October 14, 20187 yr Quote Linux kernel: version 4.18.14 OOT Intel 10gbps netowrk drivers: ixgbe-5.5.1, ixgbevf-4.5.1 Unfortunately, this update has still not fixed the issue I have with my onboard Intel 10gbps NICs (Supermicro X11SPH-nCTF). The boot syslog shows this error, just like V6.61: Oct 14 18:57:58 unSERVER kernel: i40e 0000:67:00.0: Error I40E_AQ_RC_EINVAL adding RX filters on PF, promiscuous mode forced on Oct 14 18:57:59 unSERVER kernel: i40e 0000:67:00.1: Error I40E_AQ_RC_EINVAL adding RX filters on PF, promiscuous mode forced on I did some research online and it seems like this supposedly should have been fixed with new network drivers from Intel: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=892105 But not in my case, unfortunately.
October 14, 20187 yr Update 2 I joined the unraid server to the domain. Windows 10 boxes can see \\tower but openelec wants a password i inputed domain user and password and is working Windows XP does work after entering domain user/pass for connected share. Yea team
October 14, 20187 yr 7 hours ago, dlandon said: The new Docker container size display is crowding on my system. 7 hours ago, bonienl said: Can you empty your browser's cache and try again @bonienl @dlandon This only happens when the browser's font size has been enlarged beyond its "normal" (on my system, "large" is fine, but "very large" shows the same thing as the pic. Edited October 14, 20187 yr by Squid
October 14, 20187 yr 36 minutes ago, Squid said: @bonienl @dlandon This only happens when the browser's font size has been enlarged beyond its "normal" (on my system, "large" is fine, but "very large" shows the same thing as the pic. It was actually zoom set to 110%.
October 14, 20187 yr Seems this has fixed my issue with a Mellanox 10Gbe NIC. Thank you. Updated from 6.5.3
October 14, 20187 yr 3 hours ago, highdefinitely said: Unfortunately, this update has still not fixed the issue I have with my onboard Intel 10gbps NICs (Supermicro X11SPH-nCTF). The boot syslog shows this error, just like V6.61: Oct 14 18:57:58 unSERVER kernel: i40e 0000:67:00.0: Error I40E_AQ_RC_EINVAL adding RX filters on PF, promiscuous mode forced on Oct 14 18:57:59 unSERVER kernel: i40e 0000:67:00.1: Error I40E_AQ_RC_EINVAL adding RX filters on PF, promiscuous mode forced on I did some research online and it seems like this supposedly should have been fixed with new network drivers from Intel: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=892105 But not in my case, unfortunately. Your device doesn't use those drivers you quoted. It uses the i40e driver. Current OOB version from intel is 2.4.10 No idea if it compiles with latest kernel's yet... Intel seems to be dragging there feet.
October 14, 20187 yr This update has completely messed up my system. After updating, it is just stuck on mounting the drive with the array starting. I have downgraded back to 6.6.1 and even tried updating to 6.6.0 -RC4 and nothing works. I've run xfs repair with no problems. The only thing that will start my array back up is to disable my user shares.
October 14, 20187 yr 56 minutes ago, jwiese997 said: This update has completely messed up my system. After updating, it is just stuck on mounting the drive with the array starting. I have downgraded back to 6.6.1 and even tried updating to 6.6.0 -RC4 and nothing works. I've run xfs repair with no problems. The only thing that will start my array back up is to disable my user shares. Re-enable user shares and post your diagnostics (might have to do via the command line) diagnostics
October 14, 20187 yr 6 minutes ago, Squid said: Re-enable user shares and post your diagnostics (might have to do via the command line) diagnostics Here they are. tower-diagnostics-20181014-1833.zip
October 14, 20187 yr 13 minutes ago, jwiese997 said: Here they are. tower-diagnostics-20181014-1833.zip If I was going to guess, its all these lines Oct 14 18:33:43 Tower nginx: 2018/10/14 18:33:43 [error] 3903#3903: *2107 FastCGI sent in stderr: "Primary script unknown" while reading response header from upstream, client: 192.168.1.15, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock:", host: "tower", referrer: "http://tower/Main" Try booting into safe mode. If that works, delete every reference on /boot/config/plugins to unassigned devices (since it's not even installed). And close any other open browser windows. Edited October 14, 20187 yr by Squid
October 14, 20187 yr 31 minutes ago, Squid said: If I was going to guess, its all these lines Oct 14 18:33:43 Tower nginx: 2018/10/14 18:33:43 [error] 3903#3903: *2107 FastCGI sent in stderr: "Primary script unknown" while reading response header from upstream, client: 192.168.1.15, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock:", host: "tower", referrer: "http://tower/Main" Try booting into safe mode. If that works, delete every reference on /boot/config/plugins to unassigned devices (since it's not even installed). And close any other open browser windows. It did boot into safe mode. I removed what you said to, but it didn't fix the issue. Here's the updated diagnostic tower-diagnostics-20181014-1918.zip
October 15, 20187 yr 48 minutes ago, jwiese997 said: It did boot into safe mode. I removed what you said to, but it didn't fix the issue. Here's the updated diagnostic tower-diagnostics-20181014-1918.zip I finally figured it out. It was my rclone script that was causing it to go bonkers. I've corrected that and have updated to 6.6.2 with no problems now. Thanks for the help.
