icon123

Thanks Joe. The server just locked up on me again and this time is wasn't doing a parity check, it was just copying files. Since it is locked, I can't get a syslog. I have another AMD processor, should I switch it out. It is still an x2 cpu though. Or maybe try to disable one of the cores in the bios? This build had been runing for about 6+ months and I haven't experienced anything like this before. I'm a little concerned that I can't even copy files to the server without locking it up. Any suggestions would be appreciated. I will email Tom for the time being. Edit - Hold up for now - I have bad memory.

Any suggestions based on what I posted above? The parity check completed fine, however, I didn't touch the server why it was running. I have attached the complete syslog for any reference or comparison. Thanks. syslog-2011-08-30.zip

Perfect, thanks!

Sorry... Case: Coolermaster Centurion 590 Drive Cages: Supermicro 5 in 3 hot swappable drive bays x2 P.S.: CORSAIR CMPSU-450VX Motherboard: Biostar A760G M2+ CPU: AMD Athlon X2 BE-2400 2.3 GHz Memory: 4GB: qty 2 - A-DATA 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400) Dual Channel Kit Desktop Memory Model AD2U800B1G5-DRH HDD Controller: Supermicro AOC-SASLP-MV8 (installed with 1 drive) HDD (all WD,): Parity: 2TB EARS Disk 1: 2TB EARS Disk 2: 2TB EARS Disk 3: 2TB EARS Disk 4: 2TB EARS (all EARS – 4k aligned) Flash: SanDisk Cruzer 2GB unRAID plus: 4.7 Samba shares only Full syslog is attached. I have had my fair share of problems with this build since I started. When running normally, it seems to work fine. But when I run a parity check it ends up running out of memory and kills some of my processes. I could never use the server during a parity check, cause it would kill samba almost imediately. I actally thought this was normal until I saw others using there server while a parity check was going on. I went from 2 to 4 gigs and that didn't help. So this time, I stopped crashplan and cache_dir from running at boot and I wasn't getting any out of memory errors. So I tried to use the server as normal, copying things to it. When I was doing this earlier yesterday, it appeared that the server froze after copying for awhile (I was doing this via RDP). I lost the web interface, unraid interface and couldn't even telenet into it. When I got home I hard rebooted causing anoter parity check. This time I would do a little bit at a time during the parity check, while keeping an eye on my syslog entries. Then I got a connecting error on my client while copying some files, that is the kernel panic I posted earlier. Although smb seemed to work ok, I couldn't copy anything else from that client to the server. Tried to copy from another client and that worked ok for a few minutes then I received another disconnect. This time, no other entries in the syslog. At some point unmenu stopped working even though I didn't see an entry for this. Maybe the syslog stoped getting written too. At this point it was about 45% done with the parity check. I thought I would just leave it and go to bed. Woke up this morning with it at 57% and all drives a sleep and it telling me it would take like 475678 minutes or something like that. I tried to stop the array since the web interface was still working. That locked the web interface up. Tried to issue a reboot command from telenet and that did nothing forcing me to hard boot it again. Currently it's at 45%+ parity check moving along just fine. I won't touch it this time (unless instructed to do so). I have run a memory test a few months ago with no problems. Sorry for the long post. Edit - hardware list posted syslog-2011-08-29a.zip

Thanks, I realized I could do that and it was a stupid question how I worded it. I should have asked, how could I have another script, like my go script, that I could execute after boot? That way I wouldn't have to type ever line item individually.

I am currently running a parity check due to an improper shutdown. Everything seemed to be going ok. I was coping some large files to the server and this is what showed up in the syslog and the client hanged on the copying. What does this mean? Aug 29 22:23:32 Hitch kernel: ------------[ cut here ]------------ Aug 29 22:23:32 Hitch kernel: kernel BUG at drivers/md/unraid.c:436! Aug 29 22:23:32 Hitch kernel: invalid opcode: 0000 [#1] SMP Aug 29 22:23:32 Hitch kernel: last sysfs file: /sys/devices/pci0000:00/0000:00:11.0/host6/target6:0:0/6:0:0:0/block/sdf/stat Aug 29 22:23:32 Hitch kernel: Modules linked in: md_mod xor atiixp ahci r8169 mvsas libsas scst scsi_transport_sas Aug 29 22:23:32 Hitch kernel: Aug 29 22:23:32 Hitch kernel: Pid: 3095, comm: flush-9:2 Not tainted (2.6.32.9-unRAID # A760G M2+ Aug 29 22:23:32 Hitch kernel: EIP: 0060:[<f851bc2c>] EFLAGS: 00010286 CPU: 1 Aug 29 22:23:32 Hitch kernel: EIP is at unraid_make_request+0x1e1/0x342 [md_mod] Aug 29 22:23:32 Hitch kernel: EAX: c2929d30 EBX: 00000002 ECX: 00000005 EDX: c2929d30 Aug 29 22:23:32 Hitch kernel: ESI: 00000001 EDI: c2929c20 EBP: c21e1bdc ESP: c21e1b98 Aug 29 22:23:32 Hitch kernel: DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068 Aug 29 22:23:32 Hitch kernel: Process flush-9:2 (pid: 3095, ti=c21e0000 task=c2252260 task.ti=c21e0000) Aug 29 22:23:32 Hitch kernel: Stack: Aug 29 22:23:32 Hitch kernel: c3c76d80 00000002 f74a5ba0 00000001 86fe7100 00000000 86fe7108 00000000 Aug 29 22:23:32 Hitch kernel: <0> c2929d30 f74a5ba0 c41884e0 c2929d48 c21e1bfc c1126fb4 f6de52e0 00000008 Aug 29 22:23:32 Hitch kernel: <0> c430a000 c21e1bfc f8518ef7 c3c76d80 00000002 00000000 f6de52e0 00000008 Aug 29 22:23:32 Hitch kernel: Call Trace: Aug 29 22:23:32 Hitch kernel: [<c1126fb4>] ? __make_request+0x235/0x341 Aug 29 22:23:32 Hitch kernel: [<f8518ef7>] ? md_make_request+0x83/0x8b [md_mod] Aug 29 22:23:32 Hitch kernel: [<c1125b0f>] ? generic_make_request+0x1d6/0x211 Aug 29 22:23:32 Hitch kernel: [<c104a784>] ? mempool_alloc+0x3d/0xd3 Aug 29 22:23:32 Hitch kernel: [<c1126d10>] ? submit_bio+0x96/0x9b Aug 29 22:23:32 Hitch kernel: [<c1089412>] ? bio_alloc_bioset+0x37/0x96 Aug 29 22:23:32 Hitch kernel: [<c1085ee1>] ? submit_bh+0xde/0xfd Aug 29 22:23:32 Hitch kernel: [<c10af011>] ? reiserfs_writepage+0xa42/0xb85 Aug 29 22:23:32 Hitch kernel: [<f84788a5>] ? rtl8169_poll+0xf6/0x11a [r8169] Aug 29 22:23:32 Hitch kernel: [<c12312aa>] ? net_rx_action+0x57/0x102 Aug 29 22:23:32 Hitch kernel: [<c1133b65>] ? radix_tree_gang_lookup_tag_slot+0x7b/0x9a Aug 29 22:23:32 Hitch kernel: [<c104d843>] ? __writepage+0xb/0x23 Aug 29 22:23:32 Hitch kernel: [<c104daeb>] ? write_cache_pages+0x1b4/0x2ac Aug 29 22:23:32 Hitch kernel: [<c104d838>] ? __writepage+0x0/0x23 Aug 29 22:23:32 Hitch kernel: [<c104dc00>] ? generic_writepages+0x1d/0x27 Aug 29 22:23:32 Hitch kernel: [<c104dc2f>] ? do_writepages+0x25/0x28 Aug 29 22:23:32 Hitch kernel: [<c1081c82>] ? writeback_single_inode+0xae/0x1eb Aug 29 22:23:32 Hitch kernel: [<c108233d>] ? writeback_inodes_wb+0x2ca/0x343 Aug 29 22:23:32 Hitch kernel: [<c10824a6>] ? wb_writeback+0xf0/0x14d Aug 29 22:23:32 Hitch kernel: [<c108257b>] ? wb_clear_pending+0x65/0x6a Aug 29 22:23:32 Hitch kernel: [<c10825e2>] ? wb_do_writeback+0x62/0x11b Aug 29 22:23:32 Hitch kernel: [<c10826c3>] ? bdi_writeback_task+0x28/0x85 Aug 29 22:23:32 Hitch kernel: [<c1056226>] ? bdi_start_fn+0x0/0xa1 Aug 29 22:23:32 Hitch kernel: [<c105627b>] ? bdi_start_fn+0x55/0xa1 Aug 29 22:23:32 Hitch kernel: [<c1056226>] ? bdi_start_fn+0x0/0xa1 Aug 29 22:23:32 Hitch kernel: [<c1033869>] ? kthread+0x61/0x68 Aug 29 22:23:32 Hitch kernel: [<c1033808>] ? kthread+0x0/0x68 Aug 29 22:23:32 Hitch kernel: [<c100339f>] ? kernel_thread_helper+0x7/0x1a Aug 29 22:23:32 Hitch kernel: Code: ca 7c ea 83 7d dc 00 75 04 0f 0b eb fe 8d 47 28 e8 88 49 d8 c8 8b 45 dc 83 78 0c 00 74 04 0f 0b eb fe 8b 55 dc 83 7a 10 00 74 04 <0f> 0b eb fe 83 7d c8 00 74 06 83 7d c8 02 75 0e 8b 5d bc 8b 4d Aug 29 22:23:32 Hitch kernel: EIP: [<f851bc2c>] unraid_make_request+0x1e1/0x342 [md_mod] SS:ESP 0068:c21e1b98 Aug 29 22:23:32 Hitch kernel: ---[ end trace c0c51844f16b2369 ]---

