unRAID Server Release 5.0-rc1 Available


Recommended Posts

  • Replies 331
  • Created
  • Last Reply

Top Posters In This Topic

Anybody tried this with the AOC-SASLP2?

 

Assuming you're referring to the AOC-SAS2LP-MV8 http://www.supermicro.com/products/accessories/addon/AOC-SAS2LP-MV8.cfm

 

+1 on this question

 

It should work fine.  I have not installed this RC1, but am running a box with 2 of these cards on the latest beta with no issues.  Those cards have not had any issues with any of the beta versions that I am aware of, so I would be shocked if it didn't work with this new version.

Link to comment

Anybody tried this with the AOC-SASLP2?

 

Assuming you're referring to the AOC-SAS2LP-MV8 http://www.supermicro.com/products/accessories/addon/AOC-SAS2LP-MV8.cfm

 

+1 on this question

 

It should work fine.  I have not installed this RC1, but am running a box with 2 of these cards on the latest beta with no issues.  Those cards have not had any issues with any of the beta versions that I am aware of, so I would be shocked if it didn't work with this new version.

 

The AOC-SAS2LP-MV8 is Marvell based, and so far not reported affected.  The Supermicro card that may be affected is AOC-USASLP-L8i and potentially others based on LSi.

Link to comment

After working through my SATA cabling issues (do NOT zip tie any parallel runs together, especially with >2TB drives), I have so far been successfully using 5.0 B14/RC1 with three 3TB and two 4TB drives for the past month or so...

 

How do you mean? I've got three or four cables zip-tied together, do they interfere with each other?

Link to comment
im just stating that a main issue here is the lsi cards.....and it just seems that there is no internal testing being done for them at all....and since there are ALOT of people using the lsi cards it would be a priority......

 

What is it about the LSI cards that you love so much?  They are problematic for a number of reasons. What's more they are easy to replace.  A lot easier than to kludge software to make them work.  Unless you have thousands of them, rip them out and replace with something newer and better. 

Link to comment

im just stating that a main issue here is the lsi cards.....and it just seems that there is no internal testing being done for them at all....and since there are ALOT of people using the lsi cards it would be a priority......

 

What is it about the LSI cards that you love so much?  They are problematic for a number of reasons. What's more they are easy to replace.  A lot easier than to kludge software to make them work.  Unless you have thousands of them, rip them out and replace with something newer and better.

 

the LSI SAS9211-8i is current and actually a very good card

 

http://www.lsi.com/products/storagecomponents/Pages/LSISAS9211-8i.aspx

Link to comment
What is it about the LSI cards that you love so much? ... Unless you have thousands of them, rip them out and replace with something newer and better.

Errr, I only purchased my AOC-USAS2-L8i card a few months ago, and I believed that it was a recent/current model.  Driver support has only been added during the 5.0 beta series.  Are you suggesting that I should throw that (expensive) card away and buy something else?

Link to comment

im just stating that a main issue here is the lsi cards.....and it just seems that there is no internal testing being done for them at all....and since there are ALOT of people using the lsi cards it would be a priority......

 

What is it about the LSI cards that you love so much?  They are problematic for a number of reasons. What's more they are easy to replace.  A lot easier than to kludge software to make them work.  Unless you have thousands of them, rip them out and replace with something newer and better.

 

This has already been addressed in this thread. The LSI cards are very easy to acquire and at extremely affordable prices. There is plenty of support for them not only from different software developers but also the industry in general (hell they're even easy to work with software suites such as ESXi which can be very picky about hardware). I can't tell if your post was just unfounded or purposeful trolling (registered today and posted 1 time total).

Link to comment

I still see these errors in my syslog only on bootup with my S/M SAS cards.  The drives are both WD 2TB Green both have no issues and pre-cleared fine. These drives are on the first card and the only drives hooked to the card. twice for the ATA9 and once for the ATA10 drive.

 

 

Apr 29 07:36:40 MEDIASERVER kernel: ata9: sas eh calling libata port error handler (Errors)

Apr 29 07:36:40 MEDIASERVER kernel: ata10: sas eh calling libata port error handler (Errors)

Link to comment

I still see these errors in my syslog only on bootup with my S/M SAS cards.  The drives are both WD 2TB Green both have no issues and pre-cleared fine. These drives are on the first card and the only drives hooked to the card. twice for the ATA9 and once for the ATA10 drive.

 

 

Apr 29 07:36:40 MEDIASERVER kernel: ata9: sas eh calling libata port error handler (Errors)

Apr 29 07:36:40 MEDIASERVER kernel: ata10: sas eh calling libata port error handler (Errors)

Those are informative messages to let you know the SAS error-handler is being initialized to call the LIBATA port error handler if an error occurs.

 

As far as them being flagged as "errors" in unMENU's syslog...

Anything with the string of letters "error" is highlighted in red.

 

The message could say

No error detected.

and it would be red.

 

It could say

Terror in the mountains is a great movie.

and it would show in "red" since it has the string "error" in the line.

Link to comment

 

It looks like the last kernel that was reliable with the LSI based controllers was the 3.0.x series.  It also appears the 3.0.x series is being actively maintained.  So what I'm going to do is take down -rc1 and rebuild with latest 3.0 kernel (3.0.30) and release as -rc2.

 

I will also say this: not everyone is experiencing the kinds of issues being reported in this thread, and none of these issues affect data integrity (except perhaps for LSI issue).

 

I have so far not experienced any data reliability issues with -rc1, but I guess I'm a little late to the party, just having built a new system with 3TB drives 4 days ago.

 

For what it's worth, I just downgraded to 5.0b14 because with -rc1 I was experiencing write speeds of 449kB/s (yeah, you read right - less than 0.5MB/s).  Reads were fine at about 40MB/s

 

I thought I had a bum motherboard, NIC, network cable, hard drive... but went back to b14 an hour ago and am getting 17MB/s from my first UnRAID server to the new one (via a Win7 middle man).

 

New log is attached, but since I can't get back to -rc1 at the moment, I don't have another one to compare to.

moult_5.0b14_log_after_restart.txt

Link to comment

Here is the list of issues gleened from this thread:

 

-dlandon: clicking utils/system log constantly refreshes using IE9

 

You might just want to hold on this for the moment.  I think I may have spoken too soon.  This worked on the "bare bones" unraid, but later became an issue now that I think back on it.  I will strip all add ons/plugins/customizations off and then add one at a time to see which one is/might be causing the issue.  I violated the rule of testing with "bare bones" unraid and only reporting issues from that.  My bad.  I'll rewind and start over and see if I can find what is causing this to happen.  I think Tom's time is better spent on those "bigger fish".

 

Ok the culprit appears to be the apcupsd plug in.  I install it manually (no other plug ins are installed) and all is good.  I did not start the daemon.  Then I reboot and the constant log refresh starts.  I don't think this one is in your court Tom.

 

I have more information on this issue.  If I rename the .plg to .plug and install in from the go script on reboot, there are no issues.

 

Go script line:

/usr/local/sbin/installplg /boot/config/plugins/Apcupsd-3.14.10-i468-2atv.plug

 

Tom, is there an issue in the plugin installations on boot up?

Link to comment

Built kernel 3.0.30 and testing.  Hopefully get -rc2 out tomorrow...

Also upgraded Samba to 3.6.5, which has the security patches.

 

great news! been on 4.7 since i bought UnRaid, held off cause of LSI issues.

 

Looking forward to trying version 5 :D

Link to comment

After working through my SATA cabling issues (do NOT zip tie any parallel runs together, especially with >2TB drives), I have so far been successfully using 5.0 B14/RC1 with three 3TB and two 4TB drives for the past month or so...

 

How do you mean? I've got three or four cables zip-tied together, do they interfere with each other?

 

With 2TB drives, I had zero issues with zip-tied parallel runs of SATA cables.  But the moment I introduced a 3TB on one of those runs, some time thereafter I started to have random errors and the dreaded RED BALL.  I bought brand new cables (different manufacturer), new drives, swapped existing drives, refurbished drives (thinking a drive was bad) right up to a refurb'd/new motherboard, but eventually I would start to get those dreaded RED BALLS due to some read error, but a subsequent DATA REBUILD would still always succeed completely (the parity check, not always successful).  I had two separate 3-SATA cables zip-tied and when I removed the zip-ties from the first set of 3 cables it completely resolved all RED BALLS on those drives, but then errors started cropping up on the second set.  Removed the zip-ties on the second set and installed brand new cables about a month and a half ago and so far I have not had a single RED BALL since.

 

Crossing my fingers.

 

YMMV...

Link to comment

Built kernel 3.0.30 and testing.  Hopefully get -rc2 out tomorrow...

 

As a brand new unRAID Pro user with an IBM M1015 (LSI) card, I have to say this is going to bring a lot of relief to a great many people.  :)

 

I am glad someone else pointed out that kernel 3.0.x could still be used, though it is darn peculiar that something like this could remain unfixed for so long. I suspect businesses using these cards don't go putting their servers on the latest available kernel... (I use the latest 3.0.x on my ecommerce server)

 

I bought unRAID Pro based on the fact that there was nothing else that could do the job (gave up on FlexRAID).  I purchased a new (server pull) IBM M1015 (LSI based) card via ebay (which may have been as cheap as $65 for some but not me, lol). Somewhat tellingly, the seller told me to stick to beta 12 or 12a if I was going to use it with unRAID. To then read of the growing despair here, I was a bit concerned, to say the least.

 

I know beta means beta, but given that my drives had no protection on their own before (5 drives in a HTPC and 6 drives in a desktop but FlexRAID is far too glitchy yet the author now thinks it's worth charging some serious money for!), I don't have a problem running 12a in the meantime.

 

I haven't had any problems with it at all, although  I will repeat something that has been said here before, SimpleFeatures needs to become a standard part of unRAID. Starting out the default GUI was quite lacking, I ended up installing unMENU and other items until I came upon SimpleFeatures. It is the glossy coating that the powerful unRAID engine deserves.

 

...Donovan

 

PS: The FlexRAID guy goes missing for months at a time too. Must be a thing. ;)

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.