unRAID Server Release 5.0-rc12 Available

[WeeboTech]

I use a 3TB drive as my cache drive. It acts as a warm spare.

 

One disadvantage of using a spare drive as cache is that you still have to run a lengthy preclear before it can be put into service.

 

It would be ideal if you could keep a precleared drive sitting on the shelf (or installed, but unassigned), ready to be put into service at a moment's notice.

 

This is one reason why I have always wished for better USB support. You would never want to put a disk into service with USB but its a port you never run out of.

 

 

With USB 3.0 and some of these nice JBOD array disks, USB is a viable alternative in regards to speed.

not sure about temperature, SMART and spin control, but USB 3.0 speed rivals eSATA in some respects.

[jumperalex]

I use a 3TB drive as my cache drive. It acts as a warm spare.

 

One disadvantage of using a spare drive as cache is that you still have to run a lengthy preclear before it can be put into service.

 

It would be ideal if you could keep a precleared drive sitting on the shelf (or installed, but unassigned), ready to be put into service at a moment's notice.

 

A good point, and is in fact how I'm running.  However not everyone has that luxury.  So a good middle ground is to buy a cache drive that can double as an in-house spare.  Sure you have to run a pre-clear, but the way i see it since the drive has been in use for a bit you know you've likely made it past the infant mortality stage.  Thus I wouldn't feel the need to run multiple passes, just one to verify the full platter and prep it for insertion into the array. Then you still get the advantage of getting some breathing room to source a drive at the right price, not pay for overnight shipping AND then accomplish a preclear ... multiple ones at that ... all the while your array is actually protected.

[distracted]

I use a 3TB drive as my cache drive. It acts as a warm spare.

 

One disadvantage of using a spare drive as cache is that you still have to run a lengthy preclear before it can be put into service.

 

It would be ideal if you could keep a precleared drive sitting on the shelf (or installed, but unassigned), ready to be put into service at a moment's notice.

 

Why would you need to run a preclear before using it to replace a failed drive? Not trying to be sarcastic. From my understanding this isn't necessary. Am I mistaken?

[steini84]

You would not. Just unassign as cache and rebuild the failed drive to it.

[jumperalex]

I am assuming that when you replace a failed drive ( as opposed to installing an additional new data drive) that unraid does not perform a full zero-write to the drive.  Based on that, even though it takes more time, the reason to pre-clear is to give the full drive (read: the entire platers(s)), a quick scan with a pre/post SMART report.

 

If UnRaid does in fact still do a zero-write before rebuilding, well then a pre-clear is the same thing, only you also get the pre/post SMART report.

[optiman]

My server webgui stopped responding today, not sure why.  I could still access my data, but no webgui.  So i stopped samba and umounted the drives, and did a reboot from the command line.  When it came back up it started a parity check (not sure why).  It took a little while, but the webgui came up and I could see the parity check status.  I returned a little later to find that the webgui is not responding again.  I can hear the drives so I assume the parity check is still running.  I can access my data.  I don't want to interupt the parity check now, but need a way to know when it has completed (and if there were any sync errors.  Is there a safe command line that I can use to just get the status of the existing running parity check?

 

Once the check is complete, I will reboot and start looking at logs to see what is going on.

[brandon]

@Lappen, I will PM you later this evening, tide up at the moment.

PM me with where you are storing your apple db (location), so I can adjust the instructions for you.

I will get you squared away.

 

Maybe you can post the steps to the group on one of the other threads so anyone else who wants to do this won't have to PM you.

 

This would be great.  Also having the same problem.

[WeeboTech]

I am assuming that when you replace a failed drive ( as opposed to installing an additional new data drive) that unraid does not perform a full zero-write to the drive.  Based on that, even though it takes more time, the reason to pre-clear is to give the full drive (read: the entire platers(s)), a quick scan with a pre/post SMART report.

 

If UnRaid does in fact still do a zero-write before rebuilding, well then a pre-clear is the same thing, only you also get the pre/post SMART report.

 

It's a write of the failed drive to a replacement drive. Preclear is not necessary. It's more of a confidence test in this case.

if it were a brand new drive, I would preclear it. If I had a cache drive that was to do double duty as a warm spare. i would have precleared/tested that drive before making it the cache drive.

 

The preclear is only necessary when inserting a brand new drive and expanding the array size. Not for replacing a drive in place.

 

Your other choice is to exercise the cache drive with smart and badblocks. smart -a, save log do a smart -t long (wait a long time), do a full badblocks in read mode, then smart -a saving log and diffing to prior log. Look for pending sectors or large number of reallocations.

 

if you are confident in the cache drive, you can move data from it via mover, unassign it, assign it in place and rebuild onto it.

[cj0r]

Yep, I give my warm spares a gauntlet's worth of preclears before I even begin using it. Hell I do that even with drives that I don't plan on using in my server.

[PeterB]

Why would you need to run a preclear before using it to replace a failed drive? Not trying to be sarcastic. From my understanding this isn't necessary. Am I mistaken?

Okay, not absolutely necessary, but how much confidence do you have in the substantial part of the drive which has not been accessed while in service as a cache drive?

 

The other problem with this regime is that I, like many others, use the cache drive as non-volatile storage for the apps I have running on the server.  So, it would be necessary to hold a spare cache drive.

[distracted]

Why would you need to run a preclear before using it to replace a failed drive? Not trying to be sarcastic. From my understanding this isn't necessary. Am I mistaken?

Okay, not absolutely necessary, but how much confidence do you have in the substantial part of the drive which has not been accessed while in service as a parity drive?

 

The other problem with this regime is that I, like many others, use the parity drive as non-volatile storage for the apps I have running on the server.  So, it would be necessary to hold a spare cache drive.

 

I'm guessing when you say parity drive you mean cache drive. Personally I test every drive I purchase, so in my mind it would be good to go. Plus I would run a parity check immediately after rebuild, and I would check smart before the rebuild and after the parity check for any anomalies. However, I can understand why you may want to run a preclear. There is nothing wrong with a little peace of mind.

[jumperalex]

Why would you need to run a preclear before using it to replace a failed drive? Not trying to be sarcastic. From my understanding this isn't necessary. Am I mistaken?

Okay, not absolutely necessary, but how much confidence do you have in the substantial part of the drive which has not been accessed while in service as a parity drive?

 

The other problem with this regime is that I, like many others, use the parity drive as non-volatile storage for the apps I have running on the server.  So, it would be necessary to hold a spare cache drive.

 

Ideally yes you would just have a spare to use to avoid cannibalizing your cache.  But if you don't, and push comes to shove, it may be better to forego your app data residing on the cache drive to get your array data protection back up ASAP.

 

As for confidence, again, a preclear or weeboo's regimen would give me plenty of confidence since a cache drive with some miles is certainly past the infant mortality stage.

 

Everyone has the level of risk and uptime they can I've with and afford, and none are wrong.  The point is there is not a single right way and considering a cache drive as a warm spare is certainly reasonable one.

 

Sent from my Nexus 7 using Tapatalk HD

[Dougy]

having thoroughly discussed simple features, how to configure routers and now the merits of a cache drive for 20 pages, can anybody summarize what has been discussed specially in relation to "Re: unRAID Server Release 5.0-rc12 Available"?

 

 

[Helmonder]

Summarized: use it, its the best v5 version.

[jowi]

Summarized: use it, its the best v5 version.

I disagree. V5.0-rc5 for me is still the best and most stable performer...

[PeterB]

Summarized: use it, its the best v5 version.

 

... except for stale nfs file handles, in which respect rc10 was far better.

[smakovits]

I have a general question of this version, but it relates tot he past few as I recall.

 

My mobo is running a Realtek 8111C and some time back I had an issue where my unRAID server would simply lock up when using the onboard NIC.  As a result, the fix was to stop using the on-board and switch instead to an add-in card.  I ran this way since, but I recall a few revisions back that things changes in the drivers for realtek.  Are all the Realtek woes fixed these days allowing me to return to on-board? 

 

I know there is mention about performance and add-in NIC vs on-board, but I don't know the extent and if it is significant enough, but still thought I would ask.  The thought is simply that if there is no need for the add-in NIC any more then I would remove it as it might yield slight performance can, but I could potentially be way off base with all of this.  Thanks

[RobJ]

Once the check is complete, I will reboot and start looking at logs to see what is going on.

 

I'm sure I don't need to remind you that you want to grab those logs BEFORE you reboot!

 

By the way, have you checked a current syslog?  It may have clues about your current problems...

[optiman]

Parity check completed.  I was only able to tell because the drives all spun down and things got quiet.  I did a Powerdown command.  After the reboot, everything looks fine.  I do have the logs, and will look at them later.  I have a feeling I know what caused the issue.  I posted the questions in the Plex thread, and found out that I need to stop PMS before updating it.  There is a known issue with it causing webgui issues if you don't do that.  A clean Powerdown is the fix, and that all worked for me.  My system did a parity check because it wasn't a clean powerdown.

 

I may install unmenu so I have another webgui to work with if the SF one dies. 

 

Sorry, most of this is plugin related, and has nothing to do with rc12a, so I'll stop here and go back to the other thread.

[PeterB]

My mobo is running a Realtek 8111C...

It would be good if someone with direct experience had replied, but ...

 

It is my understanding that all but a small number of older Realtek LAN chipsets are working with the current build.

[TheDragon]

I've posted this here in case this is relevant to Tom and this RC.

 

Myself, and others, have encountered a problem with refreshing of shares in Windows - the problem has been detailed in this thread

 

http://lime-technology.com/forum/index.php?topic=25267.0

[optiman]

My mobo is running a Realtek 8111C...

It would be good if someone with direct experience had replied, but ...

 

It is my understanding that all but a small number of older Realtek LAN chipsets are working with the current build.

 

RC12a works fine with my rig and I have the RTL8111C

[TheDragon]

My mobo is running a Realtek 8111C...

It would be good if someone with direct experience had replied, but ...

 

It is my understanding that all but a small number of older Realtek LAN chipsets are working with the current build.

 

RC12a works fine with my rig and I have the RTL8111C

 

+1 

[madshi]

Just updated to 5.0-rc12a from 4.7. My first experience with 5.0... 

 

The upgrade process worked just fine. However, I've run into my first problem:

 

I first started pre-clearing 4 drives via Screen/preclear script. Then I started "Utils/New Permissions", because accessing the files via Samba failed with some sort of access denied message. I could see "New Permissions" working on Disk1, then Disk2. Then I tried browsing to the Tower main page with a different browser tab to see how many read/write accesses the "New Permissions" performed. At the same time I tried to access the User shares via Samba to check whether I could browse the files again, after "New Permissions" had already finished Disk1. This is when everything Web and Samba got totally stuck. Now my Tower doesn't react to any HTTP or Samba requests, anymore. UnMenu is dead, too. The "New Permissions" window is stuck on "sync" for Disk2. However, ping and telnet still work, and the pre-clearing of those 4 new harddisks is still continuing to run just fine.

 

Anything I can do to help diagnosing the cause of this freeze? Anything I can do to get Web/Samba access back without interrupting the pre-clearing?

[Joe L.]

You can capture a copy of the syslog for analysis, but it sounds as if you just ran out of memory for all you were trying to do and processes got killed off by the kernel out-of-memory process.