Koperfild

I've just finished some tests with NFS it works PERFECTLY! Thank you Tom. Now i can play my 1:1 blu-ray images with Tvix without any stuttering. The largest image i played was a 44GB movie with continous bitrate over 40Mbps (peaks on 46Mbps). It played perfectly. Thank you once again!

IMO that board is the best buy for the future. http://www.intel.com/products/desktop/motherboards/DQ45CB/DQ45CB-overview.htm

Thanks for your suggestions, now it seems like the best option for me is to buy new battery pack for the old UPS and use it to power my PC only, and buy an APC UPS to power my server. I can buy second-hand APC Smart-UPS 750VA USB & Serial 230V for 500z? = 166$ the salesman says the UPS has got new battery installed and calibrated. Is it worth?

Batteries are probably also dead, but it shouldn't affect (should it?) the output voltage since the device is equipped with AVR system. By the way how do AVR systems work in APC devices? Is the voltage almost perfect all the time? I've found an ETA brand UPS which looks very suitable for me. It's "multisystem UPS 1000L" What makes it very advantageous is lack of (needless) software bundled and good parameters. Also, the great advantage is that ETA UPSes can us APC's signals on RS232 which makes them compatible with apcupsd. What do you think of that?

My new unraid box is now finished! I've built it around Sharkoon Rebel 9 Economy & Welland 5.25" hot-spare bays. The case is all covered inside with noise absorbing car mats. The hardware hasn't changed, it's still: Asus P5E-VM Do Celeron 420 1gig of ram BeQuiet SP E5 450W Here are the photos The case: Fixing the matts: The bays: Also, i used a special tape and stands which reduce vibrations: Cable Management: New member of the server: Which is dead after few days. NEVER buy just relased disks! Placing the disks inside: And today's mod. I decided to fix another 92mm fan to make the air circulation better. The one place i could fix it in was the place where you mount peripheral devices. For now it looks awfully because i insulated the fan with white acrylic paste. And... The old server

I'm very disappointed from that hdd! I bought one about one month ago and after two weeks i got read/write errors (it worked as the parity disk) caused by badsectors (9 reported by mhdd), so i went to the shop and was given a new one which had the same manufacturing date and was coming from the same series (-300 SD17). I took it home and updated the firmware to SD1A as described in this thread. Then i ran and MHDD test just to make sure i can put it into my unraid box and have no problems in near future. Unfortunately MHDD showed 3 blocks with response time between 150 and 500ms and 233 blocks with response time between 50 and 150ms which are very likely to change into badsectors soon. That's very disappointing. I guess that a faulty set of them must have been manufactured, because it's not possible that two similar disks suffer such an issue from the very beginning. Has any of you experienced that problem? Here are photos of the old disk (left) and the new (right):

Hi, My old Mustek USB 1000Va UPS is (eventually) dead. Now i'm looking for a new one. The UPS suitable for me must have good voltage-stabilization system so that it outputs high-quality electric current when on-line and when on battery. It should keep my system (unRaid server box + PC + Network Gear) running for about 6-8 minutes so it must be around 1000Va (know that from my experience). It should also have at least RS-232 port to turn the computers down when needed. Which ones do you recommend? Thanks in advance.

Sounds cool! Looking forward to see it :P

I strongly agree with you, and that's what i really miss here...

Don't understand my post bad, i know Tom has got his privat life, family, etc. but 4.3 version was developed within 3 months, and new betas were getting relased regularly. Now it's been two months since last beta-2 and i guees the road to final is still faraway.

I haven't seen any unraid development for a long time and neither have i any Tom activity here. What's the reason? Does Tom work on a new beta? The last version comes from 1st October what's quite disappointing. I'm afraid that the project has been given up, what would be a disaster. New stable kernel (2.6.27) is out and i can't wait for NFS shares as promised.

I also suffer the problem under regular unRaid 4.3.3. And nothing seems to help.

I'm afraid we'll have to wait very long time. unRaid development seems to be extremely slow last time

Look here: http://www.jonnyguru.com/forums/showthread.php?t=3990

Unlike Corsair, both Seasonic and OCZ have split 12v rail, and that was the problem.

True, by the way, does anybody know if there is a possibility to disable that feature in barracuda .11 drives and enable the standard role (activity info) for the 11th pin in the sata mole? I've found two documents about that but don't understand much: http://ftp.t10.org/ftp/t13/technical/d97150r0.pdf http://www.patentstorm.us/patents/7221531/description.html

Well, i've done some more researches about be quiet psu's and it looks like i lied to you about their "brigded rails". In fact only two of their four 12v rails supply the power for sata devices. Now, recommended psu's are: - Antec TruePower Trio 650 - up to 18 drives, dual core cpu, three pci sata controller - PCP&C Silencer 610W - up to 18 drives, dual core cpu, two pci sata controller - Corsair HX620W - up to 18 drives, dual core cpu, two pci sata controller - PCP&C Silencer 500W - up to 13 drives, dual core cpu, two pci sata controller - Corsair HX520W - up to 13 drives, dual core cpu, two pci sata controller - Corsair VX450W - up to 10 drives, dual core, one pci sata controller Don't worry about Corsair's and Antec's rail separation, it's just a marketing hoax.

