From mboxrd@z Thu Jan 1 00:00:00 1970 From: Noyan Ilmen Subject: dead inode read from disk [1 2 0x0 SD]. This is likely to be race with knfsd. Ignore Date: Tue, 11 Mar 2003 19:43:43 +0100 Message-ID: <30370000.1047408223@bidolor> Reply-To: Noyan Ilmen Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="==========877300887==========" Return-path: list-help: list-unsubscribe: list-post: Errors-To: flx@namesys.com List-Id: To: reiserfs-list@namesys.com --==========877300887========== Content-Type: text/plain; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Ignore - ha ha very funny Ignore means: can't mount ("no mount dir" or somethimg like that) I've got 80GB Data lost. nfs with reiserfs seems to be still buggy and corrupts data. What happens: day 1: (syslog)(see attachment 1) Mar 9 21:31:16 outa kernel: vs-5150: search_by_key: invalid format found in block 7843968. Fsck? Mar 9 21:31:16 outa kernel: vs-13070: reiserfs_read_inode2: i/o failure occurred trying to find stat data of [220650 220800 0x0 SD] Mar 9 21:31:17 outa kernel: vs-5150: search_by_key: invalid format found in block 7843968. Fsck? Mar 9 21:31:17 outa kernel: vs-13070: reiserfs_read_inode2: i/o failure occurred trying to find stat data of [220650 220805 0x0 SD] . . . (reiserfsck --check)(see attachment 2) node (7843968) with wrong level (44655) found in the tree (should be 1) whole subtree skipped free block count 3902051 mismatches with a correct one 3953043. on-disk bitmap does not match to the correct one. eh??? (h=E4???) (reiserfsck --fix-fixable)(see attachment 3) node (7843968) with wrong level (44655) found in the tree (should be 1) whole subtree skipped free block count 3902051 mismatches with a correct one 3953043. on-disk bitmap does not match to the correct one. eh??? (h=E4???) this above takes me hours: rtfms, null-info-reiser-docs, net,... what the hell is "...reiserfsck returns 1 to indicate..." Do you think I write functions to eval your return values? shutdown day 2: can't mount local reiser disk much more hours (reiserfsck --check)(see attachment 4) bad_stat_data: block 9160, [1 2 0x0 SD (0)], directory SD has bad nlink number bad_stat_data: block 9160, [2 4 0x0 SD (0)], SD has bad nlink number . 205300 lines later . bad_stat_data: block 22978597, [310954 310965 0x0 SD (0)], SD has bad nlink number free block count 3902037 mismatches with a correct one 3953043. on-disk bitmap does not match to the correct one. (reiserfsck --rebuild-tree)(see attachment 5/6) reed your warnings about your buggy-beta-loose-all-data-software and say "Yes", because there wasn't any more to backup. Btw it takes about 3 days to backup the 80G with my DDS3 streamer. ####### Pass 0 ####### "r5" got 206443 hits ####### Pass 1 ####### ####### Pass 2 ####### are_file_items_correct: convert the indirect item back to direct 208191 208307 ####### Pass 3 ######### are_file_items_correct: convert the indirect item back to direct 67929 7778 . . . are_file_items_correct: convert the indirect item back to direct 258907 258912 ####### Pass 3a (lost+found pass) ######### takes about 2 hours and I was able to mount the partition. If there is lost data? I don't know, but it would not be the first time I lost data with reiserfs. system-setup: server with: debian sarge kernel 2.4.20-686 / ReiserFS version 3.6.25 reiserfsprogs 3.6.3-1 nfs-kernel-server 1.0.2-1 samba 2.2.3a-12 client with: debian sarge kernel 2.4.19-smp nfs-kernel-server 1.0.2-1 samba 2.2.3a-6 Please don't annoy me with mails. my only desire is to convert to ext2/3. Spricht hier jemand deutsch? Noyan --==========877300887==========--