Hi, This might seem like a real simple/stupid question, but I think I need to remove some items from my go script so they don't execute at boot. How do I manually run them. One item is crashplan, one is unmenu and another one is dir cache. (Yes, I am a linux noob). TIA

I have been reading about the custom software and was planing on putting that on. But xmount appears to be the answer to my smb woes, if it works like I read it. Thanks for chiming in everyone. So the consenses is that no one is running media server software on unraid and connecting there WD to it.

Thanks for the reply. And I do agree that it is easy to do smb, just isn't as "pretty" as the old setup. Understandingly, I could go with a different/more expensive setup than the WD, but I think it is good bang for the buck. My biggest problem with the WD smb side is that it displays all my available smb shares and on any given day, I could have up to (7) networked PCs plus other things showing up in the list. That's why I didn't know if the media server would help.

Hey all, Wanted to get your opinion on something. I'm moving away from my whole house SageTV system which did a great job of playing all my 1080p mkv files from unraid. I will be replacing it with a couple of WD Live boxes and want to know what ways could or have you guys connected them to unraid? I know I can use smb shares, it just seemed a little, well not so user friendly for the family. I didn't know if I would gain anything by running a mediaserver on unraid. If so, what one would you recommend? Thanks.

This is what I used: [Animated Movies] path = /mnt/user/movies/animated read only = Yes valid users = usr1, usr2 browseable = no This gives me a hidden read only link to my animated movies that resides in the movies share. Then I made those users invalid for all other shares so they can not access them. They can still see them, but that is ok in my case. I also made the new shares hidden, otherwise the valid users would see these new shares as well and IMO would clutter up the rest of my share layout. Thanks again for the help and suggestions. BTW - I was able to create a shortcut on their windows 7 dekstops that points to the hidden shares. Therefore, no mapping had to be done nor links to the 'network area'. Just a FYI, never new that featured existed. My luck it probably existed sine Win3.11.

Thanks for the replies. Unless someone tells me otherwise, I was going to setup a share like you mention and see how it exports it and crib off that. For the record, I don't care if they see all the shares, I just don't want them to be able to access all of them. Yes they each have there own laptops, but they are young (7 and 9). BTW - the "adult" directory is for rated R movies not xxx movies. If it was the later, it would be locked down differently for sure.

So I'm going to create a smb-extra.conf file with the following included: [Animated Movies] path = /mnt/usr/movies/animated read only = Yes If I understand this correctly, this should give me a new read only share called "Animated Movies" that will point to my animated movies. What I am trying to still figure out is: I only want this to be in affect for my kids. My thought was to exclude the kids users from all the shares I created within the web setup and only give them access to these new samba-extra shares I am creating. Sound like a good game plan? If so, what is the command to limit the above share to just specific users?

