NoRaid99

@volcs0 When in my system NUT is not working anymore, then it is allways the loss of usb connection. Don't know how your ups is attached to the server. But if you log then it is a good chance to find something there.

Yesterday I updated my three servers from 7.2.6 to 7.3.1. (Qnap TS-253d, 2700x and 3700x B450) All have running dockers and VM's. Everything is woking without any problems and like expected! About internal boot: How do you deal with when you have to rollback a release because of problems and the server wont start or not accessible anymore? With a pen drive this is easy, but with internal drive?

@MothyTim In my setting this is at zero as well and shutdown of the UPS is working. edit: sorry that I can't really help you. Maybe there is something to find in the logs, when you cut power and see what gets logged. Or have a try with driver 2.8.4 edit2: May 28 16:03:24 QNAP ool www[2808853]: /usr/local/emhttp/plugins/nut-dw/scripts/start May 28 16:03:25 QNAP rc.nut: Writing NUT configuration... May 28 16:03:28 QNAP rc.nut: Updating permissions for NUT... May 28 16:03:28 QNAP rc.nut: Checking if the NUT Runtime Statistics Module should be enabled... May 28 16:03:28 QNAP rc.nut: Disabling the NUT Runtime Statistics Module... May 28 16:03:29 QNAP rc.nut: Network UPS Tools upsdrvctl - UPS driver controller 2.8.4 release May 28 16:03:29 QNAP rc.nut: Network UPS Tools 2.8.4 release - Generic HID driver 0.67 May 28 16:03:29 QNAP rc.nut: USB communication driver (libusb 1.0) 0.50 May 28 16:03:29 QNAP rc.nut: Using subdriver: CyberPower HID 0.84 May 28 16:03:29 QNAP rc.nut: Defaulting 'pollfreq' to 12 for CPS devices May 28 16:03:29 QNAP usbhid-ups[2812862]: Defaulting 'pollfreq' to 12 for CPS devices May 28 16:03:29 QNAP usbhid-ups[2812862]: You may want to set 'pollonly' flag on CPS devices May 28 16:03:29 QNAP rc.nut: You may want to set 'pollonly' flag on CPS devices May 28 16:03:29 QNAP usbhid-ups[2812862]: Listening on socket /var/run/nut/usbhid-ups-ups May 28 16:03:29 QNAP rc.nut: Listening on socket /var/run/nut/usbhid-ups-ups May 28 16:03:29 QNAP usbhid-ups[2812864]: Startup successful May 28 16:03:30 QNAP rc.nut: Network UPS Tools upsd 2.8.4 release May 28 16:03:30 QNAP rc.nut: listening on 0.0.0.0 port 3493 May 28 16:03:30 QNAP upsd[2812898]: listening on 0.0.0.0 port 3493 May 28 16:03:30 QNAP upsd[2812898]: Connected to UPS [ups]: usbhid-ups-ups May 28 16:03:30 QNAP rc.nut: Connected to UPS [ups]: usbhid-ups-ups May 28 16:03:30 QNAP usbhid-ups[2812864]: sock_connect: enabling asynchronous mode (auto) May 28 16:03:30 QNAP upsd[2812898]: Found 1 UPS defined in ups.conf May 28 16:03:30 QNAP rc.nut: Found 1 UPS defined in ups.conf May 28 16:03:30 QNAP upsd[2812899]: Startup successful May 28 16:03:30 QNAP rc.nut: Network UPS Tools upsmon 2.8.4 release May 28 16:03:30 QNAP rc.nut: UPS: [email protected] (primary) (power value 1) May 28 16:03:30 QNAP rc.nut: Using power down flag file /etc/nut/killpower May 28 16:03:30 QNAP upsmon[2812902]: Startup successfulThats the normal log from unraid when starting nut service, maybe you find something missing at yours. edit3: WARNING: In most cases it is sufficient to configure NUT using the GUI for the smoothest possible user experience. Any changing the configuration directly can have unexpected side effects, use 'Reset Config' in case of problems. Have you checked the debug file? NUT Debug Package (.zip) In my setup upsmon.conf.txt shows POWERDOWNFLAG "/etc/nut/killpower" I guess driver.flag.allow.killpower will be set just in time when needed as variable. The description for the command: Tell the driver daemon to initiate UPS shutdown; should be unlocked with driver.flag.allow_killpower option or variable setting.

