2.6.10: PREEMPT SMP
Hello,
We have been experiencing a very similar oops on a Dell poweredge 2600, running Fedora Core 3 (FC3). This host is mainly used as a samba fileserver. The problem has been seen on 2 different machines (both Dell-pe-2600s).
The problem is only ever seen when running the SMP kernel in conjunction with samba.
We have been experiencing problems since early January when the samba/nfs/io load was increased considerably. Since then a number of 2.4 kernels have been used:
2.4.22.2115nptlsmp (FC1)
2.4.22-1.2199smp (FC1)
2.4.29 (vanilla kernel)
2.6.10-1.741_FC3smp (FC3)
2.6.10-1.766_FC3smp (FC3)
All 2.4 kernels produced the same result: full system lock-up, with no messages on the console or the log files.
The 2.6.10 kernel is giving an oops which has been seen 3 time in the last 3 days. A very similar one was seen pointing to eip: c01b2d98 on the 2.6.10-1.741 kernel.
I have included the kallsyms and System.map relevant to the instruction pointed to by the EIP. I can include the full files if required.
printing eip:c01b387c
*pde = 2fcdd001
Oops: 0000 [#1] SMP Modules linked in: nfsd exportfs md5 ipv6 parport_pc lp parport autofs4 nfs lockd sunrpc video button battery ac uhci_hcd hw_random e1000 floppy sg dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod mptscsih mptbase sd_mod scsi_mod CPU: 1 EIP: 0060:[<c01b387c>] Not tainted VLI EFLAGS: 00010203 (2.6.10-1.766_FC3smp)
EIP is at __rb_rotate_left+0x8/0x36
eax: f57be640 ebx: c041e5e4 ecx: f57be640 edx: 00000000
esi: f57be640 edi: f5f39c80 ebp: c041e5e4 esp: efa2eed0
ds: 007b es: 007b ss: 0068
Process smbd (pid: 8536, threadinfo=efa2e000 task=f7612a60)
Stack: f5ac1980 c01b3988 f5f39c80 f5f39c80 f5ac1988 00000267 c01965d7 f5ac1980 0000000d efa2ef54 efa2ef61 ffffffea c019668b 00000015 00000000 00000267 c03191e0 efa2ef54 00000000 f599ed80 00000267 c01977df ffffffff 001f0000 Call Trace: [<c01b3988>] rb_insert_color+0xa8/0xc1 [<c01965d7>] key_user_lookup+0xcf/0xfc [<c019668b>] key_alloc+0x53/0x2b6 [<c01977df>] keyring_alloc+0x1a/0x48 [<c0198d43>] alloc_uid_keyring+0x2b/0x7c [<c0125d2e>] alloc_uid+0xae/0x133 [<c01291d4>] set_user+0xb/0x8b [<c01295f2>] sys_setresuid+0x11a/0x1b9 [<c0103c97>] syscall_call+0x7/0xb Code: 82 04 01 00 00 00 75 ea 41 83 f9 01 76 ed 31 c0 5b c3 57 b9 45 00 00 00 89 c7 31 c0 f3 ab 5f c3 90 90 90 53 89 d3 8b 50 08 89 c1 <8b> 42 0c 85 c0 89 41 08 74 02 89 08 89 4a 0c 8b 01 85 c0 89 02
/proc/kallsyms grepped on c01b3.
c01b3024 t newaryc01b313c T sys_semgetc01b325a t try_atomic_semopc01b336b t update_queuec01b33de t count_semncntc01b342c t count_semzcntc01b347a t freearyc01b34f6 t copy_semid_to_userc01b354b t semctl_nolockc01b3731 t semctl_mainc01b3ad6 t semctl_downc01b3c39 T sys_semctlc01b3ce9 t lookup_undoc01b3d19 t find_undo
root@msh2-c boot]# grep -i c01b3 /boot/System.map-2.6.10-1.766_FC3smp c01b304b T match_intc01b3052 T match_octalc01b305c T match_hexc01b3066 T match_strcpyc01b3098 T match_strdupc01b30c0 t radix_tree_node_allocc01b3109 T radix_tree_preloadc01b317d t radix_tree_extendc01b321c T radix_tree_insertc01b3303 T radix_tree_lookupc01b3349 T radix_tree_tag_setc01b33b8 T radix_tree_tag_clearc01b347a t __lookupc01b3547 T radix_tree_gang_lookupc01b3595 t __lookup_tagc01b369d T radix_tree_gang_lookup_tagc01b36f0 T radix_tree_deletec01b3835 T radix_tree_taggedc01b3863 t radix_tree_node_ctorc01b3874 t __rb_rotate_leftc01b38aa t __rb_rotate_rightc01b38e0 T rb_insert_colorc01b39a1 t __rb_erase_colorc01b3b0c T rb_erasec01b3bcd T rb_firstc01b3be5 T rb_lastc01b3bfd T rb_nextc01b3c26 T rb_prevc01b3c4f T rb_replace_nodec01b3c94 T rwsem_wakec01b3d89 T rwsem_downgrade_wakec01b3e14 T strnicmpc01b3e75 T strcpyc01b3e89 T strncpyc01b3ea8 T strlcpyc01b3eeb T strcatc01b3f0a T strncatc01b3f35 T strlcatc01b3f97 T strcmpc01b3fae T strncmpc01b3fe0 T strchrc01b3ff0 T strrchr
Cheers,
Andrew Taylor
Systems Engineer
Array Systems Computing Inc.
