jowy_ham Posted September 15, 2023 Share Posted September 15, 2023 (edited) Recently my UnRAID system has been crashing randomly. Can't find any errors in syslog. System was still working fine at 11+am then suddenly, unable to access web and SSH in at 5pm. Only task it was doing was preclearing 1 x 12TB HDD, nothing intensive was performed during this period. I have the following running (24x7): 1 x piHole 1 x Linux for torrenting 1 x Windows 2019 for DHCP services These crashes will rendered the server unresponsive: - No display output (most of the time, the system runs headless). I have tried to connect up a monitor when it crashes. but nothing shown (black screen) - Keyboard not responding (Alt+Ctrl+Del) but Caps lock lights up - web interface not accessible - SSH not accessible And after hard resetting the system, I would encountered random disks having disk errors (during auto parity checks upon array restarted) and these errors will disappear upon another reboot (I will shutdown the array and reboot the system again) without meddling with any HW connections. And sometimes 1 or 2 disks will be disabled due the errors, thus it will trigger a RAID rebuild which is ....... Attached is my diagnostic logs, hope experts can help tower-diagnostics-20230915-1749.zip Edited September 15, 2023 by jowy_ham Add more info Quote Link to comment
JorgeB Posted September 15, 2023 Share Posted September 15, 2023 If this started out of the blue without any changes it sounds more like a hardware issue, but you can try enabling the syslog server and post that after a crash in case there's something visible there. Quote Link to comment
jowy_ham Posted September 15, 2023 Author Share Posted September 15, 2023 24 minutes ago, JorgeB said: If this started out of the blue without any changes it sounds more like a hardware issue, but you can try enabling the syslog server and post that after a crash in case there's something visible there. Syslog enabled. Sep 15 11:52:44 Tower sSMTP[8441]: Creating SSL connection to host Sep 15 11:52:45 Tower sSMTP[8441]: SSL connection using TLS_AES_256_GCM_SHA384 Sep 15 11:52:48 Tower sSMTP[8441]: Sent mail for [email protected] (221 2.0.0 closing connection j5-20020a17090aeb0500b0026b4ca7f62csm1999412pjz.39 - gsmtp) uid=0 username=root outbytes=910 Sep 15 12:10:59 Tower emhttpd: spinning down /dev/sdg Sep 15 12:11:02 Tower emhttpd: read SMART /dev/sdg Sep 15 12:41:02 Tower emhttpd: spinning down /dev/sdg Sep 15 12:41:05 Tower emhttpd: read SMART /dev/sdg Sep 15 13:11:06 Tower emhttpd: spinning down /dev/sdg Sep 15 13:11:09 Tower emhttpd: read SMART /dev/sdg Sep 15 13:41:10 Tower emhttpd: spinning down /dev/sdg Sep 15 13:41:13 Tower emhttpd: read SMART /dev/sdg Sep 15 14:11:13 Tower emhttpd: spinning down /dev/sdg Sep 15 14:11:15 Tower emhttpd: read SMART /dev/sdg Sep 15 14:41:16 Tower emhttpd: spinning down /dev/sdg Sep 15 14:41:19 Tower emhttpd: read SMART /dev/sdg Sep 15 17:46:05 Tower file.activity: Starting File Activity Sep 15 17:46:05 Tower emhttpd: Starting File Activity... Sep 15 17:46:05 Tower file.activity: File Activity inotify starting @17:46, the system was forced resetted Nothing much was logged prior to that Quote Link to comment
jowy_ham Posted September 15, 2023 Author Share Posted September 15, 2023 I have just turn on syslog mirroring to flash (USB), hope that will capture more info. Cos previous logs were logged to a share folder on the ARRAY, so I guess when the ARRAY crash, the logs can't be logged Quote Link to comment
JorgeB Posted September 15, 2023 Share Posted September 15, 2023 If it's a hardware issue it usually doesn't leave anything relevant logged, but see. 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.