danringer

Members
  • Posts

    8
  • Joined

  • Last visited

Everything posted by danringer

  1. Can I pin docker to specific cores? or only specific containers? can I pin several containers to the same core? If I isolate cores does that prevent unraid from locking up if docker gets carried away? Can we determine from the diags what's going wrong that is forcing docker to run up the system load? Thank you, -Dan
  2. hello all, I've been using Docker with 25-ish apps for over a year now. Recently, I've been experiencing lockups beginning with load increasing over 200 at which time I can only run "uptime". Shutdown is unresponsive to change init levels and only a hard reboot is able to recover the system. Parity checks have been successful after reboots. I began with slowly restarting docker apps one at a time and I cannot trace the problematic container. I did a full delete and restore of the docker.img and that worked for a little over a week then started acting up again. Same thing, load over 300, cold boot, rebuilt docker image and this time it lasted an hour before creeping up over 80. docker stop $(docker ps -a -q) allowed the system to recover and I was able to get a diagnostic zip (attached). I want to attempt to pin the apps to specific cpus but I've read that isn't reliable with docker. Any help would be appreciated. thank you, -Dan r720-diagnostics-20231125-1054.zip
  3. this just worked for me. Plugins>CA Mover Tuning>Move Now button follows plug-in filters: Yes
  4. I requested a method to request additions in the 7th reply to this topic. On the 3rd page, 6 months ago I requested: mosh and sg3utils Is there a better way to request additions? -Dan
  5. Is there any place to see what is being worked on and/or what has already been requested?
  6. I'm happy to see this under active development. May I request: sg3utils-1.45-x86_64-1.txz and mosh-1.3.2-x86_64-9cf.txz thank you, -Dan
  7. love nerdtools/nerdpack, glad it's coming back. I only have tmux and screen enabled and both are up-to-date matching the GitHub files but both say update ready even after I've updated. Also, is there a preferred process to request some of the missing packages that were available in 6.10? or are you in the process of vetting them to be re-added? thank you, -Dan
  8. So, Mac user here, Before Unraid, I was running a rpi with an external drive setup for Time Machine. It works ok given the limitations of the rpis usb/nic/io etc. Now with unraid I wanted to make that more robust. Watched Spaceinvaderone's videos, would work once then unable to complete incremental backups. Same Failed to attach using DiskImages2 errors. added the suggested changes to smb conf, same failure for incremental backups got excited for 6.11 and the macOS optimizations, same failure. gave up and ran the docker version, initial backup and hourlies for the past 36 hours. Success! now, according the the smb.conf from the docker app version, they use several different options in smb conf: [global] access based share enum = no hide unreadable = no inherit permissions = no load printers = no log file = /var/log/samba/log.%m logging = file max log size = 1000 security = user server min protocol = SMB2 server role = standalone server smb ports = 445 workgroup = WORKGROUP vfs objects = acl_xattr fruit streams_xattr fruit:aapl = yes fruit:nfs_aces = yes fruit:model = TimeCapsule8,119 fruit:metadata = stream fruit:veto_appledouble = no fruit:posix_rename = yes fruit:zero_file_id = yes fruit:wipe_intentionally_left_blank_rfork = yes fruit:delete_empty_adfiles = yes [TimeMachine] path = /opt/timemachine inherit permissions = no read only = no valid users = myusername vfs objects = acl_xattr fruit streams_xattr fruit:time machine = yes fruit:time machine max size = 1 T notably, [Global] vfs objects = acl_xattr fruit streams_xattr fruit:aapl = yes fruit:nfs_aces = yes fruit:model = TimeCapsule8,119 fruit:metadata = stream fruit:veto_appledouble = no fruit:posix_rename = yes fruit:zero_file_id = yes fruit:wipe_intentionally_left_blank_rfork = yes fruit:delete_empty_adfiles = yes and with the descriptions of the new optimizations here: there is mention of some of the desired options being overruled when a Mac isn't the first to use the share or that some of the fruit options that may be required for Time Machine to function properly needing to be added to [Global] may negatively impact the overall samba shares on unraid. My overarching frustration is that the option exists in the share export dropdown, yes, Timemachine or yes, TimeMachine (hidden), and this lends me to believe that the service that is not working is something that my license paid for. If we can't get it working, I'd rather see the developer for the docker app https://github.com/mbentley/docker-timemachine be compensated and his option be added to unraid as the "official" way to use unraid for a timemachine server. tl;dr remove the export option (yes, Time Machine) and add the docker app TIMEMACHINE APPLICATION as the officially supported option my USD$0.02, -Dan