email: [email protected] http: www.array.ca
Quoted 4 lines · Andrew Taylor
Hello,
We have been experiencing a very similar oops on a Dell poweredge 2600, running Fedora Core 3 (FC3). This host is mainly used as a samba fileserver. The problem has been seen on 2 different machines (both Dell-pe-2600s).
Could you pleae confirm that 2.6.12-rc1 fixes this?
Thanks.
Quoted 120 lines
The problem is only ever seen when running the SMP kernel in conjunction with samba.
We have been experiencing problems since early January when the samba/nfs/io load was increased considerably. Since then a number of 2.4 kernels have been used:
2.4.22.2115nptlsmp (FC1)
2.4.22-1.2199smp (FC1)
2.4.29 (vanilla kernel)
2.6.10-1.741_FC3smp (FC3)
2.6.10-1.766_FC3smp (FC3)All 2.4 kernels produced the same result: full system lock-up, with no messages on the console or the log files.
The 2.6.10 kernel is giving an oops which has been seen 3 time in the last 3 days. A very similar one was seen pointing to eip:
c01b2d98on the 2.6.10-1.741 kernel.I have included the kallsyms and System.map relevant to the instruction pointed to by the EIP. I can include the full files if required.
printing eip:
c01b387c
*pde =2fcdd001Oops: 0000 [#1] SMP Modules linked in: nfsd exportfs md5 ipv6 parport_pc lp parport autofs4 nfs lockd sunrpc video button battery ac uhci_hcd hw_random e1000 floppy sg dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod mptscsih mptbase sd_mod scsi_mod CPU: 1 EIP: 0060:[<c01b387c>] Not tainted VLI EFLAGS: 00010203 (2.6.10-1.766_FC3smp)EIP is at
__rb_rotate_left+0x8/0x36
eax:f57be640ebx:c041e5e4ecx:f57be640edx: 00000000
esi:f57be640edi:f5f39c80ebp:c041e5e4esp:efa2eed0
ds: 007b es: 007b ss: 0068
Process smbd (pid: 8536, threadinfo=efa2e000 task=f7612a60)Stack: f5ac1980 c01b3988 f5f39c80 f5f39c80 f5ac1988 00000267 c01965d7 f5ac1980 0000000d efa2ef54 efa2ef61 ffffffea c019668b 00000015 00000000 00000267 c03191e0 efa2ef54 00000000 f599ed80 00000267 c01977df ffffffff 001f0000 Call Trace: [<c01b3988>] rb_insert_color+0xa8/0xc1 [<c01965d7>] key_user_lookup+0xcf/0xfc [<c019668b>] key_alloc+0x53/0x2b6 [<c01977df>] keyring_alloc+0x1a/0x48 [<c0198d43>] alloc_uid_keyring+0x2b/0x7c [<c0125d2e>] alloc_uid+0xae/0x133 [<c01291d4>] set_user+0xb/0x8b [<c01295f2>] sys_setresuid+0x11a/0x1b9 [<c0103c97>] syscall_call+0x7/0xb Code: 82 04 01 00 00 00 75 ea 41 83 f9 01 76 ed 31 c0 5b c3 57 b9 45 00 00 00 89 c7 31 c0 f3 ab 5f c3 90 90 90 53 89 d3 8b 50 08 89 c1 <8b> 42 0c 85 c0 89 41 08 74 02 89 08 89 4a 0c 8b 01 85 c0 89 02
/proc/kallsymsgrepped on c01b3.
