Pass 1: Checking inodes, blocks, and sizes $ e2fsck -fy /dev/zvol/filling/store/nabijaczleweli/e2testĮxt2fs_open2: Superblock checksum does not match superblockĮ2fsck: Superblock invalid, trying backup blocks. $ tune2fs -f -E clear_mmp /dev/zvol/filling/store/nabijaczleweli/e2test dev/zvol/filling/store/nabijaczleweli/e2test: ********** WARNING: Filesystem still has errors ********** Mmp_device_name: /dev/zvol/filling/store/nabijacz 'tune2fs -f -E clear_mmp /dev/zvol/filling/store/nabijaczleweli/e2test' MMP check failed: If you are sure the filesystem is not in use on any node, run: $ e2fsck -f /dev/zvol/filling/store/nabijaczleweli/e2testĮ2fsck: MMP: e2fsck being run while checking MMP block Please run 'e2fsck -f /dev/zvol/filling/store/nabijaczleweli/e2test' first. $ resize2fs /dev/zvol/filling/store/nabijaczleweli/e2test To fail, and they manage to fallocate(2)/fsync(2) most of the blockdev The resize2fs steps take inordinately long times Transcripts follow, all with sid e2fsprogs (1.46.2-1) The same block bitmap twice (attached, may be useful?). I've looped e2fsck -f for over 260 runs and never got Immediately on invalid MMP blocks (e2fsprogs from Buster fail slightlyĭifferently, in that they raze all superblocks, but the end effect This doesn't fix it either, and subsequent resize2fs runs bail out Resize2fs (and then other programs from the suite then suggest): I'd expect any enlargement to work, but we hit it with 4T originally).ĭoing e2fsck -f on the original-size blockdev results in a functioningĭestroys the filesystem seemingly permanently. (I just tested with 10T, but 4T should also work The image is then applied to a larger device Generated from e2image -Q of a snapshot generated with the proxmox UI, Please consider this image (21M compressed, 1.2G data over 1T volume),
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |