coppit Posted February 6, 2014 Share Posted February 6, 2014 I'm having trouble with VMs getting hung after a while. I'm running some video encoding, so that might have something to do with it. For example, my Windows VM is wedged right now: # xl list Name ID Mem VCPUs State Time(s) Domain-0 0 2048 8 r----- 457471.1 ubuntu.13-04.xfce.x86-64.20130424 52 512 1 -b---- 80.1 windows7 53 4093 4 ------ 59582.8 The state looks... Odd. I have no idea whether this is an unraid beta 3 problem or some problem with my VMs. Can someone point me in the right direction for debugging this? The dom0 /var/log/xen/*.log files don't seem to show much, and neither does /var/log/syslog. Do I need to be careful about over subscribing the cpus or memory? The linux VM has: vcpus = 1 memory = 512 and the windows one has: vcpus = 4 memory = 4096 I have 16MB of system memory and a Xeon E3-1245 V2. Quote Link to comment
pyrater Posted February 7, 2014 Share Posted February 7, 2014 Yet another crash ......... This time no cache_dir and was using swap. Ran stable for about 25 hours. Here is my sysconfig: label Xen/unRAID OS menu default kernel /syslinux/mboot.c32 append /xen xsave=1 cpufreq=xen:performance dom0_max_vcpus=9 dom0_vcpus_pin iommu=1 --- /bzimage console=tty0 xen-pciback.hide=(00:09.0) --- /bzroot Logs: Attached to post Configs: Attached to post Running: Unraid - Plex, APC UPS Daemon, Swap File (4gb) Windows 2 cpu and 1792 mb - utorrent, sb, cp, and vpn Arch - 1 cpu and 512 mb with mumble server Server specs: -CPU: AMD Six-Core 2419 EE 1.8GHz (x2) = 12 Cores -MOBO: H8DME-2 -RAM: 8GB PS: .... I will continue to investigate but i really have no clue.... Now running just unraid and Arch Vm no windows vm. Should i run linux prime95 on it and try to reproduce a load and see if its happening during heavy transcoding only? Google searchs basically say it can be anything.... ---Cry :'( :'( :'( :'( :'( :'( syslog.zip syslog-20140205-053255.zip arch.cfg windows.cfg Quote Link to comment
jbartlett Posted February 7, 2014 Share Posted February 7, 2014 PS: .... I will continue to investigate but i really have no clue.... Now running just unraid and Arch Vm no windows vm. Should i run linux prime95 on it and try to reproduce a load and see if its happening during heavy transcoding only? Google searchs basically say it can be anything.... ---Cry :'( :'( :'( :'( :'( :'( I'd try to stress components to see if it happens - but first, run memtest. Quote Link to comment
LinuxGuyGary Posted February 7, 2014 Share Posted February 7, 2014 append /xen xsave=1 cpufreq=xen:performance dom0_max_vcpus=9 dom0_vcpus_pin iommu=1 I don't believe that motherboard supports iommu ... Quote Link to comment
pyrater Posted February 7, 2014 Share Posted February 7, 2014 Good spot. it doesnt, just didnt think it was hurting anything by being in there and forgot to take it out of the cfg. Quote Link to comment
WeeboTech Posted February 7, 2014 Share Posted February 7, 2014 Yet another crash ......... This time no cache_dir and was using swap. Ran stable for about 25 hours. ... PS: .... I will continue to investigate but i really have no clue.... Now running just unraid and Arch Vm no windows vm. Should i run linux prime95 on it and try to reproduce a load and see if its happening during heavy transcoding only? Google searchs basically say it can be anything.... ---Cry :'( :'( :'( :'( :'( :'( if it were me, I would probably put the plexmedia server in it's own VM, perhaps the arch vm. Feb 5 00:08:50 Tower kernel: mdcmd (22): spindown 1 Feb 5 00:08:51 Tower kernel: mdcmd (23): spindown 3 Feb 5 00:29:47 Tower kernel: Plex Media Serv invoked oom-killer: gfp_mask=0x201da, order=0, oom_score_adj=0 Feb 5 00:29:47 Tower kernel: Plex Media Serv cpuset=/ mems_allowed=0 Feb 5 00:29:47 Tower kernel: CPU: 7 PID: 2589 Comm: Plex Media Serv Not tainted 3.10.24p-unRAID #13 Feb 5 00:29:47 Tower kernel: Hardware name: Supermicro H8DM8-2/H8DM8-2, BIOS 080014 10/22/2009 Feb 5 00:29:47 Tower kernel: 00000000000126d8 ffff88004bba1ab8 ffffffff8149830e ffff88004bba1b38 Feb 5 00:29:47 Tower kernel: ffffffff81494152 ffff88004bba1b08 ffffffff810805c3 0000000000000000 Feb 5 00:29:47 Tower kernel: 000000000002e3db 0000000000000047 ffff88004bba1d20 ffffffff8149ba72 Feb 5 00:29:47 Tower kernel: Call Trace: Even when I looked at it earlier in this thread, I thought there were way too many memory hogging plex processes going. Quote Link to comment
