November 5, 20178 yr 16 hours ago, RonUSMC said: Well, its been up for 3 days and everything seems fine after I made the edit Johnnie had me do. I'm sitting watching Plex and it dies. I go to the Control Panel. "Array Undefined" tower-diagnostics-20171104-2232.zip I was happy about hitting the 12 hour mark for the first time on this update but this makes me nervous again. any idea what happened?
November 5, 20178 yr Ron's issue *appears* to be the uninitialized CSRF errors causing this which only anecdotally may be cause by the rootFS being full (ie: saving stuff into RAM instead of the array) Your issue, you should post your diagnostics.
November 5, 20178 yr 10 hours ago, Lev said: Over my years using unRAID, this must attribute to 99% of the reasons why my array won't stop. I remember many times in my early years get frustrated and hard rebooting the server cause I was stuck at this same issue. I just didn't know any better. I wonder if this is that common among others. If that's the case, it makes me thing that a setting, enabled by default, that kills all ssh processes when 'stop array' is pressed would help more than it would hurt. Or perhaps a better middle ground is simple to 'ps' at time of shutdown and report to the GUI that there is a ssh terminal open and to please close it. Install the Tips and Tweaks plugin, or update it and you will be able to enter the processes to kill when the array is stopped. Be careful. You may lose data killing processes here. The processes will be killed whenever the array is stopped - webgui, ups shutdown, command line, short power button press, etc.
November 6, 20178 yr I have a new build, first time to Unraid and was until recently running the latest stable release. It's a Ryzen-based system and I had hard lock-ups at various times, but at least once a day. I finally sought help on the forums and saw the release candidate that addressed the Ryzen lockup issues and installed it. Unfortunately it appears that the issue is worse. What can I do to help provide data points toward the resolution?
November 6, 20178 yr 22 minutes ago, OnlyOneCannoli said: I have a new build, first time to Unraid and was until recently running the latest stable release. It's a Ryzen-based system and I had hard lock-ups at various times, but at least once a day. I finally sought help on the forums and saw the release candidate that addressed the Ryzen lockup issues and installed it. Unfortunately it appears that the issue is worse. What can I do to help provide data points toward the resolution? Have you tried this- https://forums.lime-technology.com/topic/61129-ryzen-freezes/
November 6, 20178 yr 2 hours ago, wgstarks said: Have you tried this- https://forums.lime-technology.com/topic/61129-ryzen-freezes/ Yes, I created a new menu entry for "Ryzen Fix" which was a copy of the GUI menu entry with the extra nocbs line. It locked up somewhere between 2 and 3 hours later (screen was showing Plex, so I'm not sure when).
November 6, 20178 yr @OnlyOneCannoli @gcalaway I've seen having stability issues on the newest rc-10b. I was not having issues on rc-9f. I'm going to go back this evening to see if I get my stability back. Edited November 7, 20178 yr by dkerlee
November 7, 20178 yr @dkerlee let me know how it goes, I have had no more crashes since adding the code personally, however I have seen a few log errors saying the Directory is not empty for the mnt/cache/appdata/Transcode/Sessions/plex. Other than that and a few updates to clear other errors (updating the Unassigned devices, and preclear) all has been going alright for 2 days 2 hours and 41 min...... fingers still crossed
November 7, 20178 yr I was wondering if UNRAID has all the new Z370 Intel Chipsets included. I picked one up and having trouble finding the right driver for the hardware sensors.
