tuxbass

Were any errors reported during previous days' backup runs?

What's the _unraid way_ of turning on key-only login access to unraid host? Can see there are some set of keys in /boot/config/ssh but there's no sshd_config found under /boot to enforce their usage. Why are there keys under boot/config/ssh to begin with if they're not used? Just to be clear, the goal is to secure ssh access so no bad actor on LAN can wreak havoc.

Believe that's by design. Eg try passing (now-deprecated, I know) --link param when _not_ using self-created network.

Interesting. How come tarring doesn't have the same issues - isn't it all simply a read operation on source endpoints anyway? Or the problem is caused by write? Anyhow cheers, will enable v1 again.

I know it's going to be an unpopular request, but could we have "Use tar?" option before "Use Comporession?", latter being disabled & defaulting to 'no' if user opts out of tarring. There are use-cases when simply copying the files as-is can be preferred.

When the plugin starts the containers back up - are the order & wait times still respected? Every morning one of the services has gone bad, likely due to a dependent container not having started by the expected time. No issue with system reboot though.

>Did you ever reboot? Yes, before reboot absolutely no shares (not only appdata) were accessible. Delayed rebooting to make sure all the data-recovery is done prior, this was to ensure no writes to cache drive were performed beforehand. Can't believe how lucky I am unraid saved me from myself. At least gained some experience recovering data from xfs 🤷‍♂️

Finally mounted the cache partition after long recovery process, and... appdata contents are all there! How on earth is that possible? Looking back at the entered commands above I can't see how it wasn't overwritten. Don't get me wrong, I'm over the moon about it, but can't make any sense about what actually happened. Is it because the mv operation targeted /mnt/user, not /mnt/cache?

Does v2 still back up VM xmls from /etc/libvirt/qemu? v1 used to back them into /mnt/user/appdata/Community_Applications_VM_XML_Backup/, but v2 only seems to back up libvirt.img.

Just noticed the plugin is no longer listed under Plugins, and when I navigate to Plugins -> 'Install PluginPlugin File Install Errors' tab, then following is listed: /boot/config/plugins-error/preclear.disk.plg -> status ERROR /boot/config/plugins-error/preclear.disk.plg contents are: <?xml version="1.0" standalone="yes"?> <!DOCTYPE PLUGIN [ <!ENTITY name "preclear.disk"> <!ENTITY repo "unRAID-plugins"> <!ENTITY author "gfjardim"> <!ENTITY version "2020.05.07"> <!ENTITY launch "Tools/Preclear"> <!ENTITY gitURL "https://raw.githubusercontent.com/&author;/&repo;/master"> <!ENTITY pluginURL "&gitURL;/plugins/&name;.plg"> ]> <PLUGIN name="&name;" author="&author;" launch="&launch;" version="&version;" pluginURL="&pluginURL;" icon="icon-preclear" support="http://lime-technology.com/forum/index.php?topic=39985.0"> <CHANGES> ##&name; ###2020.05.07 - Fix: error showing in the logs for no reason - Fix: update tmux to 3.1b - Fix: better CSRF error handling ###2020.05.05a - Add: dialog package - Add: <preclear> command line utility to start/stop/watch preclear sessions ###2020.03.14a - Fix: wrong CSRF token will open an alert instead of trying to update the token in the background ###2020.03.14 - Fix: fix timer name so Unraid can stop/restart webgui updates - Fix: script - version 1.0.14 - Fix: script - pre-read will ignore read errors so bad sectors can be detected - Fix: script - write resume info to flash drive timelapse increased from 60 to 120 seconds ###2020.03.06a - Fix: change the use of system original tail_log to custom tail_log - Fix: alert user if an action is necessary - Fix: change the disposition of alert buttons - Fix: script - version bump to 1.0.13 - Fix: script - change the way elapsed time is computed - Fix: script - better pause/unpause event handling ###2020.03.02a - Fix: apply the last fix to the original preclear script and bump version to 1.20 ###2020.03.02 - Fix: drives > 2.2TiB and with protective MBR partition starting on sector 64 failing signature verification. ###2020.03.01 - Fix: update swal2 css ###2020.02.27a - Fix: minor changes ###2020.02.26 - Fix: error while output smart upon script error - Fix: migrate to SweetAlert v2 - Fix: include Joe L. script by default ###2020.02.17a - Fix: better diskinfo update handling - Fix: resume verification will always read the MBR ###2020.02.17 - Fix: reload UD page after Preclear operations ###2020.02.14 - Fix: Write/Erase ops hang if display_pid is reused by the system ###2020.02.13a - Fix: minor bugfixes ###2020.02.13 - Fix: better queue handling - Fix: css fixes to grey/azure/black themes ###2020.02.12 - Fix: Increase time lapse before kill a hunged dd process ###2020.01.17b - Add: Fix Preclear button ###2020.01.17a - Fix: division by zero on speed calculation ###2020.01.16a - Fix: disable browser popup open on Start/Resume preclear ###2020.01.16 - Fix: plugin tries to gently stop the preclear script - Fix: addBannerWarning retrocompatibility - Add: sorting on Tools > Preclear for Preclear Queue - Add: Pause/Stop/Resume all - Fix: script: version 1.0.8 - Fix: script: resume not working while verifying the disk - Fix: script: correctly compute current speed even if the session is paused/enqueued - Fix: script: increase ui refresh rate - Fix: script: better exit handling ###2020.01.14 - Fix: only mark as unsupported if tmux isn't work properly ###2020.01.13 - Fix: prevent mounted disks from appearing at "Start Multiple Sessions" - Fix: prevent mounted or preclearing disks from being precleared by accident ###2020.01.12 - Fix: properly handle unsupported state ###2020.01.11b - Fix: add Unraid 6.8.1 support ###2020.01.11 - Add: start multiple preclear sessions ###2020.01.10 - Fix: Unraid 6.8.0 package update ###2019.11.28 - Fix: script version 1.0.6 - Fix: proper debug messages - Fix: proper exit handling ###2019.11.26 - Fix: increase verbosity in preclear script ###2019.11.22 - Fix: script version 1.0.5 - Fix: 4kn disk drives not formating - Fix: exporting errors and messages to syslog - Fix: starting all partitions at sector 64 (avoid partition missalignment) - Fix: open reports at a new tab ###2019.04.23a - Fix: Change image icon to font icon if possible ###019.04.23 - Fix: reinstallation failing; - Fix: change between icon/image if needed. ###2019.04.22a - Fix: set correct time interval for resume write to flash drive. ###2019.04.22 - Fix: remove ncurses dependency; - Fix: save resume information to flash drive every 15 minutes to avoid NAND memory wear. ###2019.03.10 - Fix: diskinfo causing high cpu usage. - Fix: start/stop queue on demand. - Fix: remove leftover packages ###2019.02.12b - Fix: queue persistent across reboots - Fix: better enqueue handling (script 1.0.3) - Fix: exiting on bad resume file ###2019.02.12 - Add: preclear queue - Add: download log from each preclear session ###2019.02.11 - Add: preclear log icon to status - Add: real-time syslog monitoring ###2019.02.10 - Fix: bad substitutionon line 590 ###2019.02.06b - Add: present serial number on notifications ###2019.02.06a - Fix: footer not appearing ###2019.02.06 - Fix: add unRAID 6.7 support - Fix: decrease backgroung information pooling - Fix: script 1.0.1 - commit all resume information to flash drive prior to create the resume file ###2018.10.07 - Fix: remove glyphicon fount dependency - Fix: vshift in title banners on 6.6.1 - Fix: install full ncurse package depending on stripped version installed (no GCC version dependant) - Fix: bump preclear script to 1.0.0 (first main version) ###2018.09.20 - Fix: Unraid 6.6 - black theme dropdown ui displaying white-on-white text ###2018.09.17 - Add: resume preclear button ###2018.09.16 - Fix: changed unnamed pipes to named pipes in verify and erase operations (more control over dd process pid and exit status) - Fix: pre-read now starts at the beggining of the disk - Add: more verbosity to the log ( progress on write/read operations) - Add: disk temperature to the page footer ###2018.08.27 - Fix: change 'direct' iflag option to 'nocache' - apparently some hdds don't like it - Fix: only export initial smart status if not resuming ###2018.08.23 - Fix: not toggling between preclear scripts ###2018.08.22 - Fix: make sure blkid file is created before retrieving dd pid - FIx: export syslog messages to preclear disk log ###2018.08.21c - Add: SMART info to preclear log ###2018.08.21b - Fix: remove fsync option from read/write disk ###2018.08.21a - Fix: avoid custom version of Tooltipster to load if the official version is present - Fix: detect wrong csrf_token and update it - Add: disk log to Tools > Preclear Disk ###2018.08.21 - Fix: diskinfo PHP warnings - Fix: revert direct I/O usage in dd invocation (reduced write performance) ###2018.08.20 - Fix: increase verbosity of script. ###2018.05.03 - Fix: script not completely zeroing the disk ###2018.04.27 - Fix: script deem failed small dd reads/writes ###018.04.24 - Fix: rc.diskinfo - PHP Warning: Missing argument 2 for force_reload() - Fix: script not detecting array disks correctly - Fix: correctly detect dd exit code - may fix link problems hanging preclear operation - Fix: pause dd if s6-sync (part of s6-utils used by linuxserver) is invoked ###2018.03.29 - Fix: early unRAID 6.5 compatibility ###2017.11.14 - Fix: wrong TMUX being installed on v6.3.5 ###2017.11.01 - Fix: remove strings dependency ###2017.09.27 - Fix: diskinfo not starting properly ###2017.09.25 - Fix: install/update not showing the Done button ###2017.09.23 - Fix: diskinfo - PHP Warning: strpos(): Empty needle in /etc/rc.d/rc.diskinfo on line 339 ###2017.09.22a - Fix: update TMUX and add support to 6.4-rc9f ###2017.09.22 - Fix: resume not working due to a bug introduced in 2017.09.20 ###2017.09.20 - Fix: better time handling across resumes ###2017.09.19 - Fix: convert popups to Sweet Alert - Fix: added more verbosity in the preclear script ###2017.09.17a - Fix: resume not working properly with pre-read ###2017.09.17 - Fix: persistent elapsed time on resumed sessions ###2017.09.16 - Fix: detect dd hang and restart it - Add: resume stopped preclear session ###2017.07.10 - Add Cron keepalive ###2017.07.08 - Fix: dismiss Statistics status saved at temporary dir; - Fix: diskinfo 0.1.0 - better array disk detection; ###2017.07.05c - Fix: diskinfo 0.0.9 - not showing SSD temperature ###2017.07.05b - Fix: diskinfo 0.0.8 - not properly detect the absent of cached info ###2017.07.05 - Fix: add more files to preclear log download ###2017.07.04 - Fix: diskinfo 0.0.7 - minor bugfixes ###2017.06.30b - Fix: NVME support? ###2017.06.30a - Fix: README typo. ###2017.06.30 - Fix: diskinfo 0.0.6 - reload unRAID config every 30 secondsinstead of every 5 seconds; - Fix: Non-static method should not be called statically (PHP 5.4 error) ###2017.06.27 - Fix: diskinfo 0.0.5 - add compatibility to future Unassigned Devices adoption ###2017.06.24 - Add: present preclear status on footer ###2017.06.23 - Add: diskinfo 0.0.4 - refreshes disk info based on UDEV events - Add: diskinfo 0.0.4 - detect precleared disks - Add: preclear_disk 0.8.8 - pause on hdparm/smartctl/sync processes running - Add: packages to unRAID 6.4 ###2017.06.21 - Fix: diskinfo 0.0.3 - suport restart - Fix: diskinfo daemon restart after update ###2017.06.20 - Fix: tooltips not properly appearing; - Fix: temperature not showing properly; - Fix: diskinfo 0.0.2 - timeouts not properly working; - Fix: script 0.8.7 - pause write every 2% to let SMART queries to complete; ###2017.06.15 - Fix: small bug in diskinfo daemon ###2017.06.14 - Fix: gather information in the background avoiding webgui to freeze - Fix: script version 0.8.6 - add timeout to 'smartctl' commands - Fix: Azure theme compatibility ###2017.03.23 - Fix: mismatch short serial numbers - bug introduced in the last version - Fix: preclear not stopping on Windows browsers ###2017.03.21 - Fix: Plugin not working in older unRAID versions ###2017.02.16 - Fix: pause not working ###2017.02.15 - Fix: more serial fixes ###2017.02.14 - Fix: change identifier from device to serial number ###2017.02.05 - Add: Erase and Preclear function ###2017.02.01 - Add: tooltipster - Fix: replace buttons to icons - Fix: disk size wrongly reported - Fix: log not being reported ###2017.01.27 - Fix: comply with unRAID's XSS and CSRF hardening ###2016.12.24 - Add: increase verbosity of script ###2016.12.23a - Fix: zero/erase failing due to false exit code ###2016.12.23 - Fix: better log dd errors ###2016.12.21 - Add: Secure Erase disk ###2016.12.20a - Fix: remove send statistics ###2016.12.20 - Fix: recompile TOR package, avoiding logrotate error - Add: TOR SlackBuild sources to GitHub ###2016.12.19 - Add: send statistics; - Add: tor package; - Fix: removed array disk appearing as unavailable disk; - Fix: install correct packages on unRAID 6.3 - Fix: script - better error reporting; ###2016.10.31 - Add: generate some debug info ###2016.10.29 - Fix: small bug in script display - Fix: Array disks connected to Areca cards showing up - Fix: Detect if script author is Joe L. or bjp999 - Add: script write errors to log file - Add: Download log file from the webui ###2016.10.26 - Add: Copy to Clipboard button on Preview window - Fix: unescaped quotes breaking read of "state.ini" file - Fix: remove PHP "split" function due deprecation - Fix: script - better SMART detection - Fix: script - greatly improve speed and memory footprint in read operations - Fix: script - faster display refresh ###2016.09.27 - Fix: better PHP exception handling. ###2016.09.26b - Fix: not loading information if /boot/preclear_reports directory doesn't exist ###2016.09.26a - Fix: prepare better Unassigned Devices integration ###2016.09.25 - Fix: utempter depencency not installed - Add: "Answer Yes" button on Joe L. script if needed. - Add: preclear command to show status on shell ###2016.09.23a - Fix: script error appearing if SMART is not available ###2016.09.23 - Add: show reports of listed disks; - Fix: page scrolling while updating; - Fix: make link point to Tools; - Fix: increase disk column width; - Fix: better script report. ###2016.09.11 - Fix: better report formating ###016.09.10 - Fix: preclear report not being created due to filename invalid characters. ###2016.08.29 - Fix: slow read with custom settings ###2016.08.28 - Fix: create temp folder on plugin installation ###2016.08.26 - Fix: merge beta script - Fix: high memory consumption ###2016.06.23 - Fix: install utempter ###2016.06.22 - Add: Notification capability - Fix: Move Preclear icon to Tools - Fix: Fix unknown terminal "screen" error - Fix: Preclear script not reading the end of the disk - Update: tmux (2.1), libevent (2.0.22) and ncurses (5.9-4) ###2016.03.24a - Fix: better smartctl device testing ###2016.03.24 - Fix: displaying negative numbers ###2016.03.22 - Fix: Better S.M.A.R.T. report - Add: Save a report in /boot/preclear_reports ###2016.03.21 - Add: pause any preclear operations while array start/stop - Add: Initial SMART reporting ###2016.03.20 - Fix: disk detected as part of the array if cache slot was empty ###2016.03.11b - Add: New Preclear script - Forked: to beta ###2015.11.18 - Add: logging capabilities - Fix: php error if disk.cfg file doesn't exist ###2015.09.28a - Fix: DVD/BluRay drives being included - Fix: some minor bugs ###2015.09.18 - Fix: Areca disks not appearing correctly. ###2015.09.17b - Fix: black CSS conformity ###2015.09.17 - Release: maintenance release. ###2015.09.08 - Add: ncurses package, avoiding the use of the xterm hack. ###2015.09.07d - Improvement: better integration with Unassigned Devices. - Fix: unRAID 6.1.1 doesn't have a 'screen' terminal descriptor; xterm used instead. ###2015.08.12 - Fix: 6.1-rc3 compatibility ###2015.07.23 - Fix: better unRAID 6.1 compatibility - Fix: clearer notification system ###2015.07.21 - Fix: unRAID 6.1-rc1 compatibility ###2015.06.16 - Fix: duplicate hdd entry due to udev's duplicate links ###2015.06.15 - Fix: minor bugfixes ###2015.06.13 - Fix: remove ^n markup from preclear status - Fix: show a wait message while preclear is starting - Fix: add MD5 verification to downloaded files - Fix: remove the Start Preclear link if an tmux session is already open ###2015.05.20b - Minor bugfixes ###2015.05.20a - Add support to bjp999 version ###2015.05.20 - Remove preclear script from package. It must be manually added to /boot/config/plugins/preclear.disk/ directory. ###2015.05.18b - Fix: empty cacheId variable crashing the code ###2015.05.18a - Fix: plugin not appearing under Settings/Utilities ###2015.05.18 - Add: fast preclear option (bjp999 mod) ###2015.05.17a - Improve: detection of SCSI disks ###2015.05.16a - Add: other preclear operations ###2015.05.16 - Initial release. </CHANGES> <!