This post no longer serves a purpose. 

 

 

 

This might finally be it - barring any major issues reported over the weekend, this will become 4.5-final on Monday 7-Dec.

 

 

So it is not planned to fix the pending issues for AD-integration for 4.5 final?

 

I am still not able to join the domain - also with latest beta13.

 

DOMAIN_TEST is the netbios name, testdomain.de is the dns name. Windowsworkstations can join the domain without problems, so infrastructure is ok.

 

I tried all combinations of dns name and netbios name (especially because syslog was complaining about it) but without success.

 

Dec 8 00:20:26 XMS-GMI-01 emhttp: shcmd (267): /usr/bin/net ads join -U "Administrator"%"*****" 2>&1 | logger

Dec 8 00:20:27 XMS-GMI-01 logger: Failed to join domain: Invalid configuration ("realm" set to 'DOMAIN_TEST', should be 'testdomain.de') and configuration modification was not requested

Dec 8 00:20:28 XMS-GMI-01 emhttp: shcmd (268): /usr/bin/net ads testjoin -P >/dev/null 2>&1

Dec 8 00:20:28 XMS-GMI-01 emhttp: _shcmd: shcmd (268): exit status: -1

 

 

This might finally be it - barring any major issues reported over the weekend, this will become 4.5-final on Monday 7-Dec.

 

 

So it is not planned to fix the pending issues for AD-integration for 4.5 final?

 

I am still not able to join the domain - also with latest beta13.

 

DOMAIN_TEST is the netbios name, testdomain.de is the dns name. Windowsworkstations can join the domain without problems, so infrastructure is ok.

 

I tried all combinations of dns name and netbios name (especially because syslog was complaining about it) but without success.

 

Dec 8 00:20:26 XMS-GMI-01 emhttp: shcmd (267): /usr/bin/net ads join -U "Administrator"%"*****" 2>&1 | logger

Dec 8 00:20:27 XMS-GMI-01 logger: Failed to join domain: Invalid configuration ("realm" set to 'DOMAIN_TEST', should be 'testdomain.de') and configuration modification was not requested

Dec 8 00:20:28 XMS-GMI-01 emhttp: shcmd (268): /usr/bin/net ads testjoin -P >/dev/null 2>&1

Dec 8 00:20:28 XMS-GMI-01 emhttp: _shcmd: shcmd (268): exit status: -1

 

 

No, bugs will continue to be fixed.  Fixes after 4.5 "final" will have an additional suffix - the first such release after 4.5 would be 4.5.1.

 

For this particular issue, you need to enter the fully-qualified domain, eg., try [email protected].

No, bugs will continue to be fixed.  Fixes after 4.5 "final" will have an additional suffix

- the first such release after 4.5 would be 4.5.1.

 

We already went from 4.5.1 through 4.5.13 in the betas. 

So you could maybe call the final 4.6 "final". 

Thus the next bug-fix release would be 4.6.1.

 

 

No, bugs will continue to be fixed.  Fixes after 4.5 "final" will have an additional suffix

- the first such release after 4.5 would be 4.5.1.

 

We already went from 4.5.1 through 4.5.13 in the betas. 

So you could maybe call the final 4.6 "final". 

Thus the next bug-fix release would be 4.6.1.

 

 

 

Betas are different then point releases. The final release will be 4.5 no matter how many betas there are until then.  Some devs do there numbering differently then others and it is mostly personal preference but Lime-techs numbering schema makes sense to me.

Still slow for me when uploading. On every build for a while I still only get 14 MB/s when uploading. I can switch back to 4.4.2 and get 40 MB/s. Any ideas on what I can do to resolve this on current beta's. Here's my Syslog http://pastebin.com/f1bf239c7 after an upgrade to beta 13 and copy of 4.5 GB worth of data.

 

I'd like to see the full system log.  Pleaes email to me [email protected]

 

One of your hard drives does not appear to support SMART or at least doesn't support reading temperatures... is this correct?

Just wanted to report back that I swapped out my old IDE Cache drive for a newer SATA drive with great results. 4.5GB file copied over in 01:37 at an average speed of 46 MB/s.     

DOMAIN_TEST is the netbios name, testdomain.de is the dns name. Windowsworkstations can join the domain without problems, so infrastructure is ok.

 

 

For this particular issue, you need to enter the fully-qualified domain, eg., try [email protected].

 

Trying this gives me

Dec 8 06:16:23 XMS-GMI-01 logger: Failed to join domain: failed to find DC for domain [email protected]

 

And even going back to DOMAIN_TEST gives same result now. Are you sure, the webinterface is setting things right? Shouldn't I get former results when switching to former config?

 

Edit:

Did further checks: PING TESTDOMAIN.DE resolves to domaincontroller as it should. ping DOMAIN_TEST points to my mailserver (!) - but only on unraid box. It does not resolve at all on other machines (while it is properly registered in WINS to the DC) - so is there a way to tell unraid e.g. the DC to use or to prevent resolving the domain name to a wrong ip address?

Beta-11 to Beta-13 upgrade has left me all blue

 

Was getting an access denied writing to my flash share even though it was set read/write in the web interface so.

 

Pulled out the flash drive and put it in my PC, copied over bzimage, bzroot and memtest.

Put it back in the server

Stopped the aray

Rebooted

Comes back up but all my drives are blue and array is stopped

 

I've attached my syslog, what would be the best course of action?

