unRAID OS version 6.4.0 Stable Release Available


limetech

Recommended Posts

39 minutes ago, FreeMan said:

I'll have to restart in safe mode when I get home this evening. Is there value in letting the parity check finish?

If the previous checks have not had errors AND this one does have any at the point where I looked it it before rebooting, I (personally) would not hesitate to stop it.  However, after figuring out what the problem was(is?) and the server had been staple for a time that I was sure it was fixed, I would probably do a parity check just to be comfortable that everything was back to normal.

Edited by Frank1940
Link to comment

Hi Gents,

 

Since I updated my server to 6.4.0, I keep having those errors in the log when I boot in normal mode

I did uninstall a whole bunch of plugins (preclear, unassigned devices, user's scripts...) but won't fix anything.

 

Jan 15 23:33:01 MAXIMUS kernel: CPU: 3 PID: 21564 Comm: kworker/u32:6 Not tainted 4.14.13-unRAID #1
Jan 15 23:33:01 MAXIMUS kernel: Hardware name: Supermicro X9SRE/X9SRE-3F/X9SRi/X9SRi-3F/X9SRE/X9SRE-3F/X9SRi/X9SRi-3F, BIOS 3.2 01/16/2015
Jan 15 23:33:01 MAXIMUS kernel: Workqueue: scsi_tmf_7 scmd_eh_abort_handler
Jan 15 23:33:01 MAXIMUS kernel: task: ffff880807380000 task.stack: ffffc90007840000
Jan 15 23:33:01 MAXIMUS kernel: RIP: 0010:isci_task_abort_task+0x18/0x334 [isci]
Jan 15 23:33:01 MAXIMUS kernel: RSP: 0018:ffffc90007843ce8 EFLAGS: 00010296
Jan 15 23:33:01 MAXIMUS kernel: RAX: ffffffffa01d6cbb RBX: ffff880818fef1a8 RCX: 0000000000000000
Jan 15 23:33:01 MAXIMUS kernel: RDX: ffff88085ec18020 RSI: 0000000000000400 RDI: 0000000000000000
Jan 15 23:33:01 MAXIMUS kernel: RBP: 0000000000000000 R08: 00000000000000d2 R09: 0000000000000000
Jan 15 23:33:01 MAXIMUS kernel: R10: ffffc90007843e58 R11: ffff8808073800c0 R12: ffff8808562d6800
Jan 15 23:33:01 MAXIMUS kernel: R13: 0000000000000000 R14: 0000000000000008 R15: 0000000000000000
Jan 15 23:33:01 MAXIMUS kernel: FS: 0000000000000000(0000) GS:ffff88085f2c0000(0000) knlGS:0000000000000000
Jan 15 23:33:01 MAXIMUS kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jan 15 23:33:01 MAXIMUS kernel: CR2: 0000000000000000 CR3: 0000000001c0a003 CR4: 00000000000626e0
Jan 15 23:33:01 MAXIMUS kernel: Call Trace:
Jan 15 23:33:01 MAXIMUS kernel: ? __freed_request+0x34/0x86
Jan 15 23:33:01 MAXIMUS kernel: ? freed_request+0x31/0x4b
Jan 15 23:33:01 MAXIMUS kernel: ? __blk_put_request+0xed/0x131
Jan 15 23:33:01 MAXIMUS kernel: ? blk_put_request+0x38/0x4b
Jan 15 23:33:01 MAXIMUS kernel: ? scsi_execute+0x168/0x177
Jan 15 23:33:01 MAXIMUS kernel: ? scsi_test_unit_ready+0x45/0x99
Jan 15 23:33:01 MAXIMUS kernel: ? cpuacct_charge+0x2c/0x6b
Jan 15 23:33:01 MAXIMUS kernel: sas_eh_abort_handler+0x2a/0x3c [libsas]
Jan 15 23:33:01 MAXIMUS kernel: scmd_eh_abort_handler+0x35/0x8f
Jan 15 23:33:01 MAXIMUS kernel: process_one_work+0x146/0x239
Jan 15 23:33:01 MAXIMUS kernel: ? rescuer_thread+0x258/0x258
Jan 15 23:33:01 MAXIMUS kernel: worker_thread+0x1c3/0x292
Jan 15 23:33:01 MAXIMUS kernel: kthread+0x10f/0x117
Jan 15 23:33:01 MAXIMUS kernel: ? kthread_create_on_node+0x3a/0x3a

I also include the diagnostics that I downloaded by booting in safe mode.

Safe mode is 100% functionnal but in normal mode, even if the array starts and runs smoothly, after 10mins, I can't access to any of my shares (except on cache disk) from the server itself (by using Dolphin browser) or from PC on the network.

