login
Search
Hi
A server with 2.6.7 stopped functioning last night, I booted it up and saw this in kern.log
Aug 26 23:26:43 stratos kernel: Bad page state at prep_new_page (in process 'amavisd-new', page c1776680)
Aug 26 23:26:43 stratos kernel: flags:0x40000104 mapping:00000000 mapcount:1 count:1
Aug 26 23:26:43 stratos kernel: Backtrace:
Aug 26 23:26:43 stratos kernel: [bad_page+107/160] bad_page+0x6b/0xa0
Aug 26 23:26:43 stratos kernel: [prep_new_page+64/96] prep_new_page+0x40/0x60
Aug 26 23:26:43 stratos kernel: [buffered_rmqueue+245/464] buffered_rmqueue+0xf5/0x1d0
Aug 26 23:26:43 stratos kernel: [__alloc_pages+168/880] __alloc_pages+0xa8/0x370
Aug 26 23:26:43 stratos kernel: [__pagevec_lru_add_active+237/288] __pagevec_lru_add_active+0xed/0x120
Aug 26 23:26:43 stratos kernel: [wake_up_page+19/64] wake_up_page+0x13/0x40
Aug 26 23:26:43 stratos kernel: [do_wp_page+121/880] do_wp_page+0x79/0x370
Aug 26 23:26:43 stratos kernel: [pte_alloc_map+125/160] pte_alloc_map+0x7d/0xa0
Aug 26 23:26:43 stratos kernel: [handle_mm_fault+338/384] handle_mm_fault+0x152/0x180
Aug 26 23:26:43 stratos kernel: [do_page_fault+828/1326] do_page_fault+0x33c/0x52e
Aug 26 23:26:43 stratos kernel: [copy_to_user+62/80] copy_to_user+0x3e/0x50
Aug 26 23:26:43 stratos kernel: [sys_rt_sigaction+170/192] sys_rt_sigaction+0xaa/0xc0
Aug 26 23:26:43 stratos kernel: [sigprocmask+61/192] sigprocmask+0x3d/0xc0
Aug 26 23:26:43 stratos kernel: [sys_rt_sigprocmask+134/256] sys_rt_sigprocmask+0x86/0x100
Aug 26 23:26:43 stratos kernel: [do_page_fault+0/1326] do_page_fault+0x0/0x52e
Aug 26 23:26:43 stratos kernel: [error_code+45/56] error_code+0x2d/0x38
Aug 26 23:26:43 stratos kernel:
Aug 26 23:26:43 stratos kernel: Trying to fix it up, but a reboot is needed
I want to know if it's a hardware/kernel failure.
Does anyone understand the backtrace? It would be nice to have your opinion
:-)
Bad page state at prep_new_page
Has there been any progress on this issue? After installing SuSE 9.2, I am experiencing this same problem.
I was running Mandrake 10 Official on this same hardware with no problems. (I forget the exact kernel versions, but I believe that Mandrake was 2.6.3 and SuSE is 2.6.7 or 8.)
I downloaded/compiled/installed 2.6.10 (using the .config from the SuSE 9.2 kernel and mostly taking the defaults for all the new configuration parameters), but that didn't help. It still fails with the same problem.
The machine is now unusable with SuSE, as it will consistenly hit this problem and then lock up after a few minutes of real use. (If I just let it sit there idling, it will run OK.) I should note that I have SuSE 9.2 installed on two other, older machines (one a single processor PIII, the other a dual-processor PII) and it works flawlessly on both. The machine that is failing is much newer: Athlon XP 2600+, ASUS A7V8X, 512MB PC3200 DDR400.
I believe the hardware is OK as this problem does not occur with the earlier kernel. I've also run Memtest for a few days on this machine and it turned up no errors.
It's very embarassing to admit to my friends that my Linux workstation is crashing due to a kernel bug. :-( And I'd really like to move to SuSE 9.2 (love that YaST!).
Is anybody investigating this? I will be glad to grab dumps/tracebacks, etc., (although failures are unpredictable, they're not hard to generate) and forward them.
Thanks.
any solutions for this ?
hey there,
we got the same error with a FC3 installation.
dmesg shows [] bad_page+0x8a/0xbd
and message log shows Bad page state at prep_new_page (in process 'gawk' or in mrtg or in smb
is this really a kernel or hardware failure? its a brand new AMD sempron machine with 1g ram.
any input welcome.
Bad page state kernel unstable!!!!
I had also the same problem, kernel 2.6.14, debian and an AMD 2600+
I spect to find the solution, it's really weird thas kernel is unstable, but i don't find another solution
Same here, and running memtes
Same here, and running memtest shows that it's due to bad RAM...