Description of problem: while trying to reproduce bug #143448, I ran into the following: DLM <CVS> (built Dec 13 2004 15:02:57) installed kmem_cache_create: duplicate cache dlm_conn ------------[ cut here ]------------ kernel BUG at mm/slab.c:1453! invalid operand: 0000 [#1] Modules linked in: dlm(U) parport_pc lp parport autofs4 i2c_dev i2c_core cman(U) sunrpc md5 ipv6 dm_mod button battery ac uhci_hcd hw_random e1000 floppy ext3 jbd qla2300 qla2xxx scsi_transport_fc sd_mod scsi_mod CPU: 0 EIP: 0060:[<c014997f>] Tainted: GF VLI EFLAGS: 00010202 (2.6.9-1.906_EL) EIP is at kmem_cache_create+0x417/0x48a eax: c0312877 ebx: f68fa480 ecx: c03fca80 edx: f8a0c462 esi: f8a0c46b edi: f8a0c46b ebp: f68fa608 esp: f4ef6ea4 ds: 007b es: 007b ss: 0068 Process clvmd (pid: 17708, threadinfo=f4ef6000 task=f4d37730) Stack: 23499435 c0000000 f4d37730 f8a0c462 00000080 00000000 00000000 f71fec80 fffffff4 f8a01c14 00000000 00000000 00000000 1d244b3c 00000000 0000000a f8a0c3fb 00000000 00000000 00000000 f6210a50 00000000 f6afa300 00000005 Call Trace: [<f8a01c14>] lowcomms_start+0x21a/0x2c5 [dlm] [<f89fdec0>] threads_start+0x20/0x3e [dlm] [<f89fdeff>] init_internal+0x17/0x30 [dlm] [<f89fe7f0>] dlm_new_lockspace+0x38/0x5f [dlm] [<f89f726c>] register_lockspace+0xa3/0x149 [dlm] [<f89f7cb3>] do_user_create_lockspace+0x21/0x32 [dlm] [<f89f8e29>] dlm_write+0x153/0x1ab [dlm] [<c01623a9>] vfs_write+0xb6/0xe2 [<c0162473>] sys_write+0x3c/0x62 [<c0301bfb>] syscall_call+0x7/0xb Code: eb 04 19 c0 0c 01 85 c0 75 29 ff 74 24 0c 68 77 28 31 c0 e8 d9 66 fd ff 58 b9 80 ca 3f c0 5a ff 05 80 ca 3f c0 0f 8e 71 1b 00 00 <0f> 0b ad 05 c2 27 31 c0 8b 6d 00 eb 86 8b 54 24 04 b8 00 f0 ff Version-Release number of selected component (if applicable): ccs-0.9-0 kernel-2.6.9-1.906_EL dlm-kernel-2.6.9-3.1 kernel-2.6.9-1.641_EL kernel-utils-2.4-13.1.37 cman-kernel-2.6.9-3.3 fence-1.3-1 dlm-1.0-0.pre9.1 cman-1.0-0.pre5.0 GFS-kernel-2.6.9-4.2 How reproducible: I've not yet tried yet Steps to Reproduce: 1. start and stop the clvmd init.d script a few times 2. 3. Actual results: Expected results: Additional info:
I'm not really sure what's causing this. I can only assume that the "dlm_conn" slab cache is not being deleted when lowcomms shuts down (presumably because it's still got things in it). This might help: Checking in lowcomms.c; /cvs/cluster/cluster/dlm-kernel/src/lowcomms.c,v <-- lowcomms.c new revision: 1.22.2.3; previous revision: 1.22.2.2 done If anyone needs to reopen this, can you include any other kernel messages above the BUG() if there are any.
*** This bug has been marked as a duplicate of 157295 ***