After Upgrading 9.3 to 9.10 Automatic Restart my NAS Box
After Upgrading 9.3 to 9.10 Automatic Restart my NAS Box five to eight time per day, i am facing this issue last 35 days. i am unable to resolve problem. please help me to resolve problem.
find a attachment of debug.
#1 Updated by Sean Fagan over 4 years ago
- Category changed from 1 to 137
- Assignee changed from Sean Fagan to Alexander Motin
I suspect you have bad hardware. Not sure what part, though.
Tracing pid 19 tid 100151 td 0xfffff8000759e960
zrl_add_impl() at zrl_add_impl+0x12/frame 0xfffffe0215c77880
dbuf_rele_and_unlock() at dbuf_rele_and_unlock+0x1a5/frame 0xfffffe0215c77930
sa_handle_destroy() at sa_handle_destroy+0x80/frame 0xfffffe0215c77960
zfs_zinactive() at zfs_zinactive+0x104/frame 0xfffffe0215c779a0
zfs_freebsd_reclaim() at zfs_freebsd_reclaim+0x66/frame 0xfffffe0215c779d0
VOP_RECLAIM_APV() at VOP_RECLAIM_APV+0xa7/frame 0xfffffe0215c77a00
vgonel() at vgonel+0x221/frame 0xfffffe0215c77a70
vnlru_free() at vnlru_free+0x1dd/frame 0xfffffe0215c77ae0
vnlru_proc() at vnlru_proc+0xa1/frame 0xfffffe0215c77b70
fork_exit() at fork_exit+0x9a/frame 0xfffffe0215c77bb0
fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe0215c77bb0
#5 Updated by Vijay Nayani almost 4 years ago
Hi, First of all Thank you for your prompt response. I have managed to find the issue within a RAM module due to your inputs only.
Query Out of curiosity: From which parameter within the log text did you managed to identify the issue within the hardware.
I thoroughly appreciate your help and support.
#8 Updated by Chris Torek over 3 years ago
I just noticed this in my list (I haven't been paying attention to all the "nice to have" bugs for quite a while...).
Sean is the one who suspected bad hardware. I'm not sure why he came to that conclusion.
I have an old Intel box at home that has no ECC RAM and it had a bad RAM chip that I only found by running memtest86+ on it. I had no real proof, just knowing that it was non-ECC and was crashing in ways that did not repeat exactly was sufficient to prompt me to run the memory tester.
(The real lesson is, always try to get ECC hardware. ECC systems detect and correct single bit RAM errors, and usually detect, but can't correct, multi-bit errors.)