DigitalMaestro Posted April 15, 2022 Share Posted April 15, 2022 (edited) I do have auto-start enabled in Disk Settings. Scenario: I am testing the UPS shutdown settings. Unraid successfully initiates shutdown once the UPS hits the defined runtime. When I plug the UPS back into the mains, the computer restarts. Unraid loads successfully but does not start the array (array status is 'Array Stopped'). However the array configuration is shown as valid in the "Main" section. Interestingly, if I then click the Reboot button, Unraid reboots and subsequently auto-starts the array as expected. Is there an additional configuration step needed to enable auto-start after UPS shutdown? Diags attached (created after first reboot from UPS shutdown when array did not auto-start). Any insight appreciated. Edit: I went ahead and tested three scenarios after this auto-start failure to see if I could glean anymore insight Reboot - Unraid auto starts properly Shudown - After manually powering up the computer Unraid auto starts properly UPS shutdown - In this case Unraid did auto start correctly. So it appears that the issue may not be UPS related after all. The core question is thus: Why would Unraid intermittently not auto start even though it reports the array configuration as valid? Thanks DM diagnostics-20220415-1431.zip Edited April 15, 2022 by DigitalMaestro Quote Link to comment
trurl Posted April 15, 2022 Share Posted April 15, 2022 1 hour ago, DigitalMaestro said: When I plug the UPS back into the mains, the computer restarts. Do you mean your BIOS is configured to automatically boot when power is restored? Quote Link to comment
DigitalMaestro Posted April 15, 2022 Author Share Posted April 15, 2022 1 hour ago, trurl said: Do you mean your BIOS is configured to automatically boot when power is restored? Yes, the BIOS is configured to automatically (re)start the computer when power is applied. DM Quote Link to comment
trurl Posted April 15, 2022 Share Posted April 15, 2022 I notice some non-stock code in your go file related to Marvell. I assume you know Marvell controllers are NOT recommended. A lot going in syslog between the start of go and emhttp actually running. Look at logs/syslog.txt in those diagnostics. Lots of stuff about disabled ports, and when it finally does start emhttp all your disks are missing, with the Unassigned disks detected first. Then a little later your assigned disks are seen, but too late for autostart. Quote Link to comment
trurl Posted April 15, 2022 Share Posted April 15, 2022 Looking in system/lspci.txt I don't actually see a Marvell controller. What are those go lines about then? Quote Link to comment
DigitalMaestro Posted April 15, 2022 Author Share Posted April 15, 2022 13 minutes ago, trurl said: Looking in system/lspci.txt I don't actually see a Marvell controller. What are those go lines about then? My old system had Marvell controllers. New system does not. I simply built the new system then moved the disks and flash drive over. Was there something I needed to do the clean up the go file? Any other file cleanup required? Thanks DM Quote Link to comment
Solution trurl Posted April 16, 2022 Solution Share Posted April 16, 2022 13 hours ago, DigitalMaestro said: clean up the go file Remove those lines. They aren't working and aren't needed. Quote Link to comment
trurl Posted April 16, 2022 Share Posted April 16, 2022 14 hours ago, trurl said: A lot going in syslog between the start of go and emhttp actually running. Look at logs/syslog.txt in those diagnostics. Normally, this line in logs Apr 15 13:59:45 Jupiter root: Starting go script is immediately followed by this line (but NOT 5 seconds later) Apr 15 13:59:50 Jupiter root: Starting emhttpd 14 hours ago, trurl said: Lots of stuff about disabled ports, and when it finally does start emhttp all your disks are missing, with the Unassigned disks detected first. Then a little later your assigned disks are seen, but too late for autostart. Quote Link to comment
DigitalMaestro Posted April 16, 2022 Author Share Posted April 16, 2022 6 hours ago, trurl said: Normally, this line in logs Apr 15 13:59:45 Jupiter root: Starting go script is immediately followed by this line (but NOT 5 seconds later) Apr 15 13:59:50 Jupiter root: Starting emhttpd Thank you for all the feedback and help. I think removing those lines may have done the trick! Now the "Starting go script" is immediately followed by "Starting emhttpd" Also the web GUI is now active much faster. I was accustomed to waiting 30 seconds to a minute after the login prompt appeared on the monitor after bootup before I could load the web GUI in the browser from another computer. Now the browser can navigate to the web GUI almost immediately after the login prompt appears. System auto started as expected. Since the failure was intermittent, I'll need to see a few shutdown cycles before I am 100% certain that the problem is resolved but I'm pretty sure it is. Thanks once again! DM Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.