-- The 'plugin' package file. --> <FILE Name="/boot/config/plugins/&name;/&name;-&version;.txz"> <URL>&gitURL;/archive/&name;-&version;.txz</URL> </FILE> <!-- The 'plugin' package MD5 hash. --> <FILE Name="/boot/config/plugins/&name;/&name;-&version;.md5"> <URL>&gitURL;/archive/&name;-&version;.md5</URL> </FILE> <!--The 'readvz' executable.--> <FILE Name="/usr/local/sbin/readvz" Mode="755"> <URL>&gitURL;/source/packages/readvz</URL> </FILE> <!--The 'dialog' install package.--> <FILE Name="/boot/config/plugins/&name;/dialog-1.3_20200327-x86_64-1.txz" Mode="666"> <URL>&gitURL;/source/packages/dialog-1.3_20200327-x86_64-1.txz</URL> <MD5>f7de2c4a0984e8f0e32a2140845dc862</MD5> </FILE> <!--The 'preclear_disk_ori.sh' script.--> <FILE Name="/usr/local/sbin/preclear_disk_ori.sh" Mode="755"> <URL>&gitURL;/source/packages/preclear_disk_20.sh</URL> </FILE> <!-- The 'packages-install' script. --> <FILE Run="/bin/bash"> <INLINE> unsupported="/var/state/&name;/unsupported" # Creating some directories mkdir -p /tmp/.preclear/ /usr/local/bin/ /var/state/&name;/ download_install() { local dest="/boot/config/plugins/&name;/${2}" local src="&gitURL;/source/packages/${2}" local pkg=$(echo "${2}" | rev | cut -f 3- -d '-' | cut -d '-' -f2- | rev) local ver=$(echo "${2}" | rev | cut -f 3- -d '-' | cut -d '-' -f1 | rev) local md5=$1 if [ ! -f "${dest}" ]; then curl --location --silent --fail "${src}" --output "${dest}" fi file_md5=$(/usr/bin/md5sum ${dest}) if [ "${file_md5:0:32}" != "${md5:0:32}" ]; then echo "Wrong '${2}' package md5 hash." rm "${dest}" exit 1 fi while read p; do local _pkg=$(echo "$p" | rev | cut -f 3- -d '-' | cut -d '-' -f2- | rev) if [ "$pkg" == "$_pkg" ]; then local ipkg=$_pkg local iver=$(echo "$p" | rev | cut -f 3- -d '-' | cut -d '-' -f1 | rev) break fi done < <(find /var/lib/pkgtools/packages/ -type f -printf '%P\n') if [ -n "$ipkg" ]; then if php -r "exit((version_compare('$ver','$iver', '>=') ) ? 0 : 1);"; then echo "$pkg version $ver is greater or equal than the installed version ($iver), installing..." /sbin/upgradepkg --install-new "$dest" else echo "$pkg version $ver is lower than the installed version ($iver), aborting..." echo "$pkg version $ver is lower than the installed version ($iver), aborting..." | logger -t "root: plugin: &name;.plg" fi else echo "installing package $pkg version $ver...." echo "installing package $pkg version $ver...." | logger -t "root: plugin: &name;.plg" /sbin/upgradepkg --install-new "$dest" fi } is_unraid_version() { if [ -f "/etc/unraid-version" ]; then . /etc/unraid-version else version="0.0" fi return $( php -r "echo ( version_compare('$version','$2', '$1') ) ? 0 : 1;" ) } if [ -f "$unsupported" ]; then rm "$unsupported" fi if is_unraid_version '>=' '6.5.3' && is_unraid_version '<' '6.6.3'; then # download_install c04358fd7d7cb0b5ca67f215871f3692 ncurses-6.0-x86_64-3.txz download_install 9ba9d61dcd90d45719d27b9e7a111913 libevent-2.1.8-x86_64-1.txz download_install a48783401331620b7c7e5d7af58f53db tmux-2.5-x86_64-1.txz download_install aed1f39d4a5acddb225b542145e0eeaf utempter-1.1.6-x86_64-2.txz elif is_unraid_version '>=' '6.6.3' && is_unraid_version '<' '6.7'; then # download_install 96ebaee8f6a62f220dbdf4723a81e535 ncurses-6.1_20180324-x86_64-1.txz download_install 9ba9d61dcd90d45719d27b9e7a111913 libevent-2.1.8-x86_64-1.txz download_install a4a7fdd5ff9910bee3477a2aa0e5fa25 tmux-2.7-x86_64-1.txz download_install aed1f39d4a5acddb225b542145e0eeaf utempter-1.1.6-x86_64-2.txz elif is_unraid_version '>=' '6.7' && is_unraid_version '<' '6.8.0-rc1'; then # download_install 007c068e334c2358621d3d1ea959da80 ncurses-6.1_20190112-x86_64-1.txz download_install 4e76295aad8c9bf8c98301a9b7ec2e64 libevent-2.1.8-x86_64-3.txz download_install 00d6b305ad4f01504a9419dd959e8cce openssl10-solibs-1.0.2o-x86_64-1.txz download_install a40eef8aa01885ff6ea592a7e8e6b2da tmux-2.8-x86_64-1.txz download_install 1af347c4d9feff774320d9c1e7920ac3 utempter-1.1.6-x86_64-3.txz elif is_unraid_version '>=' '6.8.0-b1' && is_unraid_version '<' '6.8.3'; then download_install 7dabee590ed5b5e47b17480dec367636 tmux-3.0a-x86_64-1.txz download_install a6e7015c850fcfc5a56f42b740c0a4ba libevent-2.1.11-x86_64-1.txz download_install 1af347c4d9feff774320d9c1e7920ac3 utempter-1.1.6-x86_64-3.txz elif is_unraid_version '>=' '6.8.3'; then download_install c8367090c89d26690be706c7fb3db5fc tmux-3.1b-x86_64-1.txz download_install a6e7015c850fcfc5a56f42b740c0a4ba libevent-2.1.11-x86_64-1.txz download_install 1af347c4d9feff774320d9c1e7920ac3 utempter-1.1.6-x86_64-3.txz else echo "unRAID version ($version) not supported." echo "unRAID version ($version) not supported." | logger -t "root: plugin: &name;.plg" touch "$unsupported" fi # Fix tmux tmp directory if needed if [ ! -d "/tmp/tmux-0" ]; then mkdir /tmp/tmux-0 fi chmod 700 /tmp/tmux-0 chown -R root:root /tmp/tmux-0 # Verify if tmux is