flixxx Posted February 4 Share Posted February 4 Monthly parity check this month (1st of every month) had thousands of errors it corrected. I shutdown the server, checked the cable connections to be sure. Rebooted and ran a parity check (without sync) and it found an additional 4 errors - what could it be? attached diagnostics kenny-diagnostics-20240204-0647.zip Quote Link to comment
JorgeB Posted February 4 Share Posted February 4 Run another check to see if it finds the same errors, you also need to check filesystem on disk3, run it without -n Quote Link to comment
itimpi Posted February 4 Share Posted February 4 55 minutes ago, flixxx said: Monthly parity check this month (1st of every month) had thousands of errors it corrected. We normally recommend that scheduled checks are set non-correcting so that hardware playing up (that you have not yet noticed) does not end up corrupting parity. Quote Link to comment
flixxx Posted February 4 Author Share Posted February 4 13 minutes ago, JorgeB said: Run another check to see if it finds the same errors, you also need to check filesystem on disk3, run it without -n I will run another check. The last time i checked the filesystem it made a huge mess of my array - bunch of files got lost, etc.... How can I ensure i don't lose anything? 6 minutes ago, itimpi said: We normally recommend that scheduled checks are set non-correcting so that hardware playing up (that you have not yet noticed) does not end up corrupting parity. Ok, i can certainly do that after i fix this issue Quote Link to comment
JorgeB Posted February 4 Share Posted February 4 9 minutes ago, flixxx said: How can I ensure i don't lose anything? You can't, but most times there's no data loss, though it will depend on how serious the corruption is. Quote Link to comment
flixxx Posted February 5 Author Share Posted February 5 (edited) 11 hours ago, JorgeB said: You can't, but most times there's no data loss, though it will depend on how serious the corruption is. I ran a second parity check and once again it has thousands of errors. I ran a filesystem check (with the -n), and then without the -n and the following was returned (with -n): Phase 1 - find and verify superblock... Phase 2 - using internal log - zero log... - scan filesystem freespace and inode maps... - found root inode chunk Phase 3 - for each AG... - scan (but don't clear) agi unlinked lists... - process known inodes and perform inode discovery... - agno = 0 - agno = 1 - agno = 2 - agno = 3 - process newly discovered inodes... Phase 4 - check for duplicate blocks... - setting up duplicate extent list... - check for inodes claiming duplicate blocks... - agno = 0 - agno = 1 - agno = 2 - agno = 3 No modify flag set, skipping phase 5 Phase 6 - check inode connectivity... - traversing filesystem ... - traversal finished ... - moving disconnected inodes to lost+found ... Phase 7 - verify link counts... No modify flag set, skipping filesystem flush and exiting. I am running another parity check now, here is the latest diagnostics kenny-diagnostics-20240204-1954.zip Edited February 5 by flixxx Quote Link to comment
trurl Posted February 5 Share Posted February 5 This is probably going to turn out to be bad RAM Have you done memtest? Quote Link to comment
flixxx Posted February 5 Author Share Posted February 5 17 hours ago, trurl said: This is probably going to turn out to be bad RAM Have you done memtest? I have not, i will try that the moment the parity ends. I do have a silly issue, since i run this headless. The last time i plugged my monitor in, it works when it's booted up, but during boot up (BIOS, POST, etc...) the screen doesn't work so i never see the menu to run memtest. I never bothered to try to fix it - any recommendation on what to look for? Quote Link to comment
trurl Posted February 5 Share Posted February 5 Get memtest86.com onto another flash drive to boot from. 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.