forked from illumos/illumos-gate
-
Notifications
You must be signed in to change notification settings - Fork 111
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
NULL pointer dereference in fop_frlock #192
Comments
On 3/11/19 15:38 , Sebastian Wiedenroth wrote:
Had the following panic on a system running `joyent_20180524T002819Z`.
The last thing I entered was `pfiles` on a `vmstat` process, but the system was also otherwise busy.
```
2019-03-11T22:19:16.139806+00:00 p unix: [ID 836849 kern.notice] #12#015panic[cpu5]/thread=fffffe59432a5060:
2019-03-11T22:19:16.139814+00:00 p genunix: [ID 335743 kern.notice] BAD TRAP: type=e (#pf Page fault) rp=fffffe007c53e9d0 addr=18 occurred in module "genunix" due to a NULL pointer dereference
2019-03-11T22:19:16.139818+00:00 p unix: [ID 100000 kern.notice] #12
2019-03-11T22:19:16.139821+00:00 p unix: [ID 839527 kern.notice] vmstat:
2019-03-11T22:19:16.139825+00:00 p unix: [ID 753105 kern.notice] #pf Page fault
2019-03-11T22:19:16.139828+00:00 p unix: [ID 532287 kern.notice] Bad kernel fault at addr=0x18
2019-03-11T22:19:16.139832+00:00 p unix: [ID 243837 kern.notice] pid=72777, pc=0xfffffffffbb611f2, sp=0xfffffe007c53eac0, eflags=0x10286
2019-03-11T22:19:16.139846+00:00 p unix: [ID 619397 kern.notice] cr0: 8005003b<pg,wp,ne,et,ts,mp,pe> cr4: 1606f8<smep,osxsav,pcide,xmme,fxsr,pge,mce,pae,pse,de>
2019-03-11T22:19:16.139851+00:00 p unix: [ID 152204 kern.notice] cr2: 18
2019-03-11T22:19:16.139857+00:00 p unix: [ID 634440 kern.notice] cr3: 106be8e000
2019-03-11T22:19:16.139861+00:00 p unix: [ID 625715 kern.notice] cr8: 0
2019-03-11T22:19:16.139865+00:00 p unix: [ID 100000 kern.notice]
2019-03-11T22:19:16.139868+00:00 p unix: [ID 592667 kern.notice] #011rdi: 0 rsi: e rdx: fffffe007c53ed20
2019-03-11T22:19:16.139871+00:00 p unix: [ID 592667 kern.notice] #011rcx: 2005 r8: 25 r9: 0
2019-03-11T22:19:16.139880+00:00 p unix: [ID 592667 kern.notice] #011rax: fffffe59a8bbf710 rbx: 0 rbp: fffffe007c53eb50
2019-03-11T22:19:16.139884+00:00 p unix: [ID 592667 kern.notice] #011r10: fffffe592f9eeac0 r11: fffffe59ff01b600 r12: e
2019-03-11T22:19:16.139889+00:00 p unix: [ID 592667 kern.notice] #011r13: 8044af0 r14: fffffe5a2bce6c48 r15: 7fffffff
2019-03-11T22:19:16.139892+00:00 p unix: [ID 592667 kern.notice] #011fsb: 0 gsb: fffffe594281e000 ds: 4b
2019-03-11T22:19:16.139896+00:00 p unix: [ID 592667 kern.notice] #11 es: 4b fs: 0 gs: 0
2019-03-11T22:19:16.139899+00:00 p unix: [ID 592667 kern.notice] #011trp: e err: 0 rip: fffffffffbb611f2
2019-03-11T22:19:16.139903+00:00 p unix: [ID 592667 kern.notice] #11 cs: 30 rfl: 10286 rsp: fffffe007c53eac0
2019-03-11T22:19:16.139906+00:00 p unix: [ID 266532 kern.notice] #11 ss: 0
2019-03-11T22:19:16.139909+00:00 p unix: [ID 100000 kern.notice]
2019-03-11T22:19:16.139913+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53e8c0 unix:real_mode_stop_cpu_stage2_end+bc1c ()
2019-03-11T22:19:16.139916+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53e9c0 unix:trap+108e ()
2019-03-11T22:19:16.139920+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53e9d0 unix:_cmntrap+e6 ()
2019-03-11T22:19:16.139931+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53eb50 genunix:fop_frlock+32 ()
2019-03-11T22:19:16.139936+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53eba0 dev:sdev_frlock+3e ()
2019-03-11T22:19:16.139940+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53ec40 genunix:fop_frlock+65 ()
2019-03-11T22:19:16.139943+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53eeb0 genunix:fcntl+be9 ()
2019-03-11T22:19:16.139947+00:00 p genunix: [ID 655072 kern.notice] fffffe007c53ef10 unix:brand_sys_syscall32+1aa ()
```
Sadly the dump does not seem usable:
```
# savecore -f vmdump.0
savecore: incomplete dump on dump device
savecore: System dump time: Mon Mar 11 21:40:54 2019
savecore: saving system crash dump in /var/crash/volatile/{unix,vmcore}.0
Constructing namelist /var/crash/volatile/unix.0
Constructing corefile /var/crash/volatile/vmcore.0
pfn 16266240 not found for as=fffffffffbc4b080, va=fffffe0000000000
pfn 16266241 not found for as=fffffffffbc4b080, va=fffffe0000001000
pfn 16266242 not found for as=fffffffffbc4b080, va=fffffe0000002000
pfn 16266243 not found for as=fffffffffbc4b080, va=fffffe0000003000
pfn 16266244 not found for as=fffffffffbc4b080, va=fffffe0000004000
pfn 16266245 not found for as=fffffffffbc4b080, va=fffffe0000005000
pfn 16266246 not found for as=fffffffffbc4b080, va=fffffe0000006000
pfn 16266247 not found for as=fffffffffbc4b080, va=fffffe0000007000
pfn 16266248 not found for as=fffffffffbc4b080, va=fffffe0000008000
pfn 16266249 not found for as=fffffffffbc4b080, va=fffffe0000009000
0:18 99% donesavecore: dsize 0xf0b != pagesize 0x1000
```
That's unfortunate. From looking here, I don't immediately see anything
that would be related that would have fixed this between now and then.
Can you check the size of the dump device? At least, a naive pfiles of a
vmstat process doesn't immediately replicate this, unfortunately.
The fact that sdev is on the scene and we're trying to do something on
the backend attribute vnode is a bit suspicious though. Based on the
stack there, it feels like either the vnode_t or vfs_t pointer is bad.
Robert
|
Dump device size was at 2.52G, probably created by the old installer a long time ago. I've set it to 30G now as So if it happens again we'll hopefully get more out of it. |
On 3/11/19 16:09 , Sebastian Wiedenroth wrote:
Dump device size was at 2.52G, probably created by the old installer a long time ago. I've set it to 30G now as `dumpadm -e` suggested ~15G which should reflect the current installers logic.
So if it happens again we'll hopefully get more out of it.
If you have the disk space, I might size it to RAM. Sorry I don't have
anything more immediately actionable here to suggest that we should look at.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Had the following panic on a system running
joyent_20180524T002819Z
.The last thing I entered was
pfiles
on avmstat
process, but the system was also otherwise busy.Sadly the dump does not seem usable:
The text was updated successfully, but these errors were encountered: