Hoopster

Members
  • Posts

    3466
  • Joined

  • Last visited

  • Days Won

    19

Hoopster last won the day on January 29

Hoopster had the most liked content!

6 Followers

Retained

  • Member Title
    unRAIDy to Rumble

Converted

  • Gender
    Male
  • Location
    Utah USA

Recent Profile Visitors

9240 profile views

Hoopster's Achievements

Proficient

Proficient (10/14)

757

Reputation

  1. Many people get call trace errors with macvlan which lock up the server eventually. This usually occurs when you have docker containers to which custom IP addresses have been assigned on br0. The solution for me and many others was to create a VLAN on which docker containers get IP addresses assigned. In my case, that is br0.3. This eliminated the macvlan call traces. The new ipvlan custom network type in 6.10.0 rc1 is another potential solution to this problem. In one of my server crashes, I did get a call trace that appeared to involve macvlan (among several other things) so I went with ipvlan. It did not solve my crashes which now appear to have been PSU related although I am still testing that theory.
  2. I can echo everything @kaiguy said. Turbo Boost disabled through Tips and Tweaks, MB reporting 88C temp in my case which is bogus, Docker custom network type set to ipvlan. I have had zero CPU_CATERR lockups since disabling Turbo Boost. I have not tried enabling it since moving to the 6.10.0 RC1 a month or so ago but I am fairly certain I would again get CPU_CATERR lockups if I did as this appears to be related to Turbo Boost on this board and has nothing to do with unRAID/Linux kernel version. I have been experiencing server lockup every 5-15 days since July; however, I now think that was due to a flaky PSU. I swapped out the PSU 17 days ago and the server has been running without issue since then.
  3. Hoopster

    30

    The "unlimited storage" statement is technically true as it applies to cache pools/drives. It is not true when it comes to array storage. That could certainly be more clearly stated. As mentioned, multiple arrays have been mentioned as a possible future feature but that is not much help to you in the here and now.
  4. I cannot add a peer or modify an existing Wireguard peer. Every time I try, the message below appears: I wanted to change a particular peer from Access to Server to Access to LAN but it's a no go due to this message about the peer name. I cannot make any changes to peers at all. I am running 6.10 RC1 on this server. Could it be related to that version in particular? UPDATE: It looks like I was able to change it on another computer (Chrome instead of Firefox). I did not get this popup message and it let me make the change after correcting the error in another peer name. Apparently, allowed characters in peer names have changed since I first set up the peers. A peer I was not trying to change in any way contained an invalid character (an apostrophe) and was triggering the warning even though it had nothing to do with my attempt to add a new peer or change an existing peer.
  5. My Dell H310 required the tape trick mod. Many systems will not even boot with the Dell H310 installed unless you do the tape mod. If the attached drives are not showing up in the motherboard BIOS, they will never show up in unRAID. I bought a server pull H310 off eBay and flashed the card to IT mode myself (without the card BIOS). Some Kapton tape over pins 5 and 6 did the trick for me and it has been smooth sailing every since. You might also want to make sure your motherboard BIOS is updated to the most recent for the board. Recently an unRAID user had the same issue of drives attached to a Dell H200 (same as H310 but different SAS port location) not showing up at all in unRAID. His server booted with the card installed but no drives showed up. A motherboard BIOS upgrade fixed it for him.
  6. Every time I have installed a Windows VM, if the Shell> prompt appears I just type 'exit' at the prompt and installation continues normally.
  7. The default go file will contain this: #!/bin/bash # Start the Management Utility /usr/local/sbin/emhttp & This starts the GUI. Your go file looks fine so there must be another reason you cannot access the GUI via a browser.
  8. USB 2.0 drives attached to a USB 2.x port on server is usually best. There have been reliability problems in the past with USB 3.0 thumb drives on USB 3.0 ports. You don't really need USB 3.0 speeds for anything related to the flash drive but you probably would notice a difference between USB 2.0 and USB 1.0. Having said that, 1 GB may be a bit tight on space if you install lots of plugins, run several user scripts, need to mirror the syslog to the flash drive for troubleshooting, etc. I have an 8GB flash drive but I have only used 838 MB space on that thumb drive. I have lots of plugins but only 2 user scripts. There is undoubtedly some junk on that drive I could clean up. I mirror the syslog elsewhere. You could use the larger USB 3.x thumb drive and connect it to a USB 2.0 port for increased reliability. I have two servers; one has a USB 2.0 flash drive in a USB 2.0 port and the other has a USB 3.0 flash drive in a USB 2.0 port. Neither has given me problems and I have been running the first server since 2011 and the second since 2015. The flash drive is generally accessed in the following situations: At boot (reading) Saving configuration changes (writing) Running user scripts (reading) Writing syslog info if so configured (writing) You can also get started with one of those flash drives and then transfer the license to another flash drive if needed. You can do that once a year very easily.
  9. Electrical tape will do in a pinch but it is thick and is not the best solution. The recommended tape is Kapton tape or equivalent. I used electrical tape to test the solution but replaced it with Kapton as it is made for uses like this. Taping those pins is worth trying. It will certainly do no harm and could very well be what is needed to get the card/attached drives recognized in the BIOS and unRAID.
  10. You may need to resort to the tape trick to get it to work. In some cases the server will not even boot with Dell H200/Dell H310 cards installed until pins B5 and B6 are taped. This was the case with my Dell H310. In other cases, the server will boot but no drives attached to the controller are recognized. Check in your BIOS. Are the drives attached to the H200 seen in the BIOS? If not, they will never show up for unRAID. The Dell H200 and Dell H310 are basically the same card with the connectors in different locations.
  11. Crashed again 5 days ago after about a week of uptime. Locked up again today. The frequency of lockups is increasing and nothing relevant that I can see in syslog. I have just swapped out the power supply as a test. Perhaps the other PSU is failing although it was brand new 18 months ago.
  12. According to your diagnostics file (and the network config screenshot) you have NIC bonding enabled. It is probably a good idea to get a single unbonded NIC working as a first step. NIC bonding with certain NICs has been known to cause connectivity issues in the past. No guarantee this is the problem but it won't hurt to disable bonding.
  13. Sounds like the problems lots of 1st and 2nd generation Ryzen systems had with Linux/unRAID. See this post for some things that can be done to improve Ryzen system stability with unRAID.
  14. Locked up again today on 6.10 RC1 and with the Docker custom network type set to ipvlan. Obviously this was not the issue but it was worth trying. It locked up again approximately two weeks after last reboot which has been the pattern since this started roughly in early July. I'll keep digging. This is a weird one with nothing very helpful in the syslog.
  15. I have an rsync backup script that I run through User Scripts that sends an email summary of the backup process. Here are the relevant portions of the script I use: # Set up email header echo To: xxxxxxx@yahoo.com >> /boot/logs/cronlogs/BackupNAS_Summary.log echo From: xxxxxxx@gmail.com >> /boot/logs/cronlogs/BackupNAS_Summary.log echo Subject: MediaNAS to BackupNAS rsync summary >> /boot/logs/cronlogs/BackupNAS_Summary.log echo >> /boot/logs/cronlogs/BackupNAS_Summary.log # Send email of summary of results ssmtp xxxxxxxx@yahoo.com < /boot/logs/cronlogs/allshares.log cd /boot/logs/cronlogs mv BackupNAS.zip "$(date +%Y%m%d_%H%M)_BackupNAS.zip" rm *.log The script concatenates share backup info in allshares.log and then that is emailed to a yahoo.com address