syntaxx Posted February 7, 2014 Share Posted February 7, 2014 If you were to put plex on a seperate vm is it under samba or nfs? By the way how's the stability for home use? Quote Link to comment
ironicbadger Posted February 7, 2014 Share Posted February 7, 2014 If you were to put plex on a seperate vm is it under samba or nfs? By the way how's the stability for home use? Your choice as to which network protocol you use. Quote Link to comment
generalz Posted February 7, 2014 Share Posted February 7, 2014 moved http://lime-technology.com/forum/index.php?topic=31810.0 syslog.txt Quote Link to comment
ironicbadger Posted February 7, 2014 Share Posted February 7, 2014 So cpus and mobo supports vt-d . passthrough works in Xenserver and vmware but when I try to pass something through on unraid it says vt-d is not enabled. Please create your own thread in the virtualization forum. This isn't relevant to the release. Post your exact hardware configs, and the above info as a minimum. Also post the output of this command grep -E "(vmx|svm)" --color=always /proc/cpuinfo Quote Link to comment
generalz Posted February 7, 2014 Share Posted February 7, 2014 ok, i wasn't sure where to post it since it works it other Os's without issue figured it was something with the Xen that's included with unraid. Quote Link to comment
bkastner Posted February 7, 2014 Share Posted February 7, 2014 I am looking at adding a 3TB disk to my 6.0-beta 3 server. I had pre-cleared the disk, then stopped the array, added the disk, and restarted the array. When I did this in 5.0 it would do a quick format of the new disk and the array would start. In 6.0 it is clearing the disk, which has taken 2+ hours so far for 37%. So I am looking at a 6 hour clearing process before I array restarts. Is this normal/expected? Or did I mess something up? Quote Link to comment
dirtysanchez Posted February 7, 2014 Share Posted February 7, 2014 I am looking at adding a 3TB disk to my 6.0-beta 3 server. I had pre-cleared the disk, then stopped the array, added the disk, and restarted the array. When I did this in 5.0 it would do a quick format of the new disk and the array would start. In 6.0 it is clearing the disk, which has taken 2+ hours so far for 37%. So I am looking at a 6 hour clearing process before I array restarts. Is this normal/expected? Or did I mess something up? FYI, I also just had the same issue, but I'm still on 5.0.4 (so probably doesn't belong in this thread ). When I assigned the new drive there was no option to format or start the array. The only option was clear. The only thing I did between preclearing the drive (2 passes) and adding it to the array was an hdparm -Tt /dev/sdx. I did preclear the drive on another machine in an eSATA drive dock, but that shouldn't be an issue. Quote Link to comment
bkastner Posted February 7, 2014 Share Posted February 7, 2014 I am looking at adding a 3TB disk to my 6.0-beta 3 server. I had pre-cleared the disk, then stopped the array, added the disk, and restarted the array. When I did this in 5.0 it would do a quick format of the new disk and the array would start. In 6.0 it is clearing the disk, which has taken 2+ hours so far for 37%. So I am looking at a 6 hour clearing process before I array restarts. Is this normal/expected? Or did I mess something up? FYI, I also just had the same issue, but I'm still on 5.0.4 (so probably doesn't belong in this thread ). When I assigned the new drive there was no option to format or start the array. The only option was clear. The only thing I did between preclearing the drive (2 passes) and adding it to the array was an hdparm -Tt /dev/sdx. I did preclear the drive on another machine in an eSATA drive dock, but that shouldn't be an issue. Okay, glad it's not just me. Tom - has this process changed for a reason? It sucks to have my array offline for 6 hours just because I want to add a new disk. Especially since it's now offline for the bulk of the afternoon. If I had known this I would have added the disk before going to bed tonight instead of killing UnRAID during daylight hours. Quote Link to comment
