<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns:st1="urn:schemas-microsoft-com:office:smarttags" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv=Content-Type content="text/html; charset=us-ascii"> <meta name=Generator content="Microsoft Word 11 (filtered medium)"> <o:SmartTagType namespaceuri="urn:schemas-microsoft-com:office:smarttags" name="City" downloadurl="http://www.5iamas-microsoft-com:office:smarttags"/> <o:SmartTagType namespaceuri="urn:schemas-microsoft-com:office:smarttags" name="place" downloadurl="http://www.5iantlavalamp.com/"/>  <style>  </style> </head> <body lang=ZH-CN link=blue vlink=purple style='text-justify-trim:punctuation'> <div class=Section1 style='layout-grid:15.6pt'> Hi, <o:p></o:p> <o:p> </o:p> Thanks for the team working on the device mapper driver.<o:p></o:p> I have installed Fedora Core 4 and downloaded latest device mapper patches from the website <a href="http://www.kernel.org/pub/linux/kernel/people/agk/patches/2.6/editing/">http://www.kernel.org/pub/linux/kernel/people/agk/patches/2.6/editing/</a> and applied them to the kernel 2.6.16-rc1, but I still experience problems when removing snapshots if there are large I/Os and tens of snapshots have been created. Before applying those patches to the kernel, kernel panic even when removing the only one snapshot created for a <st1:City w:st="on">LV</st1:City> if there are large I/Os on the <st1:place w:st="on"><st1:City w:st="on">LV.</st1:City></st1:place><o:p></o:p> <o:p> </o:p> The steps to produce the issue:<o:p></o:p> <o:p> </o:p> <![if !supportLists]>1. <![endif]>Create PV on the disk partition directly <o:p></o:p> <![if !supportLists]>2. <![endif]>Create VG on the PV just created<o:p></o:p> <![if !supportLists]>3. <![endif]>Create a 20G <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place> on the VG<o:p></o:p> <![if !supportLists]>4. <![endif]>Create XFS file system on the <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place><o:p></o:p> <![if !supportLists]>5. <![endif]>Mount the <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place><o:p></o:p> <![if !supportLists]>6. <![endif]>Run a script to generate large I/Os on the <st1:place w:st="on"><st1:City w:st="on">LV.</st1:City></st1:place> The job will keep running till interrupted by me.<o:p></o:p> <![if !supportLists]>7. <![endif]>2 snapshots have been created for the <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place> successfully<o:p></o:p> <![if !supportLists]>8. <![endif]>All snapshots have been removed one by one successfully<o:p></o:p> <![if !supportLists]>9. <![endif]>21 snapshots have been created for the <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place> successfully<o:p></o:p> <![if !supportLists]>10. <![endif]>Try to remove all snapshots by (“lvremove –f /dev/vg/snap*”).<o:p></o:p> I have seen following messages in the system log before kernel panic information were logged.<o:p></o:p> Feb 15 15:39:43 sit-fc4 kernel: attempt to access beyond end of device<o:p></o:p> Feb 15 15:39:43 sit-fc4 kernel: dm-14: rw=17, want=4194320, limit=4194304<o:p></o:p> <o:p> </o:p> Kernel panic<o:p></o:p> Kernel panic information:<o:p></o:p> <o:p> </o:p> Feb 15 15:40:22 sit-fc4 kernel: Unable to handle kernel paging request at virtual address 8b8a8988<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: printing eip:<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: f8abe0ab<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: *pde = 9b9a9998<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: Oops: 0000 [#1]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: SMP<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: Modules linked in: smbfs xfs exportfs dm_zero dm_snapshot dm_mod ipv6 parport_pc lp parport autofs4<o:p></o:p> rfcomm l2cap bluetooth sunrpc video button battery ac uhci_hcd i2c_i801 i2c_core shpchp e1000 ext3 jbd sata_mv libata aic79xx scsi_t<o:p></o:p> ransport_spi sd_mod scsi_mod<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: CPU: 0<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: EIP: 0060:[<f8abe0ab>] Not tainted VLI<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: EFLAGS: 00010212 (2.6.16rc2)<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: EIP is at exit_exception_table+0x3b/0x70 [dm_snapshot]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: eax: 00000046 ebx: 8b8a8988 ecx: f6fa38c0 edx: dde13988<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: esi: f8bf6ec0 edi: f6fa38c0 ebp: 000001d8 esp: dfb28da4<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: ds: 007b es: 007b ss: 0068<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: Process lvremove (pid: 4057, threadinfo=dfb28000 task=c85e4aa0)<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: Stack: <0>e66b84d8 00004000 e66b8480 d8b097c0 f8ac3700 f8a3e080 f8abe619 f8a3e080<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: e66b8280 00000000 00000000 f8aaff16 00000004 e66b8280 f59e68e0 00000004<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: f8ab2a00 f8ab1f33 f8abb6ec f8ad8000 f8ab2a44 00000000 00000000 f8ad8000<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: Call Trace:<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<f8abe619>] snapshot_dtr+0x99/0x110 [dm_snapshot]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<f8aaff16>] dm_table_put+0x56/0xd0 [dm_mod]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<f8ab2a00>] dev_remove+0x0/0x90 [dm_mod]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<f8ab1f33>] __hash_remove+0x63/0x80 [dm_mod]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<f8ab2a44>] dev_remove+0x44/0x90 [dm_mod]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<f8ab2105>] ctl_ioctl+0x1b5/0x2c0 [dm_mod]<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<c0150607>] __alloc_pages+0x57/0x320<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<c017f6fa>] do_ioctl+0x8a/0xa0<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<c017f76c>] vfs_ioctl+0x5c/0x2f0<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<c0106b2a>] do_syscall_trace+0x21a/0x240<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<c017fa72>] sys_ioctl+0x72/0x90<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: [<c0104205>] syscall_call+0x7/0xb<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: Code: d2 89 54 24 04 7e 43 8b 40 04 31 ed 8d 74 26 00 8d 34 e8 8b 16 39 d6 8b 1a 75 07 eb 1a 8d 76 0<o:p></o:p> 0 89 c3 89 f8 e8 a7 ab 6a c7 39 de <8b> 03 89 da 75 ef 8b 14 24 8b 42 04 45 39 6c 24 04 75 d2 83 c4<o:p></o:p> Feb 15 15:40:22 sit-fc4 kernel: <0>Fatal exception: panic in 5 seconds<o:p></o:p> <![if !supportLists]>11. <![endif]>Reboot the system, then remove all snapshots<o:p></o:p> <![if !supportLists]>12. <![endif]>Repeat step 6 - Run a script to generate large I/Os on the <st1:place w:st="on"><st1:City w:st="on">LV.</st1:City></st1:place> The job will keep running till interrupted by me.<o:p></o:p> <![if !supportLists]>13. <![endif]>5 snapshots have been created for the <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place> successfully<o:p></o:p> <![if !supportLists]>14. <![endif]>All snapshots have been removed by (“lvremove –f /dev/vg/snap*”) successfully.<o:p></o:p> <![if !supportLists]>15. <![endif]>9 snapshots have been created for the <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place> successfully<o:p></o:p> <![if !supportLists]>16. <![endif]>Try to remove all snapshots by (“lvremove –f /dev/vg/snap*”).<o:p></o:p> <![if !supportLists]>17. <![endif]>Kernel panic again<o:p></o:p> <![if !supportLists]>18. <![endif]>5 snapshots have been created for the <st1:place w:st="on"><st1:City w:st="on">LV</st1:City></st1:place> successfully<o:p></o:p> <![if !supportLists]>19. <![endif]>All snapshots have been removed by (“lvremove –f /dev/vg/snap*”) successfully.<o:p></o:p> <![if !supportLists]>20. <![endif]><o:p> </o:p> <o:p> </o:p> Juer<o:p></o:p> </div> </body> </html>