All the dockers and VMs are running smoothly. Of course, Plex works but can't play anything since it can't get access to the data.

 

Any idea?

 

Cheers!

maximus-diagnostics-20180118-1727.zip

Link to comment
5 hours ago, GHunter said:

Hmm, ok. I was thinking that this worked under previous stable releases but I guess I'm wrong.

 

Thanks for the reply.

 

I can confirm the same issue with 6.3.5 and prior.  Every time i exited a VM, a manual browser refresh was required to see if it had really stopped. Starting or forcing a VM stop refreshes the status.  I always found this inconsistency a bit odd, but, it was not introduced in 6.4.0.  It's been there a while.

Link to comment
1 minute ago, Hoopster said:

 

I can confirm the same issue with 6.3.5 and prior.  Every time i exited a VM, a manual browser refresh was required to see if it had really stopped. Starting or forcing a VM stop refreshes the status.  I always found this inconsistency a bit odd, but, it was not introduced in 6.4.0.  It's been there a while.

+1

Link to comment
On 1/16/2018 at 5:28 PM, johnnie.black said:

Do you see or can get into the motherboard bios? If not the problem isn't unRAID, if yes when do you lose video output?

 

No, I never saw any output to the monitor.  However, about 10 minutes after writing this post, the server came up and has been working fine since then.  I don't know what happened, but it seems all good now.

Link to comment
7 minutes ago, Perforator said:

Is there a list of plugins that are not compatible with 6.4 somewhere?

 

Anything that is outright incompatible with 6.4 FCP will flag you if installed.  Additionally, the Apps tab will not let you install anything that is outright incompatible either.

Link to comment

Everything seems ok so far, noticed some of the vmwrite error things popping up, so I'll place it here.

 

Jan 19 14:46:57 YES-MEDIASERVER dnsmasq-dhcp[11941]: read /var/lib/libvirt/dnsmasq/default.hostsfile
Jan 19 14:46:57 YES-MEDIASERVER kernel: virbr0: port 1(virbr0-nic) entered disabled state
Jan 19 14:46:57 YES-MEDIASERVER kernel: vfio-pci 0000:02:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem
Jan 19 14:46:57 YES-MEDIASERVER kernel: br0: port 2(vnet0) entered blocking state
Jan 19 14:46:57 YES-MEDIASERVER kernel: br0: port 2(vnet0) entered disabled state
Jan 19 14:46:57 YES-MEDIASERVER kernel: device vnet0 entered promiscuous mode
Jan 19 14:46:57 YES-MEDIASERVER kernel: br0: port 2(vnet0) entered blocking state
Jan 19 14:46:57 YES-MEDIASERVER kernel: br0: port 2(vnet0) entered forwarding state
Jan 19 14:46:57 YES-MEDIASERVER kernel: vmwrite error: reg 810 value 0 (err 0)
Jan 19 14:46:57 YES-MEDIASERVER kernel: CPU: 8 PID: 12092 Comm: CPU 0/KVM Not tainted 4.14.13-unRAID #1
Jan 19 14:46:57 YES-MEDIASERVER kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./X99M Extreme4, BIOS P3.20 08/16/2016
Jan 19 14:46:57 YES-MEDIASERVER kernel: Call Trace:
Jan 19 14:46:57 YES-MEDIASERVER kernel: dump_stack+0x5d/0x79
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_lapic_reset+0x278/0x292 [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_create_lapic+0xa5/0xb2 [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_arch_vcpu_init+0xd9/0x22d [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_vcpu_init+0xcd/0xee [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: vmx_create_vcpu+0x54/0xd5c [kvm_intel]
Jan 19 14:46:57 YES-MEDIASERVER kernel: ? pick_next_task_fair+0x102/0x3d9
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_vm_ioctl+0x13b/0x66c [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: ? __schedule+0x4ac/0x4e9
Jan 19 14:46:57 YES-MEDIASERVER kernel: vfs_ioctl+0x16/0x23
Jan 19 14:46:57 YES-MEDIASERVER kernel: do_vfs_ioctl+0x4f1/0x504
Jan 19 14:46:57 YES-MEDIASERVER kernel: SyS_ioctl+0x39/0x55
Jan 19 14:46:57 YES-MEDIASERVER kernel: entry_SYSCALL_64_fastpath+0x1e/0x81
Jan 19 14:46:57 YES-MEDIASERVER kernel: RIP: 0033:0x148614277727
Jan 19 14:46:57 YES-MEDIASERVER kernel: RSP: 002b:000014860ff72228 EFLAGS: 00000246
Jan 19 14:46:57 YES-MEDIASERVER kernel: vmwrite error: reg 810 value 0 (err 0)
Jan 19 14:46:57 YES-MEDIASERVER kernel: CPU: 0 PID: 12097 Comm: CPU 1/KVM Not tainted 4.14.13-unRAID #1
Jan 19 14:46:57 YES-MEDIASERVER kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./X99M Extreme4, BIOS P3.20 08/16/2016
Jan 19 14:46:57 YES-MEDIASERVER kernel: Call Trace:
Jan 19 14:46:57 YES-MEDIASERVER kernel: dump_stack+0x5d/0x79
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_lapic_reset+0x278/0x292 [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_create_lapic+0xa5/0xb2 [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_arch_vcpu_init+0xd9/0x22d [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_vcpu_init+0xcd/0xee [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: vmx_create_vcpu+0x54/0xd5c [kvm_intel]
Jan 19 14:46:57 YES-MEDIASERVER kernel: ? pick_next_task_fair+0x102/0x3d9
Jan 19 14:46:57 YES-MEDIASERVER kernel: kvm_vm_ioctl+0x13b/0x66c [kvm]
Jan 19 14:46:57 YES-MEDIASERVER kernel: ? __schedule+0x4ac/0x4e9
Jan 19 14:46:57 YES-MEDIASERVER kernel: vfs_ioctl+0x16/0x23
Jan 19 14:46:57 YES-MEDIASERVER kernel: do_vfs_ioctl+0x4f1/0x504
Jan 19 14:46:57 YES-MEDIASERVER kernel: SyS_ioctl+0x39/0x55
Jan 19 14:46:57 YES-MEDIASERVER kernel: entry_SYSCALL_64_fastpath+0x1e/0x81
Jan 19 14:46:57 YES-MEDIASERVER kernel: RIP: 0033:0x148614277727

 

Link to comment

Home server showed up some errors as well.

 

 

Seemed to be related to memory, multiple entries in the log

Jan 19 21:41:53 BEAST kernel: EDAC sbridge: CPU SrcID #0, Ha #0, Channel #0 has DIMMs, but ECC is disabled
Jan 19 21:41:53 BEAST kernel: EDAC sbridge: Couldn't find mci handler
Jan 19 21:41:53 BEAST kernel: EDAC sbridge: Couldn't find mci handler
Jan 19 21:41:53 BEAST kernel: EDAC sbridge: Failed to register device with error -19.

 

And the vmwrite error, had an entry for each VM

Jan 19 21:42:16 BEAST kernel: br0: port 3(vnet1) entered forwarding state
Jan 19 21:42:16 BEAST kernel: vmwrite error: reg 810 value 0 (err 0)
Jan 19 21:42:16 BEAST kernel: CPU: 12 PID: 7591 Comm: CPU 0/KVM Not tainted 4.14.13-unRAID #1
Jan 19 21:42:16 BEAST kernel: Hardware name: MSI MS-7885/X99A SLI PLUS(MS-7885), BIOS 1.D0 07/15/2016
Jan 19 21:42:16 BEAST kernel: Call Trace:

 

Link to comment
1 hour ago, Frank1940 said:

@ryoko227 you should probably post the diagnostics files also for both of your servers which include the time period where these problems occurred. 

 

After a reboot of the server those mem errors cleared out, only thing erroring at home now is the vmwrites, so I think we are in the clear ^_^v

Link to comment
On 1/15/2018 at 8:10 PM, cybrnook said:

Worked fine for me:

image.thumb.png.03c2a131c34bff5ea72bb42b1180bae1.png

I want to add back to this, that I did have to add server: as the first line. Seems there is a semi-bug in pfsense around unbound and custom config options. Last line:

https://redmine.pfsense.org/issues/6072

5a63765c104b4_pfSenseunbound.thumb.PNG.a06cba87e05266143bc7296d73ca884e.PNG

The server:include line was when I found out I needed the preceeding server: when trying to add some dnsbl feeds.

Edited by cybrnook
Link to comment

All working OK here.

 

I've however noticed my pfSense CPU load when hammering my connection at 25MB/sec (220Mbits) has jumped from ~10% to the best part of 40%.

 

So effectively pfSense is now running 4 times worse than it was 9_9 and theoretically no longer capable of running a gigabit connection at full speed (when/if I update to such a connection!)

 

Can we turn this patch off? (Frankly the security side of things I don't give a damn about..!)

Edited by Interstellar
Link to comment

Hi All,

 

Got a really strange one today.  I can't find any references to this elsewhere on the forum, so here goes.

 

Tried to update from 6.4.0 rc18 to 6.4.0 stable using the button in the Tools area.  Got this after the download apparently aborted.