limetech Posted February 7, 2014 Author Share Posted February 7, 2014 I am looking at adding a 3TB disk to my 6.0-beta 3 server. I had pre-cleared the disk, then stopped the array, added the disk, and restarted the array. When I did this in 5.0 it would do a quick format of the new disk and the array would start. In 6.0 it is clearing the disk, which has taken 2+ hours so far for 37%. So I am looking at a 6 hour clearing process before I array restarts. Is this normal/expected? Or did I mess something up? FYI, I also just had the same issue, but I'm still on 5.0.4 (so probably doesn't belong in this thread ). When I assigned the new drive there was no option to format or start the array. The only option was clear. The only thing I did between preclearing the drive (2 passes) and adding it to the array was an hdparm -Tt /dev/sdx. I did preclear the drive on another machine in an eSATA drive dock, but that shouldn't be an issue. Okay, glad it's not just me. Tom - has this process changed for a reason? It sucks to have my array offline for 6 hours just because I want to add a new disk. Especially since it's now offline for the bulk of the afternoon. If I had known this I would have added the disk before going to bed tonight instead of killing UnRAID during daylight hours. I'll look into this. Quote Link to comment
peter_sm Posted February 7, 2014 Share Posted February 7, 2014 When stopping the array with VM running, it's do not do a complete stop, I have no syslog, perhaps someone else can test? I forced to do a power down, and now doing a parity check. //Peter Quote Link to comment
prtsmgr Posted February 7, 2014 Share Posted February 7, 2014 Peter, I am in no way as knowledgeable as most of you guys here, but I did have a similar situation when testing beta 3. In my case, the cache drive was still mounted in the console/terminal and unraid would not stop/shutdown. I am only mentioning it in case it may help you or others. Mark When stopping the array with VM running, it's do not do a complete stop, I have no syslog, perhaps someone else can test? I forced to do a power down, and now doing a parity check. //Peter Quote Link to comment
MrLondon Posted February 7, 2014 Share Posted February 7, 2014 hi there I have started to use the xen guest with Arch yesterday, since I have done that every time I am transferring data to the drive while the xen is running the transfer speed is really bad, it's like 100kb, I have a gigabit Ethernet wired. Any reason this is happening. I have plex running in the VM, nothing else really. Quote Link to comment
SchoolBusDriver Posted February 7, 2014 Share Posted February 7, 2014 hi there I have started to use the xen guest with Arch yesterday, since I have done that every time I am transferring data to the drive while the xen is running the transfer speed is really bad, it's like 100kb, I have a gigabit Ethernet wired. Any reason this is happening. I have plex running in the VM, nothing else really. Are you using PV Drivers in the VM? Did you "tweak" NFS or Samba on both unRAID and within the VM? Lots of variables... You need to provide more info for us to assist. Quote Link to comment