Does your UPS support this switch-off function after auto shutdown the server?

Don't think so. If the cells do not get charged up to 100% some times, there will be no satisfactory balancing. At least 3.5-3.55V per cell is needed for this to even start the process. And it needs as well a minimum amount of Amps to realise a effective balancing. Here I see as well a problem with the low charging current used for the small lead acid bateries used in UPS. That is marketing speak. Lead-acid-bat can never be the the same like LiFe-bat. and only can be replaced with compromises. They are basically different and are coosen by purpose. It is like running your diesel car with sunflower oil. Just don't do it, if you don't know exactly what you are doing.

Good point. And what about the cut off voltage? Anyway, personally I wouldn't do it. edit: -like this they never get balanced!!! -perhaps the Max. Continuous Discharge Current is not high enought -perhaps the Recommended charging current is not set to the right value and could be to high -possibly the lead-bat. gets charged sometimes with more then the 13.6V. Usualy 14.2V would be the voltage to keep lead-bat. at charge level.

@Rysz Okay, thank you for the explanation. Never have seen this before. Still seems to be a bit strange to me when the UPS reports "OB or CAL" while a power interruption. Who, if not the UPS itself knows what happened?

I see a other big problem when replacing Lead-Bat. to Li-Bat. for UPS. Keeping them constantly at 100% charging level is realy bad. I would say it is a security risk. If you go for Li-Bat. then buy a UPS that is build for that.

Nice helpful tool, thanks for your work! @jcofer555 One thing I want to comunicate is that I would like the possibility to choose a light theme, because of my eye-problems. It is possible to stand it, because I do not spend much time on the applications-page, but would be helpful anyway. Perhabs for other people as well.

because of the security updates: Updated from 7.2.4 to 7.2.6 without any problems on a TS-253d with lots of dockers and a few Linux VMs. To keep the same old mac adresses of the docker conainers I did stop all dockers before the update, unticked the "auto start", updated and rebooted. Then it is important to change the templates from the docker containers before they start the first time and set the old fixed MAC. This way all did stay the same like before. I really don't hope that we soon have to update every week because of security fixes. 🤞

Hi all, there is a log generated while having power problems/cuts from my UPS and I would need help to understand what this means: May 7 02:32:56 QNAP usbhid-ups[3710]: ups_status_set: seems that UPS [ups] is in OL+DISCHRG state now. Battery charge is currently 87. Is it calibrating (perhaps you want to set 'onlinedischarge_calibration' option)? Note that some UPS models (e.g. CyberPower UT series) emit OL+DISCHRG when in fact offline/on-battery (perhaps you want to set 'onlinedischarge_onbattery' option) The UPS is a CP1500EPFCLCD (not UT series). OL+DISCHRG means what? overload and discharge? I'm asking because I'm trying to observe nasty power problems on my private source (off-grid-inverter) and have seen this entry in the server.log. thanks in advance noraid

I will report back, but do not have the time atm to update. There is to much "important" stuff on that server.

Brand: Seagate Model: ST8000DM004-2CX188 Approximate age at failure: 5 years Drive in warranty: never Origin: cheap buy out of a external USB case Because the drives are only for redundancy of a backup-server and lak of money, I did buy the cheapest I could get. Don't have the drive connected to a computer atm, but the drive has only very little working hours and read/write values, because this server gets started only once a week and there are no big changes to backup. I post here because this disk made me cracy and I want to share my experience with it. Since a long time I do have copy/speed problems on network when this server is in use. Funny erros like network errors. Sometimes all was fine, sometimes I had to restart network or the server. There was never any indication for the cause. I changed nearly everything and testet and testet... no errors at all in log. Before a month, I did rebuild the server with the same HW, but new boot media and it started to show up some errors sometimes. PCIe errors, network errors, funny self reboots of the server, only when there was action on this specific disk. Then I was checking this disk with full SMART test, Seagate extended tests, preclear with read-write-read, but all was perfect! But finally it is this blo..y disk causing all these funny errors and no test type could find something unusual.

@jademonkee ,thanks a lot for your answer. I think this makes sense like you explained and hope somebody can confirm, if so. In v.7.2.4 and before, (in my case - container MAC with active NetworkType: custom: eth0 and specified local IP) , the first generated MAC did persit every reboot and update of the container. This is important for me because DHCP and firewall settings. So, if I'm right, from now on I do have to: This creates a new question, if I stop docker, deactivate auto start and update to 7.4.5, set the new value MAC Adress in the template, will it be possible somehow to keep the old MAC?

Only to get prepared for this breaking change. Why "may"? Are there special settings or circumstances causing the "generate of a new dynamic MAC"?

Don't want to complicate things or causing confusions, but perhaps it's worth to have a try with the last stable version and not the newest 2.8.5 if you use that one. Newest Nut-release 2.8.5 causes driver crashes on my system with the result of not recognising the usb-ups-device anymore. Rollback to 2.8.4 was my quick solution, to not needing to search for bugs or incompatibilitys.

Yes, I found it meanwhile. But it is on both servers at v5 but different amount of space used, like I did say before. So v4 and v5 don't say anything about the 153GB or the 58,3GB used space for format. Is this right? UnRaidNAS1 kernel: XFS (dm-0): Mounting V5 Filesystem 9a346960-a059-4818-afd2-97237eba549e (on all disks from the array) UnRaidNAS2 kernel: XFS (dm-0): Mounting V5 Filesystem 6861d2fc-c353-46e8-be0e-1ecaa666d216 (on all disks from the array)seems that I am lucky.

Thanks for this additional info. I never have seen any warning about this after any update from v6 until now v7.2.4 (Great, since a few days I'm copying files to that server with the older XFS format.) Where can I find the XFS version running atm?

Allright, thanks again for your brillhant support. wish you a good time!

Yes its formated with XFS, thanks for this information. Higher overhead means that its using more space, right? Is there any benefit or recommendation to reformat with the newer kernel?

Hello, I do have a question about the space that unraid is taking for the array functionality on the disks when creating new shares. Wanted to ask because on my two backup-servers I did manage somehow to have different disk-space occupied by unraid array. After adding some more disks I was rebuilding my shares. Now I created a normal samba share with 6 disks, no auto splitting, high water, all is working fine until now. The first server tooked 150...GB something per disk. The second server only 55,8GB per disk. The disk have all the same size with 8TB, but are different brands. As well some SMR but the most are CMR type. Why did this happen? Is there a logic that could explain to me, what I did different or wrong?

hope it is okay to ask here. is there still the time limit of 12 month if changing the boot device and replaceing the license?

Thanks again for your help guys. I will continue in future to sync with FreeFileSync, somehow I like this tool. The thing is, I did change this backup server with a different hhd config, so I had to rebuild the shares and now it was the first time that I did choose a share "pool" with 6 hdds. And for now it is just to fill the backup disks again with the old datas and this caused all my missunderstandings and problems, because it didn't run like I expected. I'm copying at the moment with krusader one disk by one from the main server to the backup server. Afterwoods when this is finished there will be only a few files getting updated, removed or added at once and it wont be a problem anymore with FFS and "high water". And with your help I understand now how split level works and understand what the benefit is, but I don't want it. I have the feeling that I will loose control of my folders and what's in them. I am lucky to not needing every bit on my drives. As well they are less stressed when they have some free space to deal with. For rsync I would prefer a GUI. But for regular jobs with the same parameters I could imagine to use it.

No, finally I figured out it's not about splitting, it is about writing to all the disks of the share. this has to do with the allocation method. The problem is that if I copied a bunch of folders with FreeFileSync they where all created first as empty folders in disk one of the share and thats why the datas where filled into these folders, without respecting the allocation method. When I copy the hole bunch of files by linux filemanager, one folder gets copied with its content and then the next an so on. That way the allocation method gets respected and all works with "high water" now. My only problem now is to find out what syncing tool uses a copy program that does copy file by file including contents. Thanks a lot for the example to explane the levels. I was to confused to understand this because the behavior of the way that FFS is copying. (first create all folders and then fill them up) But I know now, that I do not need auto split. The first level folders won't grow, they stay, get deletet or substituted completely. I only want to fill the disks with "high water" allocation. If somebody knows a syncing tool, that works different (one complete folder by the other), I would love the know.