Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hi, I just tested XFS over drbd, 0.7 pre7 cvs taken at 4 pm, protocol was B, Kernel is 2.6.6. It crashed while bonnie++ was writing on it, after the reboot it synced, I made my primary primary again and mounted the filesystem and ended up like this: XFS mounting filesystem drbd0 Ending clean XFS mount for filesystem: drbd0 Bad page state at free_hot_cold_page (in process 'drbd0_receiver', page c16e5a20) flags:0x20000080 mapping:00000000 mapped:0 count:0 Backtrace: Call Trace: [<c014273f>] bad_page+0x70/0x95 [<c0142eac>] free_hot_cold_page+0x73/0x138 [<c03a6d32>] skb_release_data+0x6b/0xb4 [<c03a6d8b>] kfree_skbmem+0x10/0x26 [<c03a6e13>] __kfree_skb+0x72/0xe0 [<c03ceb98>] tcp_clean_rtx_queue+0xf1/0x3d0 [<c03cf513>] tcp_ack+0xd0/0x59e [<c04067f5>] schedule_timeout+0xb1/0xb3 [<c03a6352>] __release_sock+0x59/0x91 [<c03d1f5a>] tcp_rcv_established+0x1d4/0x805 [<c03c9717>] tcp_data_wait+0xa1/0xb5 [<c03db81e>] tcp_v4_do_rcv+0x12e/0x133 [<c03c9775>] tcp_prequeue_process+0x4a/0x8c [<c03c9e4a>] tcp_recvmsg+0x4ba/0x7ca [<c03ec0e9>] inet_recvmsg+0x54/0x69 [<c03ec0e9>] inet_recvmsg+0x54/0x69 [<c03a2fb4>] sock_recvmsg+0xb4/0x108 [<c01658c1>] bio_hw_segments+0x27/0x29 [<c031b029>] elv_merged_request+0x1f/0x23 [<c031e726>] generic_make_request+0x12c/0x1e2 [<f8a132ea>] drbd_recv+0xa6/0x1d0 [drbd] [<f8a14d80>] receive_DataRequest+0x293/0x2fe [drbd] [<f8a13a21>] drbd_recv_header+0x2b/0xf0 [drbd] [<f8a15e1c>] drbdd+0x1b/0xbf [drbd] [<f8a16688>] drbdd_init+0xc9/0x163 [drbd] [<f8a0baff>] drbd_thread_setup+0x85/0x117 [drbd] [<f8a0ba7a>] drbd_thread_setup+0x0/0x117 [drbd] [<c0104275>] kernel_thread_helper+0x5/0xb Trying to fix it up, but a reboot is needed And so on. -- Stephan Rattai University of Hamburg Centre for Molecular Neurobiology Hamburg (ZMNH) IT services Falkenried 94 20251 Hamburg Germany Tel.:040 - 428 - 03 - 8592 Fax.:040 - 428 - 03 - 6621