Jump to content
limetech

unRAID OS version 6.5.0-rc5 available

46 posts in this topic Last Reply

Recommended Posts

Please remember to open a new topic in this board for specific issues.

Version 6.5.0-rc5 2018-03-05

Linux kernel:

Management:

  • webgui: Improved docker used ports and IPs list
  • webgui: Enhancements to Docker settings and overview

Version 6.5.0-rc4 2018-03-03

Management:

  • fix: accommodate buggy dhcp servers which hand out a single host IPv6 address

Version 6.5.0-rc3 2018-03-03

Summary:

  • Updated ntp which addresses some CVE's and restore default NTP enabled, referencing google public NTP servers.
  • Restored console webGUI operation.
  • Misc. bug fixes.

Base distro:

  • gawk: version 4.2.1
  • libunistring: version 0.9.9
  • libxshmfence: version 1.3
  • lxterminal: version 0.3.1
  • mesa: version 17.3.6
  • ntp: version 4.2.8p11 (CVE-2016-1549, CVE-2018-7182, CVE-2018-7170, CVE-2018-7184, CVE-2018-7185)
  • p11-kit: version 0.23.10
  • php: version 7.2.3
  • tevent: version 0.9.36
  • vte3: version 0.44.3
  • xfsprogs: version 4.15.1

Management:

  • accommodate buggy dhcp servers which hand out a single host IPv6 address
  • fix desktop gui mode by re-enabling graphics compositing due to new mesa version
  • upon upgrade strip metric value from GATEWAY if present
  • shfs: support run-time logging level change
  • webgui: Fixed regression error in Docker settings
  • webgui: Fixed version/date extraction in previous unRAID version
  • webgui: Updated tablesorter to v2.29.6 and fixed sorting issue in browser
  • webgui: Corrected PHP warnings

Version 6.5.0-rc1 2018-02-28

Summary:

  • Updated linux kernel includes more patches related to Meltdown/Spectre.
  • Refreshed base packages to latest versions.
  • We added a workaround for Safari to display CPU Load Statistics correctly.
  • We have disabled NTP by default. For new installations we suggest using 0.pool.ntp.org as described in their documentation. Note that unRAID OS will preserve the ntp.drift file across system reboots on your USB Flash boot device.
  • Resident network guru bonienl continues to make networking refinements:
    • Fixed default gateway with different metrics. Now same gateway with multiple metrics properly set up
    • Fixed route_up and route_down functions. Now adding and removing routes handles IPv6 link local addresses and it is allowed to add the same route multiple times with different metric and interface.
    • Routing table shows interface name too with default routes
    • Network settings page is updated thru a websocket (/sub/dhcp), this ensures info is always present. Another nice advantage of websockets.
    • Cleanup of network.cfg file. Only necessary settings are stored. This greatly reduces the number of entries.
    • Some fixes in the network settings page itself, better user experience.
    • IPv6 routes have their own metric value, can be set independently of IPv4
    • Docker fixes for custom networks
    • Some code cleanup
  • Another nice feature added by bonienl is the ability to downgrade to the previous unRAID-OS release.
  • Upgraded noVNC. This is very much improved, give it a try!

Base distro:

  • adwaita-icon-theme: version 3.26.1
  • at-spi2-atk: version 2.26.1
  • at-spi2-core: version 2.26.2
  • atk: version 2.26.1
  • bash: version 4.4.019
  • btrfs-progs: version v4.15.1
  • cairo: version 1.14.12
  • coreutils: version 8.29
  • cyrus-sasl: version 2.1.26 (rev3)
  • dbus: version 1.12.4
  • dhcpcd: version 7.0.1
  • docker: version 17.12.1
  • e2fsprogs: version 1.43.9
  • etc: version 15.0
  • ethtool: version 4.15
  • eudev: version 3.2.5
  • freetype: version 2.9
  • fuse: version 2.9.7 (inline_name size 64 patch)
  • gawk: version 4.2.0
  • glib2: version 2.54.3
  • glibc-solibs: version 2.27
  • glibc-zoneinfo: version 2018c
  • glibc: version 2.27
  • gnutls: version 3.6.2
  • gtk+3: version 3.22.28
  • guile: version 2.2.3
  • gzip: version 1.9
  • harfbuzz: version 1.7.5
  • hdparm: version 9.54
  • hostname: version 3.18
  • htop: version 2.1.0
  • icu4c: version 60.2
  • inetd: version 1.79s (rev10)
  • iproute2: version 4.15.0
  • iptables: version 1.6.2
  • kernel-firmware: version 20180222_7344ec9
  • kmod: version 25
  • less: version 530
  • libXcursor: version 1.1.15
  • libXfont2: version 2.0.3
  • libXres: version 1.2.0
  • libdrm: version 2.4.90
  • libevdev: version 1.5.8
  • libfastjson: version 0.99.8
  • libgcrypt: version 1.8.2
  • libjpeg-turbo: version 1.5.3
  • libmnl: version 1.0.4 (rev2)
  • libnftnl: version 1.0.9
  • libpciaccess: version 0.14
  • librsvg: version 2.42.2
  • libseccomp: version 2.3.3
  • libtasn1: version 4.13
  • libunistring: version 0.9.8
  • libvirt-php: version 0.5.4 (rev2)
  • libvirt: version 4.0.0
  • libxml2: version 2.9.7
  • libxslt: version 1.1.32
  • logrotate: version 3.13.0
  • lvm2: version 2.02.176
  • lsof: version 4.90
  • lz4: version 1.8.1.2
  • mc: version 4.8.20
  • mesa: version 17.3.4
  • mpfr: version 4.0.1
  • nano: version 2.9.3
  • net-tools: version 20170208_479bb4a
  • nettle: version 3.4
  • network-scripts: version 15.0 (rev6)
  • ntp: version 4.2.8p10 (rev3)
  • openldap-client: version 2.4.45
  • pango: version 1.40.14
  • patch: version 2.7.6
  • pciutils: version 3.5.6
  • php: version 7.2.2
  • pkgtools: version 15.0 (rev5)
  • qemu: version 2.11.1
  • rpcbind: version 0.2.4 (rev3)
  • rsync: version 3.1.3 (CVE-2017-16548, CVE-2018-5764)
  • rsyslog: version 8.33.0
  • samba: version 4.7.5 (with time machine fullsync patch)
  • shadow: version 4.2.1 (rev3)
  • shared-mime-info: version 1.9
  • smartmontools: version 6.6
  • spice-protocol: version 0.12.13
  • spice: version 0.14.0
  • sqlite: version 3.22.0
  • sudo: version 1.8.22
  • sysvinit-scripts: version 2.1 (rev6)
  • talloc: version 2.1.11
  • tar: version 1.30
  • tdb: version 1.3.15
  • tevent: version 0.9.34
  • traceroute: version 2.1.0
  • usbutils: version 009
  • util-linux: version 2.31.1
  • wget: version 1.19.4 (rev2)
  • xf86-video-vesa: version 2.4.0
  • xfsprogs: version 4.15.0
  • xkeyboard-config: version 2.22
  • xorg-server: version 1.19.6
  • xterm: version 331

Linux kernel:

  • version 4.14.23 (with vmwrite error patch)

Management:

  • disable NTP by default
  • hide php warnings for now since php 7.2 is more strict
  • update smartmontools drivedb and hwdata/{pci.ids,usb.ids,oui.txt,manuf.txt}
  • webgui: noVNC: version 1.0.0
  • webgui: dashboard cpu bar updates using Safari browser (now uses SSE instead of Websockets)
  • webgui: Add ability for pluginMan to not install the .plg file
  • webgui: Enhancements to Docker custom network settings
  • webgui: enhancement to diagnostics: Don't anonymize system share names (such as appdata,domains,isos,system)
  • webgui: Fixed container update hanging when network doesn't exist
  • webgui: network refinements
  • webgui: docker: support special characters in webGUI, Support and Project Context menus
  • webgui: Add controls to enable/disable and specify custom ports for TELNET and SSH
  • webgui: Permit up to 4 NTP servers to be configured.
  • webgui: Add "Downgrade to previous version" selection
  • webgui: Add "Start Page" selection

Share this post


Link to post

Err o_0 

 

The docker page is... Interesting?

image.thumb.png.8e7c226af32ee3953c98d8eff213879c.png

 

I'm not loving the full-expanded all the time look :P 

 

It's rather messy IMO, nice to see the options are there but I'd certainly like to see that minimise button make a reappearance ;) 

Share this post


Link to post
4 hours ago, limetech said:
  • webgui: Improved docker used ports and IPs list
  • webgui: Enhancements to Docker settings and overview

Very nice.  Thank you.