c01b3024t newaryc01b313cTsys_semgetc01b325attry_atomic_semopc01b336btupdate_queuec01b33detcount_semncntc01b342ctcount_semzcntc01b347at freearyc01b34f6tcopy_semid_to_userc01b354btsemctl_nolockc01b3731tsemctl_mainc01b3ad6tsemctl_downc01b3c39Tsys_semctlc01b3ce9tlookup_undoc01b3d19tfind_undoroot@msh2-c boot]# grep -i c01b3
/boot/System.map-2.6.10-1.766_FC3smpc01b304bTmatch_intc01b3052Tmatch_octalc01b305cTmatch_hexc01b3066Tmatch_strcpyc01b3098Tmatch_strdupc01b30c0tradix_tree_node_allocc01b3109Tradix_tree_preloadc01b317dtradix_tree_extendc01b321cTradix_tree_insertc01b3303Tradix_tree_lookupc01b3349Tradix_tree_tag_setc01b33b8Tradix_tree_tag_clearc01b347at__lookupc01b3547Tradix_tree_gang_lookupc01b3595t__lookup_tagc01b369dTradix_tree_gang_lookup_tagc01b36f0Tradix_tree_deletec01b3835Tradix_tree_taggedc01b3863tradix_tree_node_ctorc01b3874t__rb_rotate_leftc01b38aat__rb_rotate_rightc01b38e0Trb_insert_colorc01b39a1t__rb_erase_colorc01b3b0cTrb_erasec01b3bcdTrb_firstc01b3be5Trb_lastc01b3bfdTrb_nextc01b3c26Trb_prevc01b3c4fTrb_replace_nodec01b3c94Trwsem_wakec01b3d89Trwsem_downgrade_wakec01b3e14T strnicmpc01b3e75T strcpyc01b3e89T strncpyc01b3ea8T strlcpyc01b3eebT strcatc01b3f0aT strncatc01b3f35T strlcatc01b3f97T strcmpc01b3faeT strncmpc01b3fe0T strchrc01b3ff0T strrchrCheers,
Andrew Taylor
Systems Engineer
Array Systems Computing Inc.email: [email protected] http: www.array.ca-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Hello,
Quoted 7 lines · Andrew Morton
Quoted 4 lines · Andrew Taylor
Hello,
We have been experiencing a very similar oops on a Dell poweredge 2600, running Fedora Core 3 (FC3). This host is mainly used as a samba fileserver. The problem has been seen on 2 different machines (both Dell-pe-2600s).
Could you pleae confirm that 2.6.12-rc1 fixes this?
Would be glad to however, between 2.6.11-5 (which appears to runs fine) and 2.6.12-rc1 the mptfusion driver got broken. Attempts to isolate the mptfusion changes have not been successful. Will look at this issue again on tuesday.
Cheers,
Andrew
Quoted 253 lines
Thanks.
Quoted 120 lines
The problem is only ever seen when running the SMP kernel in conjunction with samba.
We have been experiencing problems since early January when the samba/nfs/io load was increased considerably. Since then a number of 2.4 kernels have been used:
2.4.22.2115nptlsmp (FC1)
2.4.22-1.2199smp (FC1)
2.4.29 (vanilla kernel)
2.6.10-1.741_FC3smp (FC3)
2.6.10-1.766_FC3smp (FC3)All 2.4 kernels produced the same result: full system lock-up, with no messages on the console or the log files.
The 2.6.10 kernel is giving an oops which has been seen 3 time in the last 3 days. A very similar one was seen pointing to eip:
c01b2d98on the 2.6.10-1.741 kernel.I have included the kallsyms and System.map relevant to the instruction pointed to by the EIP. I can include the full files if required.