November 7, 20178 yr 6 hours ago, dalben said: I was wondering if UNRAID has all the new Z370 Intel Chipsets included. I picked one up and having trouble finding the right driver for the hardware sensors. Hardware sensors tend to be provided by whatever legacy multi-IO chip the motherboard vendor uses, be it ITE, VIA or whatever. It's got little/nothing to do with the chipset. Edited November 7, 20178 yr by HellDiverUK
November 7, 20178 yr 50 minutes ago, HellDiverUK said: Hardware sensors tend to be provided by whatever legacy multi-IO chip the motherboard vendor uses, be it ITE, VIA or whatever. It's got little/nothing to do with the chipset. OK, then I assume it's the nct6683 driver it's looking for. Sensors-detect gave me this: Trying family `VIA/Winbond/Nuvoton/Fintek'... Yes Found `Nuvoton NCT6683D eSIO' (address 0xa10, but not activated) I do get some warnings on bootup. Everything seems to run fine though. I posted a syslog of the system bootup in case anyone is interested looking at the warnings. Mainly around ACPI from what I can see. Oh, and an on-existent floppy drive. tdm-syslog-20171107-2054.zip
November 7, 20178 yr 21 hours ago, OnlyOneCannoli said: Yes, I created a new menu entry for "Ryzen Fix" which was a copy of the GUI menu entry with the extra nocbs line. It locked up somewhere between 2 and 3 hours later (screen was showing Plex, so I'm not sure when). I had to disable C-states in BIOS in order to avoid freezes with my Ryzen 1600X. It is called AMD CBS on my Asus Prime x370. Previously I also had to disable multithreading (SMT) and Nested Page Tables (NPT), but it is not necessary with the 6.4 prerelease based on Linux 4.13. Edited November 7, 20178 yr by pederm
November 7, 20178 yr I'm using RC10 with encrypted drives and am using the High Water option. There appears to be a glitch with file transfers because the first hard drive filled up fine, but moving on to the second hard drive it appears that everything is being duplicated onto the other drives. You can see that the drive size increase, but there are no recorded drive writes. Am I missing something or is this a bug? Thanks! I did some fiddling around and it appears that disks 2,3,4 are always duplicated regardless of what method I use wheras disk 1 is not. Edited November 7, 20178 yr by IsJohn additional info.
November 7, 20178 yr 21 minutes ago, IsJohn said: I'm using RC10 with encrypted drives and am using the High Water option. There appears to be a glitch with file transfers because the first hard drive filled up fine, but moving on to the second hard drive it appears that everything is being duplicated onto the other drives. You can see that the drive size increase, but there are no recorded drive writes. Am I missing something or is this a bug? Thanks! Appears to be a problem with High Water because this duplication doesn't occur when I do Fill Up If "everything is being duplicated onto the other drives" that should be fairly easy to demonstrate. Your screenshot doesn't.
November 7, 20178 yr 38 minutes ago, trurl said: If "everything is being duplicated onto the other drives" that should be fairly easy to demonstrate. Your screenshot doesn't. First I uploaded disk1.jpeg to the share and as you can see the file is only transferred to Location DIsk1. Next, I edit the share and exclude disk1 from the share and then upload disk2.jpeg. As you can see, it is duplicated onto all three drives as shown by the location column. Edit: It doesn't look like it is duplicating the data so much as it thinks all three drives in a pool. When I exclude drive 2 from the share and transfer a file to the share thinking that only drive 3 and 4 should receive the file, drive 2 also has the file. Also, when I change the filesystem of the drive, all three drives are also changed. Edited November 7, 20178 yr by IsJohn additional info.
November 7, 20178 yr 9 hours ago, dalben said: OK, then I assume it's the nct6683 driver it's looking for. Sensors-detect gave me this: Trying family `VIA/Winbond/Nuvoton/Fintek'... Yes Found `Nuvoton NCT6683D eSIO' (address 0xa10, but not activated) I do get some warnings on bootup. Everything seems to run fine though. I posted a syslog of the system bootup in case anyone is interested looking at the warnings. Mainly around ACPI from what I can see. Oh, and an on-existent floppy drive. tdm-syslog-20171107-2054.zip Let me guess, you bought an AsRock?
November 8, 20178 yr 3 hours ago, HellDiverUK said: Let me guess, you bought an AsRock? That obvious? Yeah, the Z370m PRO4. Is this a known issue with ASROCK boards?
November 8, 20178 yr @pederm @dkerlee @wgstarks I changed the order of the operations of the append line to have the nocbs=0-7 first before the image definitions and I've been up for ~23hrs. @dalben I haven't been able to pick up the sensors in Unraid on my system either using Ryzen 1500 and MSI B350 MoBo.
November 8, 20178 yr I have a disk that will not spin down no matter what I do. All the settings are right, but it will only spin down if I force it. It will stay spun down once I do though. Also, I'm aware of the bad drive and have a new one on order. unraid1-diagnostics-20171108-0822.zip
November 8, 20178 yr 13 hours ago, dalben said: That obvious? Yeah, the Z370m PRO4. Is this a known issue with ASROCK boards? AsRock tend to be the only ones who use that sensor chip. Good luck getting it working. I never had any success with it on the few AsRock boards I had that weren't DOA or produced flames the first time I switched it on.
