Toskache Posted July 14, 2023 Share Posted July 14, 2023 I am on 6.12.2 and after a power-failure the system is performing a parity-check. It starts at 70MB/sec (what is also very slow, usually the check starts with 170MB/sec) and now the system "stalls" with 22KB/sec and a load of over 16 (Intel Xeon E3-1270 v3 with 32GB ECC-RAM). I attached the diagnostic-package. Anny ideas are welcome - I can not wait the estimated 800 days to complete the check PS: * I paused the check - no effect, still load of >16 * I stopped docker - no big effect, still load of >13 I will give 6.12.3-rc3 a try.... nas-diagnostics-20230714-1942.zip Quote Link to comment
Toskache Posted July 15, 2023 Author Share Posted July 15, 2023 (edited) 6.12.3-rc3 didn't do the trick. I stopped docker but: What can I do? nas-diagnostics-20230715-1225.zip Edited July 15, 2023 by Toskache upload diagnostic-file Quote Link to comment
itimpi Posted July 15, 2023 Share Posted July 15, 2023 14 minutes ago, Toskache said: 6.12.3-rc3 didn't do the trick. I stopped docker but: What can I do? I would suggest updating to the stable 6.12.3 (probably the same as the rc3 release by why take a chance) and posting new diagnostics since you must rebooted to active 6.12.3. Makes sure that the problem is happening at the time the diagnostics are taken. Quote Link to comment
JorgeB Posted July 15, 2023 Share Posted July 15, 2023 Nothing obvious in the diags, run the diskspeed docker test to see if all disks are performing normally. Quote Link to comment
Toskache Posted July 15, 2023 Author Share Posted July 15, 2023 (edited) Thank you for the tip @JorgeB, but the diskspeed-test looks IMHO ok: Edited July 15, 2023 by Toskache Quote Link to comment
JorgeB Posted July 15, 2023 Share Posted July 15, 2023 I cannot see what's causing issues, reboot in safe mode and start the array in maintenance mode, then start a check, is it still slow? Quote Link to comment
Toskache Posted July 15, 2023 Author Share Posted July 15, 2023 (edited) @JorgeB I am getting closer to the root cause. I recently tried the docker "calibre" for my big eBook-Collection. The eBooks are original ordered in many sub-directories ("0-9", "A", "B", "C", ...). Calibre imports all the books in a single directory. So there are more than 80K directories in one level. Trying to access that directory seems to stall the system. Now I am trying to delete that directory with an "rm -rf" via the console but that will take a while... Edited July 15, 2023 by Toskache Quote Link to comment
Toskache Posted July 16, 2023 Author Share Posted July 16, 2023 Forget my last post. The check is of course at block level. It stalls again at 60%. I will try doing the check in maintenance mode. Unfortunately I need the docker running.... so I will have to find a good time-slot. 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.