printing eip:
c01b387c
*pde =2fcdd001Oops: 0000 [#1] SMP Modules linked in: nfsd exportfs md5 ipv6 parport_pc lp parport autofs4 nfs lockd sunrpc video button battery ac uhci_hcd hw_random e1000 floppy sg dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod mptscsih mptbase sd_mod scsi_mod CPU: 1 EIP: 0060:[<c01b387c>] Not tainted VLI EFLAGS: 00010203 (2.6.10-1.766_FC3smp)EIP is at
__rb_rotate_left+0x8/0x36
eax:f57be640ebx:c041e5e4ecx:f57be640edx: 00000000
esi:f57be640edi:f5f39c80ebp:c041e5e4esp:efa2eed0
ds: 007b es: 007b ss: 0068
Process smbd (pid: 8536, threadinfo=efa2e000 task=f7612a60)Stack: f5ac1980 c01b3988 f5f39c80 f5f39c80 f5ac1988 00000267 c01965d7 f5ac1980 0000000d efa2ef54 efa2ef61 ffffffea c019668b 00000015 00000000 00000267 c03191e0 efa2ef54 00000000 f599ed80 00000267 c01977df ffffffff 001f0000 Call Trace: [<c01b3988>] rb_insert_color+0xa8/0xc1 [<c01965d7>] key_user_lookup+0xcf/0xfc [<c019668b>] key_alloc+0x53/0x2b6 [<c01977df>] keyring_alloc+0x1a/0x48 [<c0198d43>] alloc_uid_keyring+0x2b/0x7c [<c0125d2e>] alloc_uid+0xae/0x133 [<c01291d4>] set_user+0xb/0x8b [<c01295f2>] sys_setresuid+0x11a/0x1b9 [<c0103c97>] syscall_call+0x7/0xb Code: 82 04 01 00 00 00 75 ea 41 83 f9 01 76 ed 31 c0 5b c3 57 b9 45 00 00 00 89 c7 31 c0 f3 ab 5f c3 90 90 90 53 89 d3 8b 50 08 89 c1 <8b> 42 0c 85 c0 89 41 08 74 02 89 08 89 4a 0c 8b 01 85 c0 89 02
/proc/kallsymsgrepped on c01b3.
c01b3024t newaryc01b313cTsys_semgetc01b325attry_atomic_semopc01b336btupdate_queuec01b33detcount_semncntc01b342ctcount_semzcntc01b347at freearyc01b34f6tcopy_semid_to_userc01b354btsemctl_nolockc01b3731tsemctl_mainc01b3ad6tsemctl_downc01b3c39Tsys_semctlc01b3ce9tlookup_undoc01b3d19tfind_undoroot@msh2-c boot]# grep -i c01b3
/boot/System.map-2.6.10-1.766_FC3smpc01b304bTmatch_intc01b3052Tmatch_octalc01b305cTmatch_hexc01b3066Tmatch_strcpyc01b3098Tmatch_strdupc01b30c0tradix_tree_node_allocc01b3109Tradix_tree_preloadc01b317dtradix_tree_extendc01b321cTradix_tree_insertc01b3303Tradix_tree_lookupc01b3349Tradix_tree_tag_setc01b33b8Tradix_tree_tag_clearc01b347at__lookupc01b3547Tradix_tree_gang_lookupc01b3595t__lookup_tagc01b369dTradix_tree_gang_lookup_tagc01b36f0Tradix_tree_deletec01b3835Tradix_tree_taggedc01b3863tradix_tree_node_ctorc01b3874t__rb_rotate_leftc01b38aat__rb_rotate_rightc01b38e0Trb_insert_colorc01b39a1t__rb_erase_colorc01b3b0cTrb_erasec01b3bcdTrb_firstc01b3be5Trb_lastc01b3bfdTrb_nextc01b3c26Trb_prevc01b3c4fTrb_replace_nodec01b3c94Trwsem_wakec01b3d89Trwsem_downgrade_wakec01b3e14T strnicmpc01b3e75T strcpyc01b3e89T strncpyc01b3ea8T strlcpyc01b3eebT strcatc01b3f0aT strncatc01b3f35T strlcatc01b3f97T strcmpc01b3faeT strncmpc01b3fe0T strchrc01b3ff0T strrchrCheers,
Andrew Taylor
Systems Engineer
Array Systems Computing Inc.email: [email protected] http: www.array.ca-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/Quoted 4 lines · Andrew Taylor
Hello,
We have been experiencing a very similar oops on a Dell poweredge 2600, running Fedora Core 3 (FC3). This host is mainly used as a samba fileserver. The problem has been seen on 2 different machines (both Dell-pe-2600s).
Could you pleae confirm that 2.6.12-rc1 fixes this?
Thanks.
Quoted 120 lines
The problem is only ever seen when running the SMP kernel in conjunction with samba.
We have been experiencing problems since early January when the samba/nfs/io load was increased considerably. Since then a number of 2.4 kernels have been used:
2.4.22.2115nptlsmp (FC1)
2.4.22-1.2199smp (FC1)
2.4.29 (vanilla kernel)
2.6.10-1.741_FC3smp (FC3)
2.6.10-1.766_FC3smp (FC3)All 2.4 kernels produced the same result: full system lock-up, with no messages on the console or the log files.