Share this post


Link to post

upgraded server with rcu issue. will let you know if it returns. thanks!

Share this post


Link to post
3 hours ago, limetech said:

 

Credit goes to @bonienl

 

More the come... did some pretty nifty things in the last PR :)

  • Like 1

Share this post


Link to post

@limetech
Would you consider adding a hook for a custom script run after Unraid upgrade? I've been running under ESXi for ages and each time I have to manually copy bz* files and such to my boot vmdk.
If it's easy to do it on your side...why not. Maybe it could have other uses as well.
Thank you.

Sent from my SM-G955U1 using Tapatalk

Share this post


Link to post
17 minutes ago, AndroidCat said:

@limetech
Would you consider adding a hook for a custom script run after Unraid upgrade? I've been running under ESXi for ages and each time I have to manually copy bz* files and such to my boot vmdk.
If it's easy to do it on your side...why not. Maybe it could have other uses as well.
Thank you.

Sent from my SM-G955U1 using Tapatalk
 

 

You mean in the unRAIDServer.plg file?  Maybe provide an example of what you want to see.

Share this post


Link to post
 
You mean in the unRAIDServer.plg file?  Maybe provide an example of what you want to see.
After upgrade process is completed (upgrade button pressed and all files got downloaded and copied to flash) I'd like to execute simple command, like cp /boot/xyz /disks/vmdk/ prior to reboot.
I have vmdk mounted by unassigned devices, but that script could as well mount it as a first step.

Sent from my SM-G955U1 using Tapatalk

Share this post


Link to post
2 hours ago, AndroidCat said:

After upgrade process is completed (upgrade button pressed and all files got downloaded and copied to flash) I'd like to execute simple command, like cp /boot/xyz /disks/vmdk/ prior to reboot.
I have vmdk mounted by unassigned devices, but that script could as well mount it as a first step.

Sent from my SM-G955U1 using Tapatalk
 

 

Ok I will be totally honest: If you want something like this added, please show the actual code changes you want to see.  This is because it does not make sense for me to spend the time figuring out exactly what you want, then testing it, then tweaking it, etc., when it benefits <1% of the user base, when I could be spending the time on something that would be of benefit to >99% of the user base.

  • Like 1

Share this post


Link to post
2 hours ago, AndroidCat said:

After upgrade process is completed (upgrade button pressed and all files got downloaded and copied to flash) I'd like to execute simple command, like cp /boot/xyz /disks/vmdk/ prior to reboot.
I have vmdk mounted by unassigned devices, but that script could as well mount it as a first step.

Sent from my SM-G955U1 using Tapatalk
 

why not to use PlopKExec method to boot from USB? it's very fast now. i have two unRAID VMs on top of Esxi - one boots from vmdk, one from USB via PlopKExec - both VMs boot times are very similar..

Share this post


Link to post
 
Ok I will be totally honest: If you want something like this added, please show the actual code changes you want to see.  This is because it does not make sense for me to spend the time figuring out exactly what you want, then testing it, then tweaking it, etc., when it benefits 99% of the user base.
When I used to do something like this I used to use the user scripts plugin to execute a script on shutdown.

Sent from my LG-H815 using Tapatalk

Share this post


Link to post
15 hours ago, nexusmaniac said:

I'm not loving the full-expanded all the time look :P 

 

It's rather messy IMO, nice to see the options are there but I'd certainly like to see that minimise button make a reappearance ;) 

I agree - the port and volume mappings need to be collapsible (I think they used to be) as the page gets very long if you have a lot of dockers....

 

Share this post


Link to post
15 hours ago, nexusmaniac said:

I'm not loving the full-expanded all the time look :P 

 

It's rather messy IMO, nice to see the options are there but I'd certainly like to see that minimise button make a reappearance

I don't see it as messy at all.  I rather like it because I can see all the IP addresses and ports.

Share this post


Link to post

Possible to have something to display ranges?

 

Rather than 

192.168.1.10:20 -> 192.168.1.10:20

192.168.1.10:21 -> 192.168.1.10:21

192.168.1.10:22 -> 192.168.1.10:22

192.168.1.10:23 -> 192.168.1.10:23

192.168.1.10:24 -> 192.168.1.10:24

 

We'd have:

192.168.1.10:20-24 -> 192.168.1.10:20-24

Share this post


Link to post
17 hours ago, nexusmaniac said:

I'm not loving the full-expanded all the time look

 

image.thumb.png.e73313d32a97537b2bb5f04501108f18.png

 

Added more controls...

  • Like 5

Share this post


Link to post

I was just about to update to RC5 from RC3, but the web-gui is unresponsive, i managed to get the startpage to load but nothing beyond that. Can't access dockers either.

 

I can SSH into the server and the CPU is more or less idle, but i found this in the syslog with a timestamp about 1,5h ago, does anyone know what it means?

 

I can see that there was an patch included in RC5 for scsi rcu, perhaps that fixes this. 

But how do i upgrade to RC5 from terminal?

Mar  6 17:00:37 mandarin kernel: ------------[ cut here ]------------
Mar  6 17:00:37 mandarin kernel: WARNING: CPU: 2 PID: 0 at kernel/rcu/tree.c:2725 rcu_process_callbacks+0x320/0x36b
Mar  6 17:00:37 mandarin kernel: Modules linked in: tun xt_CHECKSUM iptable_mangle ipt_REJECT nf_reject_ipv4 ebtable_filter ebtables ip6table_filter ip6_tables xt_nat veth ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 iptable_filter ip_tables nf_nat xfs nfsd lockd grace sunrpc md_mod tg3 ptp pps_core ipmi_ssif i2c_core x86_pkg_temp_thermal intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel pcbc aesni_intel aes_x86_64 crypto_simd glue_helper cryptd intel_cstate intel_uncore intel_rapl_perf ahci libahci pcc_cpufreq acpi_power_meter thermal button ipmi_si ie31200_edac [last unloaded: kvm]
Mar  6 17:00:37 mandarin kernel: CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.14.23-unRAID #1
Mar  6 17:00:37 mandarin kernel: Hardware name: HP ProLiant MicroServer Gen8, BIOS J06 11/02/2015
Mar  6 17:00:37 mandarin kernel: task: ffff88020171aa00 task.stack: ffffc90000cbc000
Mar  6 17:00:37 mandarin kernel: RIP: 0010:rcu_process_callbacks+0x320/0x36b
Mar  6 17:00:37 mandarin kernel: RSP: 0018:ffff880202283f18 EFLAGS: 00010002
Mar  6 17:00:37 mandarin kernel: RAX: ffffffffffffd800 RBX: ffff8802022a14c0 RCX: 0000000100200001
Mar  6 17:00:37 mandarin kernel: RDX: 0000000000000003 RSI: ffff880202283f20 RDI: ffff8802022a14f8
Mar  6 17:00:37 mandarin kernel: RBP: ffffffff81c399c0 R08: 0000000000000001 R09: ffffffff814bf800
Mar  6 17:00:37 mandarin kernel: R10: ffffea00071bcd00 R11: ffff8802022a0900 R12: ffff8802022a14f8
Mar  6 17:00:37 mandarin kernel: R13: 7fffffffffffffff R14: 0000000000000246 R15: ffffffffffffffff
Mar  6 17:00:37 mandarin kernel: FS:  0000000000000000(0000) GS:ffff880202280000(0000) knlGS:0000000000000000
Mar  6 17:00:37 mandarin kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Mar  6 17:00:37 mandarin kernel: CR2: 00001493494cc000 CR3: 0000000001c0a005 CR4: 00000000000606e0
Mar  6 17:00:37 mandarin kernel: Call Trace:
Mar  6 17:00:37 mandarin kernel: <IRQ>
Mar  6 17:00:37 mandarin kernel: ? rebalance_domains+0x1fd/0x22f
Mar  6 17:00:37 mandarin kernel: __do_softirq+0xcd/0x1c2
Mar  6 17:00:37 mandarin kernel: irq_exit+0x4f/0x8e
Mar  6 17:00:37 mandarin kernel: smp_apic_timer_interrupt+0x7a/0x85
Mar  6 17:00:37 mandarin kernel: apic_timer_interrupt+0x7d/0x90
Mar  6 17:00:37 mandarin kernel: </IRQ>
Mar  6 17:00:37 mandarin kernel: RIP: 0010:cpuidle_enter_state+0xe0/0x135
Mar  6 17:00:37 mandarin kernel: RSP: 0018:ffffc90000cbfef8 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff10
Mar  6 17:00:37 mandarin kernel: RAX: ffff8802022a0940 RBX: 0000000000000000 RCX: 000000000000001f
Mar  6 17:00:37 mandarin kernel: RDX: 0000efd89a8f63a4 RSI: 0000000000020140 RDI: 0000000000000000
Mar  6 17:00:37 mandarin kernel: RBP: ffffe8fffee88040 R08: 00020eb35d87d8b6 R09: 0000000000000048
Mar  6 17:00:37 mandarin kernel: R10: ffffc90000cbfed8 R11: 0000000000476118 R12: 0000000000000002
Mar  6 17:00:37 mandarin kernel: R13: 0000efd89a8f63a4 R14: ffffffff81c59198 R15: 0000efd89a8037e2
Mar  6 17:00:37 mandarin kernel: ? cpuidle_enter_state+0xbb/0x135
Mar  6 17:00:37 mandarin kernel: do_idle+0x11a/0x179
Mar  6 17:00:37 mandarin kernel: cpu_startup_entry+0x18/0x1a
Mar  6 17:00:37 mandarin kernel: secondary_startup_64+0xa5/0xb0
Mar  6 17:00:37 mandarin kernel: Code: a8 00 00 00 eb 13 48 2b 05 eb f3 ba 00 48 39 c2 7d 07 48 89 93 90 00 00 00 48 83 7b 38 00 0f 94 c1 48 85 d2 0f 94 c0 38 c1 74 02 <0f> 0b 4c 89 f7 57 9d 66 66 90 66 90 4c 89 e7 e8 1f 0e 00 00 84 
Mar  6 17:00:37 mandarin kernel: ---[ end trace c87d29027b138377 ]---

 

Edited by dannen

Share this post


Link to post
33 minutes ago, dannen said:

But how do i upgrade to RC5 from terminal?

 

Try booting in 'Safe Mode', which will prevent array from starting.  Then server should stay alive long enough to update the OS.

Share this post


Link to post
12 hours ago, limetech said:

 

Ok I will be totally honest: If you want something like this added, please show the actual code changes you want to see.  This is because it does not make sense for me to spend the time figuring out exactly what you want, then testing it, then tweaking it, etc., when it benefits <1% of the user base, when I could be spending the time on something that would be of benefit to >99% of the user base.

 

Let me try again :-)

 

Is there any way I can insert custom *.sh script which gets launched when Unraid upgrade is finished (as a last step in upgrade routine).

Thanks.

 

Share this post


Link to post
 
Let me try again :-)
 
Is there any way I can insert custom *.sh script which gets launched when Unraid upgrade is finished (as a last step in upgrade routine).
Thanks.
 
No, but as the following step of upgrade is a shutdown and reboot, you can trigger a script to run on shutdown, to copy the necessary files to the necessary location, if the files aren't there because no upgrade has happened then nothing will happen.

You can use the user scripts plugin to do this and it's an approach I used to use to copy bzroot and bzimage to my virtual disk from the USB when I ran an Unraid VM inside Unraid on KVM.

Sent from my LG-H815 using Tapatalk

  • Like 1

Share this post


Link to post
Guest
This topic is now closed to further replies.