October 15, 20187 yr Did update two server to 6.6.2 this morning. All user shares are gone now - on both servers. "Enable user shares" is set to "Yes" in "Global Share settings". "User Shares" list on "Shares" page is empty. Diagnostics of one server attached. tower-diagnostics-20181015-0817.zip
October 15, 20187 yr 6 minutes ago, hawihoney said: Did update two server to 6.6.2 this morning. All user shares are gone now - on both servers. "Enable user shares" is set to "Yes" in "Global Share settings". "User Shares" list on "Shares" page is empty. Diagnostics of one server attached. tower-diagnostics-20181015-0817.zip Looks like a package conflict with one of them from DevPack or maybe one of the other plugins? Try Safe Mode and see if your shares are back.
October 15, 20187 yr What is "Safe Mode"? I remember in old days there was a checkbox called "Maintenance Mode" or so. I can't find it.
October 15, 20187 yr Couldn't wait. Need these servers. Did delete all txz files in /boot/config/plugins/[DevPack|NerdPack]/[6.5|6.6]. User Shares are back again. Thanks for now. The DevPack/NerdPack plugins are of great help. But they are always a bulk of problems. Is there a way to bind them more closely to main Unraid in case of version compatibility and compile options?
October 15, 20187 yr 1 hour ago, hawihoney said: Couldn't wait. Need these servers. Did delete all txz files in /boot/config/plugins/[DevPack|NerdPack]/[6.5|6.6]. User Shares are back again. Thanks for now. The DevPack/NerdPack plugins are of great help. But they are always a bulk of problems. Is there a way to bind them more closely to main Unraid in case of version compatibility and compile options? The Devpack/Nerdpack plugins are an extreme example of the basic problem with using plugins (rather than dockers). Every time unRAID is updated all packages that can be installed need to be reviewed (and potentially updated) for compatibility with the new release. If a plugin overwrites any components used by core Unraid it can break things. Since Limetech take no responsibility for testing against plugins it is up to the maintainers of the plugins to do this. I cannot see any way to easily resolve this. One thing that might help with identifying if plugins are causing issues could be to force the reboot after installing an OS update to come up in Safe Mode. It might also help to make the Safe Mode option more visible in the GUI as I am sure that many who run Unraid in headless mode may not even realize (or remember) that this is a boot time option.
October 15, 20187 yr Update from 6.6.1 went fine so far except from the VM restart issue reported couple posts above. Restarting a VM from inside or within Unraid causing Unraid to show the VM as "not running". You can't start the VM from the web gui nor from the cl. Only way to fix it is to stop the VM from inside.
October 15, 20187 yr 12 minutes ago, itimpi said: The Devpack/Nerdpack plugins are an extreme example of the basic problem with using plugins (rather than dockers). Every time unRAID is updated all packages that can be installed need to be reviewed (and potentially updated) for compatibility with the new release. If a plugin overwrites any components used by core Unraid it can break things. Since Limetech take no responsibility for testing against plugins it is up to the maintainers of the plugins to do this. I cannot see any way to easily resolve this. I'm an old Perl coder and do use a ton of GUI-less scripts. So Perl (with some additional modules) and SQLite are the things I need most. Do you know about a docker for something like that?
October 15, 20187 yr I could not see anything from the Release Notes, but does anyone know if 6.6.2 fixes the issue of Custom Parity Check Schedules being ignored. I have Checks scheduled for the first Monday of the month every 3rd month, and since updating to 6.6.1 Parity Checks have commenced EVERY Monday.
October 15, 20187 yr just updated from 6.5.3, everything works oke, it's only kills my VM xml files! i cant finds them! if i recreate the vm with the (still there) .img files, will the VM be back? 1 harddisk didn't survived the reboot, swapped it for a new one and its currently rebuilding... this is the first time i 'test' the paity function one thing what still wont work is the temperature monitoring of the Threadrippers chips... i do use NFS and SMB, will monitor it... Edited October 15, 20187 yr by sjaak
October 15, 20187 yr 18 minutes ago, sjaak said: it's only kills my VM xml files! This is likely a problem with libvirt.img and not directly related to this release, you should start a thread on the general support forum and don't forget to please post your diagnostics: Tools -> Diagnostics
October 15, 20187 yr Wanted to report in that on 6.6.2 with the ipmi/kvm java console it is working fine for my H11SSL-NC w/ ASPEED AST2500 BMC.
Archived
This topic is now archived and is closed to further replies.