The 2.6.10 kernel is giving an oops which has been seen 3 time in the last 3 days. A very similar one was seen pointing to eip:
c01b2d98on the 2.6.10-1.741 kernel.I have included the kallsyms and System.map relevant to the instruction pointed to by the EIP. I can include the full files if required.
printing eip:
c01b387c
*pde =2fcdd001Oops: 0000 [#1] SMP Modules linked in: nfsd exportfs md5 ipv6 parport_pc lp parport autofs4 nfs lockd sunrpc video button battery ac uhci_hcd hw_random e1000 floppy sg dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod mptscsih mptbase sd_mod scsi_mod CPU: 1 EIP: 0060:[<c01b387c>] Not tainted VLI EFLAGS: 00010203 (2.6.10-1.766_FC3smp)EIP is at
__rb_rotate_left+0x8/0x36
eax:f57be640ebx:c041e5e4ecx:f57be640edx: 00000000
esi:f57be640edi:f5f39c80ebp:c041e5e4esp:efa2eed0
ds: 007b es: 007b ss: 0068
Process smbd (pid: 8536, threadinfo=efa2e000 task=f7612a60)Stack: f5ac1980 c01b3988 f5f39c80 f5f39c80 f5ac1988 00000267 c01965d7 f5ac1980 0000000d efa2ef54 efa2ef61 ffffffea c019668b 00000015 00000000 00000267 c03191e0 efa2ef54 00000000 f599ed80 00000267 c01977df ffffffff 001f0000 Call Trace: [<c01b3988>] rb_insert_color+0xa8/0xc1 [<c01965d7>] key_user_lookup+0xcf/0xfc [<c019668b>] key_alloc+0x53/0x2b6 [<c01977df>] keyring_alloc+0x1a/0x48 [<c0198d43>] alloc_uid_keyring+0x2b/0x7c [<c0125d2e>] alloc_uid+0xae/0x133 [<c01291d4>] set_user+0xb/0x8b [<c01295f2>] sys_setresuid+0x11a/0x1b9 [<c0103c97>] syscall_call+0x7/0xb Code: 82 04 01 00 00 00 75 ea 41 83 f9 01 76 ed 31 c0 5b c3 57 b9 45 00 00 00 89 c7 31 c0 f3 ab 5f c3 90 90 90 53 89 d3 8b 50 08 89 c1 <8b> 42 0c 85 c0 89 41 08 74 02 89 08 89 4a 0c 8b 01 85 c0 89 02
/proc/kallsymsgrepped on c01b3.
c01b3024t newaryc01b313cTsys_semgetc01b325attry_atomic_semopc01b336btupdate_queuec01b33detcount_semncntc01b342ctcount_semzcntc01b347at freearyc01b34f6tcopy_semid_to_userc01b354btsemctl_nolockc01b3731tsemctl_mainc01b3ad6tsemctl_downc01b3c39Tsys_semctlc01b3ce9tlookup_undoc01b3d19tfind_undoroot@msh2-c boot]# grep -i c01b3
/boot/System.map-2.6.10-1.766_FC3smpc01b304bTmatch_intc01b3052Tmatch_octalc01b305cTmatch_hexc01b3066Tmatch_strcpyc01b3098Tmatch_strdupc01b30c0tradix_tree_node_allocc01b3109Tradix_tree_preloadc01b317dtradix_tree_extendc01b321cTradix_tree_insertc01b3303Tradix_tree_lookupc01b3349Tradix_tree_tag_setc01b33b8Tradix_tree_tag_clearc01b347at__lookupc01b3547Tradix_tree_gang_lookupc01b3595t__lookup_tagc01b369dTradix_tree_gang_lookup_tagc01b36f0Tradix_tree_deletec01b3835Tradix_tree_taggedc01b3863tradix_tree_node_ctorc01b3874t__rb_rotate_leftc01b38aat__rb_rotate_rightc01b38e0Trb_insert_colorc01b39a1t__rb_erase_colorc01b3b0cTrb_erasec01b3bcdTrb_firstc01b3be5Trb_lastc01b3bfdTrb_nextc01b3c26Trb_prevc01b3c4fTrb_replace_nodec01b3c94Trwsem_wakec01b3d89Trwsem_downgrade_wakec01b3e14T strnicmpc01b3e75T strcpyc01b3e89T strncpyc01b3ea8T strlcpyc01b3eebT strcatc01b3f0aT strncatc01b3f35T strlcatc01b3f97T strcmpc01b3faeT strncmpc01b3fe0T strchrc01b3ff0T strrchrCheers,
Andrew Taylor
Systems Engineer
Array Systems Computing Inc.email: [email protected] http: www.array.ca-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/