Hello, recently when I move data to my server about mid way through the transfer it halts. I lose connection to my server shares. I can ping my server that is about it. I can still access console on my local UNRAID server, but can not shut it down or reboot it.

 

Pinging 10.200.1.140 with 32 bytes of data:

Reply from 10.200.1.210: Destination host unreachable.

 

That is a ping request

 

if I go to my share says network path unavailable. I can (from local console) do a 'reboot' command which thank god allows me to do a clean reboot. but that is about it. Once I reboot the server, everything is fine, but next large file transfer I do, just halts.

 

Any suggestions ?  I posted a syslog as well. Thanks

syslog-01-30-13.txt

Need to capture a syslog that contains an error event. Next the system stops responding copy the syslog to the flash drive and then post as an attachment.

Roger that, how do I pull the syslog from the local UNRAID console ? because that is the only thing i have access too when the problem occurs. I don't have telnet access. Let me know, thanks.

Roger that, how do I pull the syslog from the local UNRAID console ? because that is the only thing i have access too when the problem occurs. I don't have telnet access. Let me know, thanks.

 

Try these instructions:

   

    http://lime-technology.com/forum/index.php?topic=9880.msg94514#msg94514

 

Then shut your server down, remove the flash drive and  put it into your usual computer, and copy the file from the flash drive. 

OK i got the problem to happen again. Where I lose my connection in mid-file transfer.

 

I managed to copy approx 26gb to my drive (3) in question.... of a 47gb transfer. The transfer was writing to the server at approx 30-40MB/s Then it halted. So I pulled a system log from console and did a clean shut down. Pulled the log off USB. 

 

NOTE:  Before I got the issue to repeat I made sure to do a parity sync, which went without fault. A clean power down. Clean boot up. Before the file transfer I spun up all disks in the array.

 

As well any time this issue happens, I do shut it down from console. Once I boot back up it immediately/automatically runs a parity sync

 

I booted back up to copy media from the server, thinking it might have been only when writing to the server. I am in mid-copy from the server now, and only getting 5-7mb/sec ... so strange.

 

Thank you for the help

 

**thoughts** maybe it has to do with making a fresh share.cfg or a fresh super.dat file / SATA cables needing replacment? I could be wrong. This started happening after 2 disks in the set were replaced.

 

 

Biostar A7G0G / AMD Athlon x2 250 (3.0ghz) / 2GB DDR2 / 1x Promise Sata TX400 / Corsair HX620 / disk 0-9 (all WD) (EARS = precleared) / UnRAID v 4.7 Pro

syslog4.txt

Run checkdisk on the flash in a PC or Mac.

All was well, when I ran checkdisk on the flash, on my windows system...

Any other ideas ?

 

I attached another syslog since having the same issue.

 

C:\>chkdsk e: /f

The type of the file system is FAT.

Volume UNRAID created 18/02/2008 7:16 PM

Volume Serial Number is 0CA9-D46E

Windows is verifying files and folders...

File and folder verification is complete.

Windows has checked the file system and found no problems.

 

  513,499,136 bytes total disk space.

      16,384 bytes in 1 hidden files.

      16,384 bytes in 2 folders.

  58,621,952 bytes in 51 files.

  454,844,416 bytes available on disk.

 

        8,192 bytes in each allocation unit.

      62,683 total allocation units on disk.

      55,523 allocation units available on disk.

 

C:\>

 

--------

 

Every time I shut down (after getting my syslog) and boot up, it does another parity sync

syslog9.txt

Use the instructions in the Console Commands link in my sig to restart from the command line. Post SMART reports for the drives. Run Memtest overnight.

finally got home to get the SMART report on what I think is the drive I am having issues with (brand new EARS + Precleared successful)

 

Smart report attached, let me know what you think, thanks

smartdisk3.txt

hm problem is still persisting ... Not sure what to do to solve this... seems to have been doing this since I have added 2x precleared EARS disks ... I will grab smart for all disks but I can't determine what this issue is here

 

should i make a new USB ? should i upgrade my server version (i am running 4.7 pro) I may just run down to store and replace all SATA cables tomorrow to see if that helps... I checked inside my server case and made sure no power cables were close to any sata cables too just for the fun of it... trying to weed out possible problems :-/

 

I've been copying 6 to 10 gb at a time to disk3 and it seems to be doing fine, I realize it is also writing to parity as I copy, so it must be between parity and disc 3 (unless it is a cable issue) trying to determine from the smart reports

 

zip POSTED -- ALL 10 disks SMART REPORTS

SMARTREPORT.zip

Here is another syslog I grabbed today, this was after 2 successful 6-10 gb transfers , and currently copying another 14gb to server as I pulled the log

 

Thanks for all the help !

syslog_03_01_2013.txt

Is there a cache drive? How do you shutdown?

There is no cache drive... and when I lose connection via my windows PC, I run over to my server console, and type 'powerdown' and it turns off. Upon reboot I have access again until if it ever faults again.

If anyone has any idea to a solution to this problem would be greatly appreciated

The syslogs are very short and do not indicate a problem. Collect a syslog after the next problem event and before the system is restarted. Use the Console Commands instructions in my sig to shutdown from the command line.

Still having the issue where when file transfering I lose connection to my network share and access to my unraid server. Yet I still can use the server via console ... Any suggestions to why this is occuring ?

Need a syslog...

I take it no one has any solution to this issue I am having ? Just sorta deal with it ?

Are you getting the syslog after you lose the connection?

or are you getting the syslog in the middle of the big transfer?

 

We would need the one after you lose the connection, not in the middle of the transfer.

 

and i dont know if i missed it, but did you run memtest overnight to see your memory is okay?

 

 

OK Thank you, I will work on all of that, that you said. Right now I am writing 22GB to disk (disk9) seems to be OK. This issue started  to happen when I upgraded disk 3 from a 500GB (WD) to a 2TB (EARS) and since I use disk3 a lot, for my TV SHOWS, etc. It seems to crap out on me when writing to disk3.

 

Mind you I won't determine it is the disk itself... yet, but very well could be. Disk3 was precleared and installed, but when I write to it, seems to lose connection

 

Thanks. I will check the RAM again this week, could very well be that too.

I never ran 4.7 before. I jumped straight to 5.0 beta..

but i had the same problem before on 5.0 where unraid would lose the connection.

 

it kept losing connection when i try to copy >10GB files.

so when it lost the connection, i went on the machine console and typed these to renew the ip address

 

ifconfig eth0 down
ifconfig eth0 up

 

It crapped out on me and gave me some weird error i remember.

 

it was because of RAM.

for some reason, it gave me out of memory error in the syslog and it was shutting down the ethernet card.

and when i try to renew the ip address, it's not able to cause it completely shut down ethernet card.

 

so this is why we would need the syslog after losing the connection. as you expand your storage, it may require more ram. (i am not too sure on that though)..

 

I'm guessing you still have 2GB as you mentioned earlier?

 

 

anyway, i fixed my issue using this though:

 

http://lime-technology.com/forum/index.php?topic=3999.msg36211#msg36211

 

I used 65536 instead of 8192 like this:

 

echo 65536 > /proc/sys/vm/min_free_kbytes

 

if this fixes the problem, better to put it in go file

I will try that tonight, As well I'll toss some more RAM in the system, makes sense, toss good 4 to 8 gb in there. Didn't thing UnRAID needed much ram to run, DDR2 these days is pricey perhaps I'll take it to 4GB

OK just had the issue happen again.

 

here is the syslog from console before I 'poweroff' server

 

hope this helps.

 

As well I tried that

 

echo 65536 > /proc/sys/vm/min_free_kbytes

 

command, it seems to set the value, but if you powerdown / reboot, you have to re-renter the value, is that correct or should it save that value ?

 

Thx

syslog.zip

Had the issue happen again, I captured another syslog after I lost connection to my server and the file transfer failed after approx 10-12gb copied.

 

 

note: I had run this command -- echo 65536 > /proc/sys/vm/min_free_kbytes

 

Will be running memtest over night tonight

syslog2.zip