Dear Forum members, I made a copy from one chroot to another chroot - the /home/onechroot/usr/lib/ to /home/secondchroot/usr/lib/ and I lost my SSH connection while the copy went through. Now I can not reach my server via SSH so I asked Hetzner Online to give me KVM switch access. I see "Bus error" message even if I use the 'php -v' command. Services are not working, like apache2, dovecot, pure-ftpd-mysql. I get this: Source Code failed /usr/sbin/apache2ctl : line 138: 14489 Bus error $HTTPD ${APACHE_ARGUMENTS} -t This is Debian 8.11 server I asked Hetzner to check the hard drives in the rescue system. They did it and sent me this answer: "The HDD check just finished and didn't find any problems. So, we have updated both the BIOS of the server and the firmware of the HDDs and booted the system again normally." I created a backup with Bacula from my filesystem, but I get Bus error message for bconsole command also Can you help me to solve this problem? Thank you in advance, Gyula Domonkos
Some more log files for debugging: smartctl-sda: https://pastebin.com/ie7VP8Ey smartctl-sdb: https://pastebin.com/eujWwUtb dmesg: https://pastebin.com/0xwvGUqf kern.log: https://pastebin.com/YGFEyJyJ messsages: https://pastebin.com/mmCXNGXJ
I was able to restart the server in Rescue system. I started a memtest with the command below: memtester 28G Result in attached image.
Strange that would affect SSH connections. How was that copy done? Do you remember the exact command? Rather old. I do not remembe if Debian 8 already had systemctl. So some commands may not work. You can access the computer from console with that KVM thingy? What shows Code: who -r If the host is not runlevel 5 it may not have started services. Do you get that bus error on any command you issue on the console? If yes, I would suspect hardware failure. Have you read the logs you posted, there seem to be errors the. Your disk at least seems to make errors.
That is unlikely to cause the issue. However, it could be it was triggered by high IO load You have a drive related error in your dmesg: Code: [ 1.844163] ACPI Error: [DSSP] Namespace lookup failure, AE_NOT_FOUND (20110623/psargs-359) [ 1.844330] ACPI Error: Method parse/execution failed [\_SB_.PCI0.SAT0.SPT1._GTF] (Node ffff88081add9fb0), AE_NOT_FOUND (20110623/psparse-536) There are other people around with similar errors. https://lists.debian.org/debian-user/2017/04/msg00698.html Most said its gone with a bios update. But if that was no solution? (Have you shut down the server and restarted? Not just a simple restart (because this will not restart kvm)) Also: Code: Mar 9 21:12:52 grey kernel: [16342.853544] EXT4-fs error (device loop0): ext4_mb_generate_buddy:758: group 8302, block bitmap and bg descriptor inconsistent: 1038 vs 1035 free clusters Mar 9 21:12:52 grey kernel: [16342.922245] EXT4-fs error (device loop0): ext4_mb_generate_buddy:758: group 8305, block bitmap and bg descriptor inconsistent: 6738 vs 6727 free clusters I have not had this errors before and can't help (except from searching google and you can do that by yourself) Have you changed something in kvm configuration lately? Like drivers, settings, etc... You can try to set acpi to force in grub config. Did the server run before without problems? If you have a backup, just reformat and start over.