Beta-11 to Beta-13 upgrade has left me all blue

 

Was getting an access denied writing to my flash share even though it was set read/write in the web interface so.

 

Pulled out the flash drive and put it in my PC, copied over bzimage, bzroot and memtest.

Put it back in the server

Stopped the aray

Rebooted

Comes back up but all my drives are blue and array is stopped

 

I've attached my syslog, what would be the best course of action?

This line in your syslog indicates your config/super.dat file is un-readable.

Dec  8 19:46:15 Silo kernel: md: could not read superblock from /boot/config/super.dat

 

That would usually indicate it is corrupted in some way.  It might be just that one file, but to be sure you should move the flash drive to the PC once more and run chkdisk on it, just in case it has some corruption. (Make sure you "safely eject" the flash drive before moving it back to the unRAID array)

 

Then, since your disk.cfg file knows which disks are in the array all you need to do is use the "Trust-my-parity" procedure as described in the wiki to rebuild a new super.dat file.

 

The procedure is described here: http://lime-technology.com/wiki/index.php?title=Make_unRAID_Trust_the_Parity_Drive,_Avoid_Rebuilding_Parity_Unnecessarily

 

Joe L.

Just wanting to note that kernel 2.6.31.7 was released today. I haven't looked through the entire changes yet so not certain if/how much it would help others out.

Can cpufreq module be added to the final release please?

I am sure many folks who don't use WOL  and don't put their unRAID servers asleep would be happy to save on electricity bill by utilizing the CPU frequency scaling feature.

 

http://lime-technology.com/forum/index.php?topic=1184.0

http://lime-technology.com/forum/index.php?topic=4825.0

 

 

+1 

 

Should be so great 

Can cpufreq module be added to the final release please?

 

+1 

 

 

 

Just wanting to note that kernel 2.6.31.7 was released today.

Also, it looks like 2.6.32 is stable now.

Just wanting to note that kernel 2.6.31.7 was released today. I haven't looked through the entire changes yet so not certain if/how much it would help others out.

 

Noticed that as well - it doesn't look too interesting though apparently there's a small tweak in the sky2 ethernet driver.  Since 4.5 final is released, I'll probably create a 4.5.1 at some point with this kernel.  Perhaps consider moving to 2.6.32 as well...

Just wanting to note that kernel 2.6.31.7 was released today.

Also, it looks like 2.6.32 is stable now.

 

I'll probably wait for 2.6.32.1 

Can cpufreq module be added to the final release please?

I am sure many folks who don't use WOL  and don't put their unRAID servers asleep would be happy to save on electricity bill by utilizing the CPU frequency scaling feature.

 

http://lime-technology.com/forum/index.php?topic=1184.0

http://lime-technology.com/forum/index.php?topic=4825.0

 

One thing I don't know whether it's related or not, there is a bios option on the Supermicro motherboards called "Enable Intel CTECH" (something like that), which if enabled, has caused spontaneous resets.  So I need to look into this to see how it might be related to CPU frequency scaling.  It's on my to-do list

You mean C1E or EIST (speedstep)?

 

C1E being a halt state to reduce power consumption when idle and speedstep being a way of reducing CPU multiplier and vcore (not seen this in action though) to reduce power consumption.

 

 

Worked a treat thanks! 

 

Beta-11 to Beta-13 upgrade has left me all blue

 

Was getting an access denied writing to my flash share even though it was set read/write in the web interface so.

 

Pulled out the flash drive and put it in my PC, copied over bzimage, bzroot and memtest.

Put it back in the server

Stopped the aray

Rebooted

Comes back up but all my drives are blue and array is stopped

 

I've attached my syslog, what would be the best course of action?

This line in your syslog indicates your config/super.dat file is un-readable.

Dec  8 19:46:15 Silo kernel: md: could not read superblock from /boot/config/super.dat

 

That would usually indicate it is corrupted in some way.   It might be just that one file, but to be sure you should move the flash drive to the PC once more and run chkdisk on it, just in case it has some corruption. (Make sure you "safely eject" the flash drive before moving it back to the unRAID array)

 

Then, since your disk.cfg file knows which disks are in the array all you need to do is use the "Trust-my-parity" procedure as described in the wiki to rebuild a new super.dat file.

 

The procedure is described here: http://lime-technology.com/wiki/index.php?title=Make_unRAID_Trust_the_Parity_Drive,_Avoid_Rebuilding_Parity_Unnecessarily

 

Joe L.

You mean C1E or EIST (speedstep)?

 

C1E being a halt state to reduce power consumption when idle and speedstep being a way of reducing CPU multiplier and vcore (not seen this in action though) to reduce power consumption.

 

 

 

I mean "Intel C-State Technology".  Apparently started with E7300 processor, but E7400 plugged into Supermicro C2SEE with this enabled caused resets.

You mean C1E or EIST (speedstep)?

 

C1E being a halt state to reduce power consumption when idle and speedstep being a way of reducing CPU multiplier and vcore (not seen this in action though) to reduce power consumption.

 

I mean "Intel C-State Technology".  Apparently started with E7300 processor, but E7400 plugged into Supermicro C2SEE with this enabled caused resets.

 

Intel C-State technology is a broad sweeping statement for many of their power saving techniques and it seems each CPU has a new version. C1E is an old halt state which most AMD cpus can do.

 

Overclocking/underclocking and enabling C1E, C-State, Speedstep and EIST is a no-no on lots of m/b not just supermicro boards. Should use one of the other no both.