working correctly /usr/bin/tmux new-session -d -x 140 -y 200 -s "install_test_session" &>/dev/null sleep 0.5 /usr/bin/tmux kill-session -t "install_test_session" &>/dev/null if [ $? -ne 0 ]; then echo "tmux is not working properly" echo "tmux is not working properly" | logger -t "root: plugin: &name;.plg" touch "$unsupported" fi # Verify and install plugin package sum1=$(/usr/bin/md5sum /boot/config/plugins/&name;/&name;-&version;.txz) sum2=$(/usr/bin/cat /boot/config/plugins/&name;/&name;-&version;.md5) if [ "${sum1:0:32}" != "${sum2:0:32}" ]; then echo "Wrong 'plugin' package md5 hash." rm /boot/config/plugins/&name;/&name;-&version;.txz rm /boot/config/plugins/&name;/&name;-&version;.md5 exit 1 else for old in $(find /var/lib/pkgtools/packages/ -iname "preclear.disk-*"); do removepkg "$old" done upgradepkg --install-new /boot/config/plugins/&name;/&name;-&version;.txz fi </INLINE> </FILE> <!-- The 'post-install' script. --> <FILE Run="/bin/bash"> <INLINE> # Create a symlink to /root/mdcmd if [[ ! -e /root/mdcmd ]]; then ln -sf /usr/local/sbin/mdcmd /root/mdcmd fi #dismissed="/boot/config/plugins/statistics.sender/dismissed" #version_dismiss="/boot/config/plugins/&name;/statistics.sender.dismissed-&version;" #if [ ! -f "$version_dismiss" ]; then # rm -f "$dismissed" # touch "$version_dismiss" #fi # Make sure log file exists touch /var/log/preclear.disk.log # Set correct terminal type for TMUX echo "set -g default-terminal '$TERM'" > /root/.tmux.conf # Update file access mode chmod 644 $(awk '/\.(htm|php|js|page|css|png|plg|gif|cfg|md)$/ {print "/"$0}' /var/log/packages/&name;-&version;) chmod -R 755 /usr/local/emhttp/plugins/&name;/script/ ln -sf /usr/local/emhttp/plugins/&name;/script/preclear.sh /usr/local/bin/preclear # Install diskinfo service cp /usr/local/emhttp/plugins/&name;/rc.diskinfo /etc/rc.d/rc.diskinfo chmod +x /etc/rc.d/rc.diskinfo # Add CRON entry to keep alive cat <<'EOT' > /etc/cron.d/diskinfo # Diskinfo daemon keepalive */1 * * * * /etc/rc.d/rc.diskinfo --daemon &> /dev/null EOT chmod 600 /etc/cron.d/diskinfo # Add UDEV rule mv /usr/local/emhttp/plugins/&name;/99_diskinfo.rules /etc/udev/rules.d/99_diskinfo.rules chmod 644 /etc/udev/rules.d/99_diskinfo.rules udevadm control --reload-rules # Cleaning old source files find /boot/config/plugins/&name;/ -type f -iname "&name;*.txz" ! -iname "*&version;*" -delete find /boot/config/plugins/&name;/ -type f -iname "&name;*.md5" ! -iname "*&version;*" -delete find /boot/config/plugins/&name;/ -type f -iname "tor-*.txz" -delete # Start diskinfo /etc/rc.d/rc.diskinfo --quit &>/dev/null rm -f /var/local/emhttp/plugins/diskinfo/diskinfo.json &>/dev/null echo "sleep 5 ; /etc/rc.d/rc.diskinfo --daemon &>/dev/null" | at now &>/dev/null # Plugin version echo "&version;" > /var/state/preclear.disk/version echo "" echo "-----------------------------------------------------------" echo " &name; has been installed." echo " Copyright 2015-2020, gfjardim" echo " Version: &version;" echo "-----------------------------------------------------------" echo "" </INLINE> </FILE> <!-- Change image icon to font icon if possible --> <FILE Run="/usr/bin/at -M now 2>/dev/null "> <INLINE> #!/bin/bash # Change PNG icon to SVG icon if possible font_file="/usr/local/emhttp/webGui/styles/font-unraid.svg" if [ -f "$font_file" ]; then icon_exists=$(grep -c "preclear" "$font_file" 2>/dev/null) if [ "$icon_exists" -gt 0 ]; then sed -i -e 's/preclear.disk.png/icon-preclear/g' /usr/local/emhttp/plugins/&name;/Preclear.page sed -i -e '/^---/i Tag=\"icon-preclear\"' /usr/local/emhttp/plugins/&name;/Preclear.page elif [ "$icon_exists" -eq 0 ]; then sleep 5 sed -i -e 's/icon=\"icon-preclear\"//g' /boot/config/plugins/&name;.plg fi fi </INLINE> </FILE> <!-- The 'remove' script. --> <FILE Run="/bin/bash" Method="remove"> <INLINE> # kill all preclear sessions /usr/local/emhttp/plugins/&name;/script/clear_preclear.sh # Remove plugin removepkg &name;-&version; # Remove diskinfo rm /etc/cron.d/diskinfo sleep 1 /etc/rc.d/rc.diskinfo --quit rm /var/local/emhttp/plugins/diskinfo/diskinfo.json \ /etc/rc.d/rc.diskinfo \ /etc/udev/rules.d/99_diskinfo.rules udevadm control --reload-rules # Remove plugin related files rm -f /usr/local/bin/preclear \ /var/run/preclear_queue.pid \ /var/log/preclear.disk.log \ /usr/local/sbin/preclear_disk_ori.sh \ /usr/local/sbin/readvz # Remove plugin related directories rm -rf /boot/config/plugins/&name; \ /usr/local/emhttp/plugins/&name; \ /var/state/&name; \ /boot/.preclear \ /tmp/.preclear echo "" echo "-----------------------------------------------------------" echo " &name; has been uninstalled." echo " Please reboot your server to complete uninstall this plugin." echo " Copyright 2015-2020, gfjardim" echo " Version: &version;" echo "-----------------------------------------------------------" echo "" exit 0 </INLINE> </FILE> </PLUGIN> Any idea what's the deal?

Are you sure? eg `strings` is BSD version, not GNU.

Thanks for tcllib! Another request that I'm surprised is not already included: GNU coreutils https://slackware.pkgs.org/14.2/slackware-x86_64/coreutils-8.25-x86_64-2.txz.html

For context, it's commonly used by other tcl utilities, like xfs_undelete, used to recover removed files on XFS filesystem.

We have tcl package, but no option for tcllib, often accompanying tcl runtimes. Any possibility to get it added in? https://slackware.pkgs.org/14.2/slackonly-x86_64/tcllib-1.20-x86_64-1_slonly.txz.html

Cool, will do that. I'll see what if any i manage to recover before rebooting; still working on that.

Unraid is working. But even /mnt/user/ is inaccessible: ┌─[Tower]─[/mnt] └──╼ ls /mnt/user ls: cannot access '/mnt/user': Transport endpoint is not connected

Given appdata is gone, what would be the process to get unraid operational again?

I screwed up majorly. ironically while setting up CommunityBacupData; previous appdata backup is sadly years old. in /mnt/user/CommunityApplicationsAppdataBackup i created another appdata dir to move contents of CommunityApplicationsAppdataBackup/ to. But the move operation rewrote the appdata: mv appdata/ .. So empty directory /mnt/user/CommunityApplicationsAppdataBackup was moved to /mnt/user/ After this (note realizing what I had done) I moved the contents of appdataBackup previous backups into the new directory: ┌─[Tower]─[/mnt/user/CommunityApplicationsAppdataBackup] └──╼ mv * ../appdata mv: failed to access '../appdata': Transport endpoint is not connected Unsure if this process completed or not. What i do know is listing the contents of CommunityApplicationsAppdataBackup/ definitely still worked prior this last move. Array was stopped afterwards. Any chance of saving anything from the drive from this state? appdata was set to cache Only (or maybe Prefer) sadly. Also listing all shared fails: ls -lt /mnt/user ls: cannot access '/mnt/user': Transport endpoint is not connected Filesystem on cache drive is xfs. Edit: full recap of the commands performed: ┌─[Tower]─[~] └──╼ cd /mnt/user/CommunityApplicationsAppdataBackup ┌─[Tower]─[/mnt/user/CommunityApplicationsAppdataBackup] └──╼ mkdir appdata ┌─[Tower]─[/mnt/user/CommunityApplicationsAppdataBackup] └──╼ ls [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] appdata ┌─[Tower]─[/mnt/user/CommunityApplicationsAppdataBackup] └──╼ mv appdata/ .. ┌─[Tower]─[/mnt/user/CommunityApplicationsAppdataBackup] └──╼ mv * ../appdata mv: failed to access '../appdata': Transport endpoint is not connected ^ exit code of this was non-zero

Same here - just one of the containers is reporting not avail.

