unRAID Server release 4.4-beta2 available

[limetech]

Both wget and curl are available as standard slackware packages:

http://packages.slackware.it/

 

Is is sufficient to install these 'on the fly' via 'go' script?

[NAS]

not really as its a bit of a catch 22.

 

you really want to use wget or curl to download the packages that we want to help users install on the fly.

 

so we can handle the rest but knowing all users have one of these tools makes all subsequent sripts easier to automate

[WeeboTech]

The thought was to have a basic method of putting/getting files from unRAID without having to install the packages via go.

At the very least curl lets us get or push files in/out of the machine in a pinch,

i.e. if you want to save a log somewhere but the flash isn't working.

(which has happened to me a number of times).

 

Another reason for the request is newer automated tools that people are building.

If curl is a basic part of unRAID, then we can code tools to grab files and/or associated packages via script, menus & programs.

If curl is not on the system, then we have to walk people through the first step of installing curl.

Other options are future possibilities of automated log postings ,etc, etc.

[bubbaQ]

With wget already installed, scripts can be used to add packages and automate a lot of activities, and do it in a simplified manner from the perspective of novice users.

 

 

[NAS]

Is PAE large memory support coming in the 4.4 branch?

[limetech]

Is PAE large memory support coming in the 4.4 branch?

Sure, we can enable that, but as untested feature.

 

We're soon about to release next beta, which is actually quite a bit changed from previous.  Some highlights:

- 2.6.27 kernel

- samba 3.2.3

- individual disk controls, e.g., individual disk spin-down settings

- numerous small changes

 

[NAS]

Sounds like an exciting release is coming.

 

I would be happy to buy more RAM to test with and I believe webotech already has 8GB available. It definitely wont be for everyone so if its problematic then perhaps lets not do it but cost vs. gain for me at least in the context of real world usability RAM has been a godsend.

[jimwhite]

I have 8gb as well....

 

Bring on PAE

 

64bit next 

 

[jesuslovesbacon]

fantastic! bring on the next release!

[sosdk]

Just added a new disk(disk3) to my array.

 

Was copying a lot of files from disk1 to disk3 but it was taking forever so I decided to remove parity disk while moving files. But when I clicked stop disk2 suddenly went to unformatted.

 

What shall I do ?

[Joe L.]

Just added a new disk(disk3) to my array.

 

Was copying a lot of files from disk1 to disk3 but it was taking forever so I decided to remove parity disk while moving files. But when I clicked stop disk2 suddenly went to unformatted.

 

What shall I do ?

Whatever you do, do not reformat...

 

Instead, reboot.  It will figure it out.

 

Because the disk was "busy" it could not be un-mounted when you clicked stop.  Since it could not be un-mounted, it could not be mounted again.  unRAID figures ANYTHING it cannot mount as a reiserfs, regardless of the reason, is unformatted. (it is wrong, but it sure is disturbing, isn't it)

 

By rebooting, you will get it to figure things out.

 

Next time, stop the copy and log off the server before attempting to stop the array, as even being "cd'd" in a directory on a disk will keep it from being un-mounted.

 

Joe L.

 

 

[sosdk]

The html page doesnt show a reboot buttom.

 

How do I reboot from a telnet session ?

[sosdk]

Found it and disk2 is ok again - thx

[rhinoman]

I've found a minor bug I think.

 

I'm currently copying some data from a 1.5tb seagate to a 750gb seagate in the same machine running 4.4b2. Its been running for a couple of hours now, the 750gb shows free space diminishing as data is being transfered but the 1.5tb is howing as spun down and not having its space updated in the console. Its definilty running as the folders are slowly disapearing from the 1.5tb and reapearing on the 750gb.

[NAS]

Tom how we looking for the next beta or stable. Coming on 2 months since the 4.4 beta series came out and i believe there have been no major issues.

[RobJ]

I have to report an issue with v4.4-beta2 (probably beta1 too), with detecting drives that have spun up.  It appears that either my own ls loop to force certain folders to stay cached, or the kernels own logic to keep folder entries cached, is resulting (after gigabyte copies to the array) in uninvolved drives spinning up, and the unRAID Web Management page shows them still spun down.  UnMENU (with or without MyMain) shows them spun up, and my Kill-A-Watt confirms.  Very conveniently in MyMain, I can click the Spin down for the individual drives, and watch the Kill-A-Watt lose 6 to 7 watts.

 

Perhaps unRAID needs to detect drive activity at a lower level, or hook more drive activities, or poll the drives occasionally (hdparm -C /dev/sda?).

[Joe L.]

I have to report an issue with v4.4-beta2 (probably beta1 too), with detecting drives that have spun up.

Perhaps unRAID needs to detect drive activity at a lower level, or hook more drive activities, or poll the drives occasionally (hdparm -C /dev/sda?).

I've noticed the same thing....  emhttp  probably needs to poll using hdparm -C or equivalent every once in a while, to get its display back into sync with the drives that have spun up as a result of smartctl or other ioctl activities unrelated to reading or writing of disks.

 

Joe L.

[bubbaQ]

BTW, just something to file away for reference...

 

I partitioned my cache drive to be cache, rootFS, and swap.

 

Got it booting and Slackware installed, created ReiserFS on partition1.  All is fine, and unRAID sees the correct size partition for cache.

 

But since root is on the same drive, all IO to the rootfs partition shows up in the unRAID management interface.

 

Cache will never spin down if the rootFS is on another partition on the same drive.  Of course, that's OK for me, since I was going to set cache to never spin down anyway.