Is it possible to create a share within a share? I'm assuming not, so followup question... I have a share, movies. Within this share are home, adult (Rated R), animated, etc. I want to give my kids access to some of those directories but not all of them. Any easy way I can accomplish this? TIA

Will this work in general for preclearing drives? It's one of the biggest reasons I shut down the server. Would we put new drive to preclear in and mount it, then preclear it and then unmout it and remove it? Is it that simple (assuming AHCI mode of course)?

I haven't tired testing the new ram yet, but the old ram passed. The only other thing I run is crashplan, that's why you will probably notice java running out of memory as well. Thanks for the tips.

Maybe this will work... But be careful. Try it in a test location first. I have only used this on my Windows machines though. Let me know if it works. It's a .bat file that basically creates a directory of the filename and moves the file in that directory. file2folder.zip

I have noticed, when I seem to run a non-corrective parity check from the unraid menu, my system runs out of memory and starts killing things including emhttp. This was when I had 2GB of memory, so I upgraded to 4GB and it does the same thing. Can anyone help explaing why? Please see syslog-2011-05-03.zip for this. Is it normal that smb shares are lost when you are rebuilding drives? If so, then don' worry about syslog-2011-05-04.zip, otherwise, could someone explain why that happens. Everytime I rebuild, it seems to kill my shares. Thanks. syslog-2011-05-03.zip syslog-2011-05-04.zip

Thanks for your explanation, but I probably should have left WHS out of it as I am certainly not trying to reproduce it, but merley mentioning that was a feature that I enjoyed. As part of my back up plan, I have my important files on my main unraid server 'crash planed' to another offsite unraid server and that works well. I have my laptops 'crash planed' to both unraid servers and that also works well. I was mainly looking for share duplication for another method of redundacy, mainly just for photos and music. I have my crash plan backups on a 2TB drive that remains out of the share pool so duplicaing other shares to it really wouldn't be a problem in my case. BTW - is there any way to setup the WHS PC backups to store on the unraid server? Probably not, but that would be sweet.

Hey all, Comming for the WHS world, one of the nice things that it provided was setting duplication up for user shares. I would like to do something similar on unraid. Didn't know if someone already developed something. If not, what's the best way to accomplish this. My only thought was some sort of rsync ran from cron. Any thoughts?

Thanks for the reply, but in case it wasn't clear the way I stated it. I have now tried upgrading this one paticular drive with (4) different 2TB drives (on 2 different mother boards) that were previously precleared with no errors. For some reason, I am having a hard time upgrading this one paticular drive. I don't think its my 2TB drives that are the problem. Specially since I have used (2) of them to upgrade other drives in my array without any problems and relatively fast ~57MBps.

So to continue the saga ... I switched MBs to a Biostar A760G M2+. I still have the AOC-SASLP-MV8 installed but not currently using and I have had problems running prelcear on it. I have sucessfully upgraded (2) of my 1TB HDDs to 2TB. When I got to the third drive upgrade, it started rebuilding really slow (~1 to 2 MBps at the most) so I stoped it, rebooted and tried it again. This time a little faster (~8 to 10 MBps) but then failed later on. I put my 1TB drive back in for now and doing a parity rebuild/check. What I find interesting is that this is the same drive I tried to upgrade early on in this post when I had the Asus MB. So I'm thinking maybe a bad cable or hard drive chasis (supermicro 5 in 3). I wanted to get your thoughts on what recommendations on what my next move should be. I like to be able to troubleshoot this without putting my data at risk like I have been doing. I know I can write and read ok to the disk directly, however, this might be where some of my slower speeds have been coming from. Any thoughts? Thanks.

I was in the process of upgrading my 1TB HDDs to 2TB. (2) of these sucessfully upgraded without any problems. The third one however was going real slow (1 to 2 MBps) so I stoped in and tried it again. When I got home from work the drive upgrade process failed and the disk was disabled. Knowing that I might have more of a problem that I wanted to tackle right then, I decided to put the healthy, original, 1TB drive back in. Now when I boot up, the array is stopped saying that the replacement disk must be just as big or bigger than the original. What do I do now, type initconfig?

Thanks op, saved $50!

Exchange server can NOT be taken off line. I could bpass it by connecting my laptop directly to the unraid server, but that might not be very helpful. I rebooted again and have no problems accessing the main web interface now. Not sure why or when it will start not working again. Maybe a memory issue.