November 8, 20178 yr [mention=77711]gcalaway[/mention] been up for 2 days. I think I'm in the clear, knock knock... Crashed a couple more times, even after that modification (rcu_nocbs=0-3) I'm going back to rc9. Edited November 8, 20178 yr by dkerlee
November 8, 20178 yr 7 hours ago, HellDiverUK said: AsRock tend to be the only ones who use that sensor chip. Good luck getting it working. I never had any success with it on the few AsRock boards I had that weren't DOA or produced flames the first time I switched it on. ASROCK replied saying the controller is a Super I/O NCT6683D-T. In the same email they pointed out Linux isn’t an officially supported OS. This should be fun.
November 10, 20178 yr Been upgrading to the latest versions and only recently noticed something that makes me wonder if it is the OS rc10b or another component that is causing my issues. I have a APC Smart-UPS 1500 with a network card attached, it is set up with my vCenter cluster and unRAID. What I am noticing is that unRAID is losing connection to the UPS and then almost immediately reconnecting without any interaction on my part. Just in the last two days, I have been getting another alert about insufficient system storage even though my USB stick is only 5% used. Attached is from my logs, I can grab more if needed but this shows the frequency of this oddity. Oct 31 08:16:09 Server apcupsd[7460]: Communications with UPS lost. Oct 31 08:16:10 Server apcupsd[7460]: Communications with UPS restored. Nov 2 12:46:43 Server apcupsd[7460]: Communications with UPS lost. Nov 2 12:46:43 Server apcupsd[7460]: Communications with UPS restored. Nov 5 04:40:01 Server apcupsd[7460]: apcupsd exiting, signal 15 Nov 5 04:40:01 Server apcupsd[7460]: apcupsd shutdown succeeded Nov 5 04:40:04 Server apcupsd[4785]: apcupsd 3.14.14 (31 May 2016) slackware startup succeeded Nov 5 04:40:04 Server apcupsd[4785]: NIS server startup succeeded Nov 5 15:36:52 Server apcupsd[4785]: Communications with UPS lost. Nov 5 15:37:09 Server apcupsd[4785]: Communications with UPS restored. Nov 6 03:17:28 Server apcupsd[4785]: Communications with UPS lost. Nov 6 03:17:44 Server apcupsd[4785]: Communications with UPS restored. Nov 6 14:58:01 Server apcupsd[4785]: Communications with UPS lost. Nov 6 14:58:17 Server apcupsd[4785]: Communications with UPS restored. Nov 7 02:38:34 Server apcupsd[4785]: Communications with UPS lost. Nov 7 02:38:50 Server apcupsd[4785]: Communications with UPS restored. Nov 7 20:10:15 Server apcupsd[4785]: Communications with UPS lost. Nov 7 20:10:31 Server apcupsd[4785]: Communications with UPS restored. Nov 8 07:50:48 Server apcupsd[4785]: Communications with UPS lost. Nov 8 07:51:04 Server apcupsd[4785]: Communications with UPS restored. Nov 8 19:31:22 Server apcupsd[4785]: Communications with UPS lost. Nov 8 19:31:22 Server sSMTP[23669]: 452 4.3.1 Insufficient system storage Nov 8 19:31:38 Server apcupsd[4785]: Communications with UPS restored. Nov 8 19:31:38 Server sSMTP[25834]: 452 4.3.1 Insufficient system storage Nov 9 07:11:58 Server apcupsd[4785]: Communications with UPS lost. Nov 9 07:11:58 Server sSMTP[28658]: 452 4.3.1 Insufficient system storage Nov 9 07:12:14 Server apcupsd[4785]: Communications with UPS restored. Nov 9 07:12:14 Server sSMTP[30332]: 452 4.3.1 Insufficient system storage Nov 9 18:52:35 Server apcupsd[4785]: Communications with UPS lost. Nov 9 18:52:35 Server sSMTP[24797]: 452 4.3.1 Insufficient system storage Nov 9 18:52:51 Server apcupsd[4785]: Communications with UPS restored. Nov 9 18:52:51 Server sSMTP[26881]: 452 4.3.1 Insufficient system storage
November 10, 20178 yr @sparkyy I only have a little experience with the two programs that are giving you grief. But you can see error messages about SMTP running out of space. Perhaps it's logging, or keeping all messages and attachments somewhere it shouldn't be. Or it shouldn't be keeping them at all.Are any of those apps being ran as dockers? Dockers have unconventional storage requirements you should read up on if either of those programs is, in fact, a Docker. Sent from my Pixel XL using Tapatalk
November 10, 20178 yr @dkerlee I don't use my unRAID for anything other than SMB sharing. No dockers, no VMs, nothing fancy, just plain old file sharing to several Windows boxes. That's why it's odd to me that this is happening and started once I went up to rc10b.
Archived
This topic is now archived and is closed to further replies.