Much better than corsairs *50w (except 450W) are ones ending with *20w as they are manufactured by seasonic thus to which they're quieter and more efficient. I'd recommend you Corsair 520W [CMPSU-520HXEU]. A great alternative are also BeQuiet's P7 Dark Power Pro PSU's. Nowadays these are the best PSU's concerning their noise and efficiency. If i were you and could afford one, i'd definietly go with it. Actually, they don't have 'true' single 12v line, but their four rails are bridged into one (they have four converters which consist of one rail). Second important thing is how much power you really need. In fact the highest efficiency is reached when psu is loaded between 40 and 50%. What's more typical unraid machines don't draft as much power as people think. Let's assume an examplary server: - 16 X 1Tb non-wd green disk's - celeron e1400 - 1 gig of ram - generic motherboard /w gpu integrated - two external silicon image sata controllers For that server the a suitable psu is a good 550W. The biggest energy draft takes place when disks are being spun up. In that moment it may reach up to 27.5W per disk (25W on 12v rail and 2.5W on 5v rail) so, it's easile countable that for sixteen disks the peak power draft (on 12v) may be as big as 400W so, there are still 100W free for rest of the hardware. After spinning up the power need drops to what manufacturer's declare and, the typical value is 0.75A * 12V = 9W so, for sixteen disks it's 144W + 75 watts for hardware = 219W. 219W / 550W * 100% = 39,8% - the psu efficiency is high when loaded like this.

I'm open to suggestions to perform some additional tests to help you working the problem out.

Bubba when are you going to relase new beta which will recognise my drives (so thet they are not all read)? I am really looking forward to see it...

So the apps included in bubbaraid will just not work with user shares but the unraid engine will still offer user shares feature right? It's ok for me this way. so when can we expect beta 0.0.8 with corrected disk recognition?

Wait, user shares are when unraid merges two folders with the same name located on different disks into one? If so, deleting this feature you will kill the bubba raid project for me! When i saw that distro i really thought i would eventually be able to use my unraid box as torrent downloading machine, web server, and i generally was happy to see something new, as tom's progress has been hardly any last times... Now, without user shares it is going to be another unusefull project

Yes, i do.

I don't think so. Look at this: commit bc5b8bb64a2dc740d8b99635931e689a8b13daf2 Author: Greg Kroah-Hartman <[email protected]> Date: Wed Oct 15 16:02:53 2008 -0700 Linux 2.6.27.1 commit d23d43386311fde5f11e06c16d4185e94a8d6d06 Author: Steven Rostedt <[email protected]> Date: Wed Oct 15 18:21:44 2008 -0400 disable CONFIG_DYNAMIC_FTRACE due to possible memory corruption on module unload While debugging the e1000e corruption bug with Intel, we discovered today that the dynamic ftrace code in mainline is the likely source of this bug. For the stable kernel we are providing the only viable fix patch: labeling CONFIG_DYNAMIC_FTRACE as broken. (see the patch below) We will follow up with a backport patch that contains the fixes. But since the fixes are not a one liner, the safest approach for now is to disable the code in question. The cause of the bug is due to the way the current code in mainline handles dynamic ftrace. When dynamic ftrace is turned on, it also turns on CONFIG_FTRACE which enables the -pg config in gcc that places a call to mcount at every function call. With just CONFIG_FTRACE this causes a noticeable overhead. CONFIG_DYNAMIC_FTRACE works to ease this overhead by dynamically updating the mcount call sites into nops. The problem arises when we trace functions and modules are unloaded. The first time a function is called, it will call mcount and the mcount call will call ftrace_record_ip. This records the calling site and stores it in a preallocated hash table. Later on a daemon will wake up and call kstop_machine and convert any mcount callers into nops. The evolution of this code first tried to do this without the kstop_machine and used cmpxchg to update the callers as they were called. But I was informed that this is dangerous to do on SMP machines if another CPU is running that same code. The solution was to do this with kstop_machine. We still used cmpxchg to test if the code that we are modifying is indeed code that we expect to be before updating it - as a final line of defense. But on 32bit machines, ioremapped memory and modules share the same address space. When a module would load its code into memory and execute some code, that would register the function. On module unload, ftrace incorrectly did not zap these functions from its hash (this was the bug). The cmpxchg could have saved us in most cases (via luck) - but with ioremap-ed memory that was exactly the wrong thing to do - the results of cmpxchg on device memory are undefined. (and will likely result in a write) The pending .28 ftrace tree does not have this bug anymore, as a general push towards more robustness of code patching, this is done differently: we do not use cmpxchg and we do a WARN_ON and turn the tracer off if anything deviates from its expected state. Furthermore, patch sites are statically identified during build time so there's no runtime discovery of dynamic code areas anymore, and no room for code unmaps to cause the hash to become out of date. We believe the fragility of dynamic patching has been sufficiently addressed in the development code via the static patching method, but further suggestions to make it more robust are welcome.

Any news here? I see there's already 2.6.27.3 kernel but still no e1000 bug fixes...