MrLondon Posted February 7, 2014 Share Posted February 7, 2014 hi there I have started to use the xen guest with Arch yesterday, since I have done that every time I am transferring data to the drive while the xen is running the transfer speed is really bad, it's like 100kb, I have a gigabit Ethernet wired. Any reason this is happening. I have plex running in the VM, nothing else really. Are you using PV Drivers in the VM? Did you "tweak" NFS or Samba on both unRAID and within the VM? Lots of variables... You need to provide more info for us to assist. I used the ready compiled VM from the Virtualization subforum, I didn't change anything, I went with the defaults, I am using NFS The VM itself is running on a cache drive, but I am copying direct to the array. Tasks: 149 total, 1 running, 148 sleeping, 0 stopped, 0 zombie Cpu(s): 0.2%us, 0.6%sy, 0.0%ni, 35.4%id, 63.4%wa, 0.0%hi, 0.2%si, 0.3%st Mem: 1942208k total, 1781260k used, 160948k free, 90684k buffers Swap: 0k total, 0k used, 0k free, 1375252k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 21719 nobody 20 0 285m 15m 12m D 4 0.8 3:20.47 smbd 1780 root 20 0 1010m 9236 900 S 3 0.5 25:14.65 shfs 3575 root 20 0 2158m 6372 3940 S 1 0.3 2:09.56 qemu-system-i38 1618 root 20 0 0 0 0 S 1 0.0 2:44.29 unraidd 21718 nobody 20 0 283m 9832 7644 S 0 0.5 0:11.92 smbd 21873 root 20 0 0 0 0 D 0 0.0 0:01.12 kworker/2:4 21952 root 20 0 0 0 0 D 0 0.0 0:04.18 kworker/u8:1 22179 root 20 0 13272 1272 948 R 0 0.1 0:00.56 top 1 root 20 0 4356 600 516 S 0 0.0 0:10.89 init 2 root 20 0 0 0 0 S 0 0.0 0:00.04 kthreadd 3 root 20 0 0 0 0 S 0 0.0 0:42.10 ksoftirqd/0 5 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/0:0H 7 root RT 0 0 0 0 S 0 0.0 0:00.31 migration/0 8 root 20 0 0 0 0 S 0 0.0 0:00.00 rcu_bh 9 root 20 0 0 0 0 S 0 0.0 0:08.70 rcu_sched 10 root RT 0 0 0 0 S 0 0.0 0:00.21 migration/1 11 root 20 0 0 0 0 S 0 0.0 0:00.26 ksoftirqd/1 syslog_07022014.txt Quote Link to comment
ironicbadger Posted February 7, 2014 Share Posted February 7, 2014 Guys, this is a release thread. Please create your own thread in the virtualization forum for your own problems! (If it's my Arch VM you're using, post in that thread if you like). Quote Link to comment
MrLondon Posted February 7, 2014 Share Posted February 7, 2014 ironbadger actually I now did the xl destroy command to stop xen but the performance is still really bad. root@Tower:~# free total used free shared buffers cached Mem: 1940788 1638556 302232 0 119176 1207932 -/+ buffers/cache: 311448 1629340 Swap: 0 0 0 Tasks: 146 total, 1 running, 145 sleeping, 0 stopped, 0 zombie Cpu(s): 0.3%us, 0.9%sy, 0.0%ni, 43.9%id, 54.7%wa, 0.0%hi, 0.1%si, 0.0%st Mem: 1940788k total, 1815312k used, 125476k free, 120116k buffers Swap: 0k total, 0k used, 0k free, 1385144k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 21719 nobody 20 0 287m 15m 12m D 4 0.8 5:11.75 smbd 1780 root 20 0 1009m 8424 900 S 3 0.4 26:21.21 shfs 1618 root 20 0 0 0 0 S 1 0.0 2:55.96 unraidd 21952 root 20 0 0 0 0 D 1 0.0 0:10.78 kworker/u8:1 3 root 20 0 0 0 0 S 0 0.0 0:43.52 ksoftirqd/0 21718 nobody 20 0 285m 8044 5844 S 0 0.4 0:22.06 smbd 22465 root 20 0 0 0 0 D 0 0.0 0:01.94 kworker/0:0 1 root 20 0 4356 600 516 S 0 0.0 0:10.91 init 2 root 20 0 0 0 0 S 0 0.0 0:00.04 kthreadd 5 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/0:0H 7 root RT 0 0 0 0 S 0 0.0 0:00.31 migration/0 8 root 20 0 0 0 0 S 0 0.0 0:00.00 rcu_bh 9 root 20 0 0 0 0 S 0 0.0 0:09.31 rcu_sched 10 root RT 0 0 0 0 S 0 0.0 0:00.21 migration/1 11 root 20 0 0 0 0 S 0 0.0 0:00.27 ksoftirqd/1 13 root 0 -20 0 0 0 S 0 0.0 0:00.00 kworker/1:0H 14 root RT 0 0 0 0 S 0 0.0 0:00.26 migration/2 it has taken 39 min to copy a 1.4gb file, that must be a new record surely. any suggestions. Quote Link to comment
limetech Posted February 7, 2014 Author Share Posted February 7, 2014 Guys, this is a release thread. Please create your own thread in the virtualization forum for your own problems! (If it's my Arch VM you're using, post in that thread if you like). Like herding cats, right? Quote Link to comment
prostuff1 Posted February 7, 2014 Share Posted February 7, 2014 Guys, this is a release thread. Please create your own thread in the virtualization forum for your own problems! (If it's my Arch VM you're using, post in that thread if you like). Like herding cats, right? Quote Link to comment
starcat Posted February 7, 2014 Share Posted February 7, 2014 Is there somewhere a post describing how to install 6b3 on its own SSD disk (bare metal, no VM) and using the flash drive for license only? Thanks Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.