Cool, thanks. Will give that unit a go 👍

Am shopping for a UPS for a ~300W build. Mostly the server idles around 55-60W, and goes high-200W during gaming in VM (seen around 280W); spikes to ~120W when array is started at idle. This UPS would have to serve this build, and also the modem/router ISP has given. Haven't measured, but guess it averages around 10W. Is this APC BX1400U-GR a reasonable choice? 700W might be bit of an overkill, but offers chance for heavier workload in the future. Note my aim is to allow at least 2min of runtime prior to initiating the shutdown. At 300W load the manufacturer estimates 14min of runtime, which gives plenty of buffer as well, so if server gets shut down within 5min, the battery wouldn't get extensively depleted.

[solution below] Started using unraid again after couple of years. Updated the OS and windows 10 VM, but bluetooth dongle no longer works. It's detected by VM, but it connects to couple of minutes max, after which device is no longer functioning (might state "Connected" in win bluetooth settings though). Have tried 3 different devices: - onboard BT module (Asrock's Fatal1ty Z170 mobo), think it's BCM20702A0 chipset; - plugable USB dongle (should be same chipset as above) - CSR dongle (unknown chipset, but this is the one that used to work before) Have tried all of these 3 devices, but best I can do is connect the keyboard for 1-2 minutes, after which it no longer works. It never pairs again. Any pointers? Things worked relatively well around v6.4. Of course windows was also at lower ver back in those days, hard to tell where the issue is here really. If these products/chipsets are no-go, can you recommend one that should be compatible with latest unraid/win10 vm? Edit: tried old backed up VM images, and those fail to work as well. Maybe it's something that newer unraid versions are interfering with? Edit2: appears to be similar to this: One thing I've noticed - if I keep the "add new device" dialog open, then the device is kept connected. As soon as the dialog is closed and you stop typing on BT keyboard for ~3sec, the pairing is lost. Looks like it requires active communication over BT to keep it running. Running unraid v 6.8.3 SOLUTION The problem wasn't unraid upgrade, but windows upgrade for the VM. To get this sudden disconnection issue resolved, you need to open Device Manager, open the dialog for your bluetooth device, select 'Power Management' tab, and untick the 'allow this device to be powered off...' box. Problem for me was that the Win10 build I installed (build 2004) didn't have that checkbox available for my bluetooth device anymore. So the solution is to either a) continue using older windows builds that still have said tickbox; or b) get a bluetooth USB dongle and connect it into USB controller that's passed through. Note for b) you might also want to go through the listed 'USB device' & USB controllers in Device Manager, and uncheck the same power management feature.

Cheers. First restore (without the -i flag) attempt: root@Tower:/dev# btrfs restore -v /dev/md2 /mnt/disk1/disk2-restore/ No mapping for 847315304448-847315320832 Couldn't map the block 847315304448 Couldn't map the block 847315304448 bad tree block 847315304448, bytenr mismatch, want=847315304448, have=0 Couldn't read tree root Could not open root, trying backup super No mapping for 847315304448-847315320832 Couldn't map the block 847315304448 Couldn't map the block 847315304448 bad tree block 847315304448, bytenr mismatch, want=847315304448, have=0 Couldn't read tree root Could not open root, trying backup super No mapping for 847315304448-847315320832 Couldn't map the block 847315304448 Couldn't map the block 847315304448 bad tree block 847315304448, bytenr mismatch, want=847315304448, have=0 Couldn't read tree root Could not open root, trying backup super ...followed by including -i flag: root@Tower:/mnt/disk1# btrfs restore -vi /dev/md2 /mnt/disk1/disk2-restore/ No mapping for 847315304448-847315320832 Couldn't map the block 847315304448 Couldn't map the block 847315304448 bad tree block 847315304448, bytenr mismatch, want=847315304448, have=0 Couldn't read tree root Could not open root, trying backup super No mapping for 847315304448-847315320832 Couldn't map the block 847315304448 Couldn't map the block 847315304448 bad tree block 847315304448, bytenr mismatch, want=847315304448, have=0 Couldn't read tree root Could not open root, trying backup super No mapping for 847315304448-847315320832 Couldn't map the block 847315304448 Couldn't map the block 847315304448 bad tree block 847315304448, bytenr mismatch, want=847315304448, have=0 Couldn't read tree root Could not open root, trying backup super Guess it's time to throw parity drive away and try to restore drive 2 contents from backups, as close as possible?

Did you mean trying to mount the rebuilt data-drive? If so, then no dice: root@Tower:/tmp# mount -o usebackuproot,ro /dev/sde1 /tmp/x mount: /tmp/x: wrong fs type, bad option, bad superblock on /dev/sde1, missing codepage or helper program, or other error. If you meant parity drive (ie the one w/ read errors), then i'm not sure how that could even work, as it contains the xor'd bits, not files per se. Given the drive doesn't mount, it's likely safe to say the rebuild failed right? Any idea where that would be stated? No notification showed it, and can't find statement about it in logs either.

The filesystem is in btrfs, any alternative for that?