Re: reiserfs crash
Hello On Tue, 2006-05-09 at 11:16 +0200, Devel wrote: Il Sat, 06 May 2006 11:43:58 +0400 Vladimir V. Saveliev [EMAIL PROTECTED] scrisse: Hello On Fri, 2006-05-05 at 16:34 +0200, Devel wrote: Il Fri, 05 May 2006 10:43:26 +0400 Vladimir V. Saveliev [EMAIL PROTECTED] scrisse: Hello On Thu, 2006-05-04 at 19:06 +0200, Devel wrote: Hi All, i'm testing reiser4 on a linux box kernel 2.6.16. This linux box write a lot of images on the partiion with reiser4 and after delete them. After a while all goes wrong and dmesg give me this oops: 4reiser4[image_eraser.pl(2374)]: cbk_level_lookup (fs/reiser4/search.c:971)[vs-3533]: WARNING: Keys are inconsistent. Fsck? 4reiser4[image_eraser.pl(2374)]: key_warning (fs/reiser4/plugin/file_plugin_common.c:514)[nikita-717]: WARNING: Error for inode 47326534 (-5) Unable to handle kernel NULL pointer 5 is error code indicating i/o error: disk block could not be read or written from/to a device. That is why I guessed that the harddrive is not reliable. I make an fsck on the partition and now the disk seem work fine. May be the disk was inconsistent? It would be interesting to see whether the problem will come up again under the same load. dereference at virtual address printing eip: *pde = Oops: [#1] Modules linked in: bttv video_buf firmware_class compat_ioctl32 i2c_algo_bit v4l2_common btcx_risc ir_common tveeprom i2c_core videodev video CPU:0 EIP:0060:[]Not tainted VLI EFLAGS: 00010282 (2.6.16.5 #1) EIP is at rest_init+0x3feffde0/0x1e eax: ebx: d80c3d84 ecx: da670afc edx: c03ee8e0 esi: edi: ebp: c01b74f1 esp: d80c3b58 ds: 007b es: 007b ss: 0068 Process image_eraser.pl (pid: 2374, threadinfo=d80c2000 task=df863a30) Stack: 0c01b74b5 d80c3d84 da670afc d80c3e38 d80c3bbc d80c3bbc c01b746c d80c3c18 c01b750e d80c3d84 da670afc d80c3e38 d80c3bbc c01b77a4 d80c3d84 da670afc d80c3e38 d80c3bbc 0002 Call Trace: [c01b74b5] kill_units+0x49/0x53 [c01b746c] kill_units+0x0/0x53 [c01b750e] kill_head+0x1d/0x24 [c01b77a4] prepare_for_compact+0x1e2/0x406 [c01a3c00] reiser4_get_neighbor+0x75/0x261 [c01903ea] jload_gfp+0x112/0x124 [c01b79eb] kill_node40+0x23/0x9a [c0192582] lock_carry_node_tail+0x16/0x18 [c0193d76] carry_cut+0x3f/0x53 [c0191f7c] carry_on_level+0x30/0xaa [c0191e5c] carry+0x79/0x169 [c01960a7] kill_node_content+0x125/0x13e [c0196591] cut_tree_worker_common+0x196/0x2e8 [c01963fb] cut_tree_worker_common+0x0/0x2e8 [c0196791] cut_tree_object+0xae/0x149 [c01b6aba] create_item_node40+0x1fc/0x258 [c0199c4e] znode_make_dirty+0x40/0x50 [c01abad2] cut_file_items+0xdb/0x174 [c01abba4] shorten_file+0x29/0x1d7 [c01ab996] update_file_size+0x0/0x61 [c01abe0b] truncate_file_body+0x63/0x6f [c01ae1ac] delete_object_unix_file+0x33/0xa6 [c01a23ae] reiser4_delete_inode+0x83/0x9d [c01a232b] reiser4_delete_inode+0x0/0x9d [c0155796] generic_delete_inode+0x56/0xb4 [c015597c] iput+0x63/0x66 [c014e22e] do_unlinkat+0xb4/0xf9 [c014e2ae] sys_unlink+0xb/0xe [c0102977] sysenter_past_esp+0x54/0x75 Code: Bad EIP value. If i reboot the machine this error repeat again! i have to do fsck to resolve this error? thanks It looks like your harddrive is not very reliable. Would you please try to experiment with another box/harddrive? The harddrive is a maxtor so i tested it with the powermax maxtor harddrive tools with success. More i run the smart control long test with success (smartctl -t long /dev/hda) . Why you say It looks like your harddrive is not very reliable ? ok. But nevertheless I would like you to try to reproduce the problem on another hardware to be sure that it is a software bug. I will make onother one linux box to reproduce errors!! great, thanks
Re: Reproducible reiser4 bug with 2.6.16.2 patch on tail_conversion.c:80
Hello please check whether the attached patch helps. On Wednesday 10 May 2006 14:56, Timo Kokkonen wrote: Hello, I was able to produce this bug on two different computers running the same kernel version and same patches and same task. Each time I run dispatch-conf (I'm running on gentoo) to update the config files on /etc, it gives the following output: RCS file: /etc/config-archive/usr/lib/X11/xkb/xkbcomp,v 1.1 locked ci: RCS file /etc/config-archive/usr/lib/X11/xkb/xkbcomp,v is in use /etc/config-archive/usr/lib/X11/xkb/xkbcomp,v -- /etc/config-archive/usr/lib/X11/xkb/xkbcomp revision 1.1.1.1 writable /etc/config-archive/usr/lib/X11/xkb/xkbcomp exists; remove it? [ny](n): And right after that I can see this message in kernel log: May 10 11:26:32 kaatokone2 [4296234.613000] [ cut here ] May 10 11:26:32 kaatokone2 [4296234.613000] kernel BUG at fs/reiser4/plugin/file/tail_conversion.c:80! May 10 11:26:32 kaatokone2 [4296234.613000] invalid opcode: [#1] May 10 11:26:32 kaatokone2 [4296234.613000] PREEMPT May 10 11:26:32 kaatokone2 [4296234.613000] Modules linked in: w83627hf w83781d hwmon_vid eeprom i2c_isa May 10 11:26:32 kaatokone2 [4296234.613000] CPU:0 May 10 11:26:32 kaatokone2 [4296234.613000] EIP:0060:[b019ec1f] Not tainted VLI May 10 11:26:32 kaatokone2 [4296234.613000] EFLAGS: 00210286 (2.6.16-gentoo-r7 #1) May 10 11:26:32 kaatokone2 [4296234.613000] EIP is at get_nonexclusive_access+0x23/0x36 May 10 11:26:32 kaatokone2 [4296234.613000] eax: ef7644c4 ebx: ecx: a7e8b000 edx: e5d0c414 May 10 11:26:32 kaatokone2 [4296234.613000] esi: e5d0c414 edi: e5d0c46c ebp: e5d59e9c esp: e5d59dc4 May 10 11:26:32 kaatokone2 [4296234.613000] ds: 007b es: 007b ss: 0068 May 10 11:26:32 kaatokone2 [4296234.613000] Process rcs (pid: 8422, threadinfo=e5d58000 task=e5cf2ab0) May 10 11:26:32 kaatokone2 [4296234.613000] Stack: 0b01b18e9 e5d0c414 3400 e5d58000 b16ba260 May 10 11:26:32 kaatokone2 [4296234.613000]e5eb8780 b01b1ccd e5d0c46c e5d59e9c eed8e8c0 eed8e8d0 eed8e8d0 0002 May 10 11:26:32 kaatokone2 [4296234.613000]0001 e5d59e9c 1000 0001 efeea800 efded414 00391f78 May 10 11:26:32 kaatokone2 [4296234.613000] Call Trace: May 10 11:26:32 kaatokone2 [4296234.613000] [b01b18e9] extent_balance_dirty_pages+0x13a/0x145 May 10 11:26:32 kaatokone2 [4296234.613000] [b01b1ccd] extent_write_flow+0x3d9/0x4e7 May 10 11:26:32 kaatokone2 [4296234.613000] [b01b1f0a] write_extent+0x0/0x36 May 10 11:26:32 kaatokone2 [4296234.613000] [b019dc17] append_and_or_overwrite+0x24f/0x2fd May 10 11:26:32 kaatokone2 [4296234.613000] [b019dd0c] write_flow+0x47/0x50 May 10 11:26:32 kaatokone2 [4296234.613000] [b019de92] write_file+0x5c/0x7a May 10 11:26:32 kaatokone2 [4296234.613000] [b019e177] write_unix_file+0x2c7/0x444 May 10 11:26:32 kaatokone2 [4296234.613000] [b014bd9e] vfs_write+0x87/0x11b May 10 11:26:32 kaatokone2 [4296234.613000] [b014bed0] sys_write+0x3b/0x63 May 10 11:26:32 kaatokone2 [4296234.613000] [b0102a93] sysenter_past_esp+0x54/0x75 May 10 11:26:32 kaatokone2 [4296234.613000] Code: 0f 85 b5 0d 00 00 c3 c3 83 7c 24 08 00 8b 54 24 04 75 20 b8 00 e0 ff ff 21 e0 8b 00 8b 80 b4 04 00 00 8b 40 40 83 78 08 00 74 08 0f 0b 50 00 ba cf 43 b0 89 d0 ff 00 0f 88 92 0d 00 00 c3 8b 54 May 10 11:26:32 kaatokone2 [4296234.613000] 44reiser4[rcs(8422)]: release_unix_file (fs/reiser4/plugin/file/file.c:2670)[vs-44]: May 10 11:26:32 kaatokone2 [4296234.614000] WARNING: out of memory? May 10 11:26:32 kaatokone2 [4296234.614000] 4reiser4[rcs(8422)]: release_unix_file (fs/reiser4/plugin/file/file.c:2670)[vs-44]: May 10 11:26:32 kaatokone2 [4296234.614000] WARNING: out of memory? May 10 11:26:32 kaatokone2 [4296234.614000] 4reiser4[rcs(8422)]: release_unix_file (fs/reiser4/plugin/file/file.c:2670)[vs-44]: May 10 11:26:32 kaatokone2 [4296234.614000] WARNING: out of memory? May 10 11:26:32 kaatokone2 [4296234.614000] 4reiser4[rcs(8422)]: release_unix_file (fs/reiser4/plugin/file/file.c:2670)[vs-44]: May 10 11:26:32 kaatokone2 [4296234.614000] WARNING: out of memory? May 10 11:26:32 kaatokone2 [4296234.614000] 4reiser4[rcs(8422)]: release_unix_file (fs/reiser4/plugin/file/file.c:2670)[vs-44]: May 10 11:26:32 kaatokone2 [4296234.614000] WARNING: out of memory? After a reboot and running dispatch-conf again I get the exact same error message again. I got this same bug on the another computer aswell, during exact same conditions (running dispatch-conf, it even complained about the same filename). Running reiser4.fsck didn't reveal any problems at all on that computer. Booting with 2.6.15.1 (with gentoo-sources-2.6.15-r1 patches) and reiser4-2.6.15.1 didn't crash and dispatch-conf finished updating the files normally. So, now I have this computer that I can make to crash every time by running dispatch-conf. I can keep
Re: Kernel BUG at fs/reiser4/plugin/file/tail_conversion.c:80
Hi, after running for a couple hours, kernel reports a different bug. Then again processes get stuck in D state, and only hard reset helps. Here's this new error message: May 11 07:19:22 titanic kernel: 4reiser4[httpd(11176)]: plugin_by_unsafe_id (fs/reiser4/plugin/plugin.c:296)[nikita-2913]: May 11 07:19:22 titanic kernel: WARNING: Invalid plugin id: [2:235] May 11 07:19:22 titanic kernel: Unable to handle kernel NULL pointer dereference at 0004 RIP: May 11 07:19:22 titanic kernel: 80229461{obtain_item_plugin+17} May 11 07:19:22 titanic kernel: PGD 37d86067 PUD e1ba8067 PMD 0 May 11 07:19:22 titanic kernel: Oops: [1] SMP May 11 07:19:22 titanic kernel: CPU 3 May 11 07:19:22 titanic kernel: Modules linked in: May 11 07:19:22 titanic kernel: Pid: 11176, comm: httpd Not tainted 2.6.16-cks9 #2 May 11 07:19:22 titanic kernel: RIP: 0010:[80229461] 80229461{obtain_item_plugin+17} May 11 07:19:22 titanic kernel: RSP: 0018:8100edd39b48 EFLAGS: 00010292 May 11 07:19:22 titanic kernel: RAX: RBX: 81016df732d0 RCX: 80406c68 May 11 07:19:22 titanic kernel: RDX: RSI: 0292 RDI: 80406c60 May 11 07:19:22 titanic kernel: RBP: 81016df732d0 R08: 0003 R09: 0001 May 11 07:19:22 titanic kernel: R10: R11: 80118cd0 R12: 81013e91f000 May 11 07:19:22 titanic kernel: R13: 80419000 R14: 81013e91f4f0 R15: May 11 07:19:22 titanic kernel: FS: 2ab7a85cc8e0() GS:8101045370c0() knlGS: May 11 07:19:22 titanic kernel: CS: 0010 DS: ES: CR0: 8005003b May 11 07:19:22 titanic kernel: CR2: 0004 CR3: d9e67000 CR4: 06e0 May 11 07:19:22 titanic kernel: Process httpd (pid: 11176, threadinfo 8100edd38000, task 810037d3f100) May 11 07:19:23 titanic kernel: Stack: 81016df732d0 801f14d1 81016df732d0 801f17ad May 11 07:19:23 titanic kernel:0001 801ed372 8100edd39b70 0001 May 11 07:19:23 titanic kernel: May 11 07:19:23 titanic kernel: Call Trace: 801f14d1{coord_num_units+17} 801f17ad{coord_init_after_item_end+13} May 11 07:19:23 titanic kernel: 801ed372{carry_insert_flow+1106} 801eb9ab{carry+267} May 11 07:19:23 titanic kernel: 801eac05{post_carry+85} 801ef267{insert_flow+263} May 11 07:19:23 titanic kernel: 80221de5{write_tail+245} 801e8e35{jload_gfp+437} May 11 07:19:23 titanic kernel: 80211bbe{extent2tail+942} 8020fd40{release_unix_file+192} May 11 07:19:23 titanic kernel:80179d02{__fput+194} 80162a01{remove_vma+65} May 11 07:19:23 titanic kernel: 8016417e{do_munmap+670} 801649e2{sys_munmap+82} May 11 07:19:23 titanic kernel:8010aaf6{system_call+126} May 11 07:19:23 titanic kernel: May 11 07:19:23 titanic kernel: Code: 0f be 40 04 88 43 0c 5b c3 66 66 90 66 66 90 53 0f b6 47 0c May 11 07:19:23 titanic kernel: RIP 80229461{obtain_item_plugin+17} RSP 8100edd39b48 May 11 07:19:23 titanic kernel: CR2: 0004 On 5/11/06, Alexander Zarochentsev [EMAIL PROTECTED] wrote: Hello. please apply the attached patch. -- Alexey Polyakov