April 27, 201313 yr Hi all, The ip address of my tower is 10.168.3.80. Actually, I can only browse to http://10.168.3.80:8080, but can't loading to http://10.168.3.80/, http://tower/, http://tower:8080/. If using Windows Explorer to access share drives, I can only use \\10.168.3.80 to access them (e.g. \\10.168.3.80\Public, \\10.168.3.80\flash). But there will not be available intermittently. Which log i need to post here for further diagnostic? :'( :'( :'( I have tried to remove all PLG, PKG but still no luck.
April 28, 201313 yr Author Thanks dgaschk, Revert to stock system: 1. Rename the /boot(flash)/config/plugins directory. 2. Rename boot(flash)/extra/. 3. Use the stock go file (boot(flash)/config/go). i did step 1 & 2 before and missing step 3. Just follow your step 3 to revert to default, http:\\tower now back to normal, but i can't connect to http:\\tower:8080 or http:\\10.168.3.80:8080. Is it normal for a default setting?
April 28, 201313 yr As mentioned before 8080 is unMenu. The default go file does not include unMenu so this is normal.
May 18, 201313 yr Author I still experience this issue intermittently. Ping: ok http://tower: 10 time load not 10 time success (using ip as well) Telnet: 10 time load not 10 time success (using ip as well) Anything I can do with diagnostic? I really want to found out the root cause and fix it.
May 18, 201313 yr It sounds like the Web GUI (emhttp) is not running. See this thread for how to restart it without rebooting: http://lime-technology.com/forum/index.php?topic=27085.0
May 18, 201313 yr Author Thanks Gary, First when I try to telnet my server, it shown me "telnet: can't connect to remote host (10.168.3.80): Connection refused". Then I try again, it works, and follow the instruction to restart emhttp, then the man menu back to normal. Thx!
May 18, 201313 yr You're most welcome. Note that with v5 you have to be careful to kill all of the processes (as outlined on the link I gave you) or you can get a segfault when you restart emhttp. With v4.7 it's a bit simpler. But in either case it's a PITA !!
May 18, 201313 yr Author PITA! So true! I wonder what I can do further with this issue? Add new NIC? Change motherboard?
May 18, 201313 yr PITA! So true! I wonder what I can do further with this issue? Add new NIC? Change motherboard? Has the issue returned? If you have to restart emhttp frequently, then something is clearly wrong -- either with your configuration; or with your hardware. Have you run a THOROUGH memory test (e.g. at least 12 hours of MemTest ... preferably longer)?
May 19, 201313 yr Author Yes still. Problem occurs intermittently, e.g. connection dropped when playing movies from media player box, telnet session disconnected, webgui 5 times success out of 10 times trial. I did once MEMTEST when first built the unRAID. Let me schedule a MEMTEST.
May 19, 201313 yr Let MemTest run for at least 12 hours ... preferably longer. Intermittent memory errors can take a long time to detect.
May 19, 201313 yr Hopefully the display will look just like that in 12 hours (i.e. no red error lines !)
May 19, 201313 yr Your .sig indicates that you have 1333MHz memory, yet memtest is reporting 665MHz - this doesn't sound right. It may be that the version of memtest in use isn't fully compatible with ivy bridge cpus and is reporting an incorrect value. However, if it really is running at 665MHz, this isn't a valid test of your memory. I'm not clear whether you are running memtest86, or memtest86+ - the screen shot is confusing (the latter usually reports as 86+, the former should show a version of 4.2.0, with a slightly different screen layout ... my guess is that you are running 86+, but the '+' isn't showing). memtest86+ development appears to have stagnated in January 2011. memtest86 has had a couple of revisions in 2013, after two years stagnation. Unfortunately, the current releases of both versions are 4.20 (or 4.2.0).
May 20, 201313 yr Author Thx all! Yes I have 1333MHz memory. I just run the memtest from unRAID boot USB. What else I should do with memtest? How to get 1333MHz shown instead of 665MHz? I can't the is any + sign so I expect I'm running memtest 86. I just take another screen shot, hope it is more clear. Up to now, I can't see any error yet.
May 20, 201313 yr Author I just download Memtest86+ 5.00RC1 and start another run memtest. Stay tuned.
May 21, 201313 yr Author Hi all, after another 24hrs, test with Memtest86+ 5.00rc1, here the result. Is it a gd news? Or bad news?
May 21, 201313 yr Looks fine -- 25 passes with zero errors is a good indication that your memory's fine. That's "good" for the memory; but doesn't help isolate why you're having the intermittent issues. How many disks does your system have?
May 21, 201313 yr Memtest is still appears to be reporting the memory speed as 665MHz. If this is accurate, then there is still no validity in the test.
May 21, 201313 yr Memtest is still appears to be reporting the memory speed as 665MHz. If this is accurate, then there is still no validity in the test. It's DDR RAM "Dual data rate" ==> transfers on both rising and falling clock. MemTest is reporting it just fine ... a 665MHz clock = DDR3-1330. In theory it should show a 666MHz clock, but I wouldn't be concerned with a 1 cycle difference in the clock
May 21, 201313 yr Author Thx Gary & PeterB @Gary, I got 9 disks in the system, including Parity & Cache disk. @PeterB, i found something interesting. When i look at the sample screenshot from Memtest86, it showing 668MHz for his DDR-1337 ram. Could anyone show me your Memtest86 screen as a ref? Do u know how to set it to expected speed to run Memtest86? http://forum.canardpc.com/forums/73-Memtest86-Official-forum#threads/68001-NEW-%21%21-Memtest86-5.00-RC1-available-%21-Need-betatesters-%21?forcenoajax=1
May 21, 201313 yr Intermittent errors are USUALLY caused by either memory, power, or poorly seated or "noisy" cables. You've already tested memory pretty well. Unplug all of your SATA cables and re-plug them just to be sure they're firmly seated. Also, do NOT tie them together for "neatness" (a common mistake), as this can induce crosstalk between them and cause errors. If that doesn't help, then I'd replace the power supply with a higher quality unit. Your Antec is an "okay" low cost unit; but I'd get a good Seasonic X series or Corsair HX series unit instead.
Archived
This topic is now archived and is closed to further replies.