 

image.png.3a9254fdc27a83333a5d233796d878ff.png

 

SSH'd into the directory and noted that there was zero free space, which was consistent all the way up that path, until I get up to the root, where I then see something like 93% free space...

 

I'm suspecting a failed or failing flash drive, which wouldn't surprise me since it's hosted unRAID since 2008.  I was able to mount it and successfully copy off all the files.

 

I did get a diagnostic dump but every file in it is 0 bytes in length except for TOP.  At this point, the main web page was loaded with all sorts of Dynamix related error strings, which alternated between the usual information and the errors every 5 seconds or so, and rapidly became non-responsive.  The GUI hosted locally on the server seemed to work or at least didn't show the same behaviour as the remote webGUI.

 

Up to this point, unRAID has performed pretty much flawlessly.  Any feedback on this would be appreciated.

 

Thanks!!

UNRAID syslog.txt

tower-diagnostics-20180120-1214.zip

Link to comment
1 hour ago, Craigb said:

Hi All,

 

Got a really strange one today.  I can't find any references to this elsewhere on the forum, so here goes.

 

Tried to update from 6.4.0 rc18 to 6.4.0 stable using the button in the Tools area.  Got this after the download apparently aborted.

 

image.png.3a9254fdc27a83333a5d233796d878ff.png

 

SSH'd into the directory and noted that there was zero free space, which was consistent all the way up that path, until I get up to the root, where I then see something like 93% free space...

 

I'm suspecting a failed or failing flash drive, which wouldn't surprise me since it's hosted unRAID since 2008.  I was able to mount it and successfully copy off all the files.

 

I did get a diagnostic dump but every file in it is 0 bytes in length except for TOP.  At this point, the main web page was loaded with all sorts of Dynamix related error strings, which alternated between the usual information and the errors every 5 seconds or so, and rapidly became non-responsive.  The GUI hosted locally on the server seemed to work or at least didn't show the same behaviour as the remote webGUI.

 

Up to this point, unRAID has performed pretty much flawlessly.  Any feedback on this would be appreciated.

 

Thanks!!

UNRAID syslog.txt

tower-diagnostics-20180120-1214.zip

 

It's going to wind up being that your RAM is completely full.  a reboot followed by a quick install should do the trick.

 

Unfortunately though the diagnostics is basically empty because of this, and the syslog only shows what is expected to happen when this happens.  How much memory do you have installed on the server?

Link to comment
1 hour ago, Craigb said:

Up to this point, unRAID has performed pretty much flawlessly.  Any feedback on this would be appreciated.

This is probably going to be a symptom of some misconfiguration, such as a docker not mapped correctly and so is writing into the unRAID OS file system, which is mostly RAM

Link to comment
1 hour ago, Nem said:

anyone notice, since updating, that the FTP server is re-enabled automatically after a reboot? I've tried disabling it and removing all ftp users from the list, but it still gets enabled again...

It's always been that way.  If you want it off, install the Tips and Tweaks plugin and disable the ftp.

Link to comment

Thanks guys!  It's 2 GB on the RAM.  And now that you mention memory being overwritten, what I observed makes sense.  Took a deep breath and rebooted successfully.  Tried to update the OS and it failed again in exactly the same way. 

 

image.thumb.png.dedc8ed6fcc1c9d8ba0b83480601c57c.png

 

Tried a third time, with a bit less than half the 2GB of RAM free, and again, the same result.

 

image.png.fdabb5e549a960dbec8d503fd18dfa73.png

 

Everything seems fine with the exception of the Dynamix Stats plugin which I've uninstalled.

 

The server is back on service now, thanks for the pointers!

 

Craig

 

 

 

 

 

image.png

Link to comment

@bonienl might I suggest on the next release, the shadowing also be inverted. For example, when on the plugins tab, you have the background as the greyish color, but the selected tab is the darkened color. To me, at least, it makes more sense for the in focus tab to match the background, with the out of focus tab be the darker of the two. For example, below I have the "Installed Plugins" tab selected, which to me seems backwards to the "Install Plugin" tab.:

5a6377906bf1f_IncorrectShadowing.PNG.b228c475249f49244781b615df1a67ae.PNG

Edited by cybrnook
  • Like 1
Link to comment
@Craigb  Said 

Thanks guys!  It's 2 GB on the RAM. 

 

Problem is that the entire download is stored in RAM and then unzipped to a directory in RAM (NOT the flash Drive!).  The download coupled with the amount of RAM that all versions of 6.0.X now use (with its increased features), will not fit in 2GB of RAM.  You will either have to install more RAM or download the zipped files and copy over those bz* files as @squib  suggests.  

Edited by Frank1940
Link to comment
  • limetech unpinned and locked this topic
Guest
This topic is now closed to further replies.