blkdev.h

  BLK_MIN_SG_TIMEOUT from 7 seconds to 30 seconds

 

This change is in -rc2 which I'll post today.

http://lime-technology.com/forum/index.php?topic=19926.msg177355#msg177355

Link to comment

Due to an unfortunate occurrence I have to replace a couple drives.  Since I was starting to run low on space I'm considering buying 4TB drives.  What version should I use?

 

I'm running the Supermicro X7SPE-H-D525 MB and the AOC-SASLP-MV8 controller.

 

Rc2 when is posted?

 

 

Link to comment

You would need at minimum 5.0-beta8 or later for larger than 2TB drives.

 

Are you already at 5.x or 4.7?

 

You might want to follow the RC2 thread (once it is released) but sounds like it might be a good candidate.

All depends what the latest is avail and if any bugs are found when you get the 4TB drives.

 

If you are already at 5.x (b8 or greater), you could just stay with that.

Link to comment

Is RC2 up already or still RC2-test?

 

I believe we have RC2-test and are awaiting RC2.

 

That's right.  -rc2-test was a 'sanity' check for LSI controller support.  I'll be posing 'official' -rc2 sometime before midnight :)

 

That is a great news :)

 

Make sure you install OpenSSH and disable Telnet....  improved security  :P

Link to comment

You would need at minimum 5.0-beta8 or later for larger than 2TB drives.

 

Are you already at 5.x or 4.7?

 

You might want to follow the RC2 thread (once it is released) but sounds like it might be a good candidate.

All depends what the latest is avail and if any bugs are found when you get the 4TB drives.

 

If you are already at 5.x (b8 or greater), you could just stay with that.

 

I'm at 4.7.  I'm planning on picking up the drive tonight or tomorrow.

Link to comment

Is RC2 up already or still RC2-test?

 

I believe we have RC2-test and are awaiting RC2.

 

That's right.  -rc2-test was a 'sanity' check for LSI controller support.  I'll be posing 'official' -rc2 sometime before midnight :)

 

Midnight in Europe? 40 mins left. It is 23:20. Getting sleepy.

Link to comment

In other news, we're finally, about to announce a new server product.  This is a 10/14 drive rack mount server.  The "10/14" means there are two configurations: a 'standard' version that supports 10 trayless hot-swap 3.5" hard drives, and an optional configuration that includes additional 4 trayless hot-swap 2.5" bays.  You can install laptop drives or SSD's in these bays.

 

This 3U server is 19" deep and designed for low power and low noise (for example, we offer a 80+GOLD rated PSU).

 

The product page should be ready next week, and we will be offering a substantial "pre-order" discount.

 

I'm just patiently waiting to see what hardware comes out. I'd rather see the software finished so others can use it, but this and the other thread http://lime-technology.com/forum/index.php?topic=17705.msg159449#msg159449 got my attention.

 

If he had that RB-1200 server for $700 a little while ago I would have bought it then.

Link to comment

Here is the link for "unRAID OS 5.0-rc2-test".  If you have been following along in this thread, and you have an LSI controller, or you have a Realtek NIC, please give it a try and let me know what issues you find.  I'm expecting everything to be great.  I'm seeing 90+MB/sec read via Samba (win7), and around 47MB/sec write to cache drive - not happy with the write speed, it should be better going to cache.  Also parity sync/check speed has no slow down on any of my test systems.  Note that I have NOT addressed any issue with webGui appearing to "hang" - if you let it sit long enough, it seems to come back.  I realize this is not ideal and will address in rc3.  For now I want to know if LSI chipset issues are solved and Realtek NIC's still work  ;)  If so, I'll release as "official" -rc2, and then work on the webGui issues.

 

RC2-test.  RC2 not yet available, but Tom has indicated it's very soon.

Link to comment

Hey Tom,

 

Did you widen any IO buffers in the RC?  For the betas my write speed on my WD20EAR(S/X) array across the network was about 20-24MB/s and it was fairly constant.  What I'm seeing now is bursts of fast speed (up to 61MB/s) followed by long lag periods (seconds-minutes) of 0b/s on rc2-test.

 

The drive and computer that I'm copying the data from has not changed (hasn't even been rebooted) since b14.

Link to comment

I have an issue, but do not believe it is UnRaid related.

My Parity drive died on my, I was able to get it replace, (They sent a remanufactured one in its place.) The drive seams fine in every way, but when I add it to the server, I can NOT get the server to start, it just goes back to the unmounted drives status. As soon as I remove it, the drives will mount just fine. (I am running 5 RC1, and I tried mounting the drive as a data drive as well, and also tried B14.)

 

I just ordered 3 new 3TB 7200 rpm drives so hopefully I don't ever run into this issue again.

Right now I have 6200 movies, 33000 tv episodes, 21000 songs, and 20000 Karaoke songs, plus lots of other stuff unprotected.

Is there a trick I can try to get this drive to work? I thought maybe the drive size was reduced when it was remanufactured, but that wouldn't stop it from working as a date drive.

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.