Re: fsm crash again

If you are interested, this is a follow-up on the problem:
http://www.xsanity.com/forum/viewtopic.php?t=3853

In short: I came to a conclusion that metadata for the crashing volume got
corrupted during the expansion of the volume and this cannot be fixed with
cvfsck. 

Basically I’m waiting for the money to buy new hardware, so I can backup my
data and recreate the volume.

-- 
  Vilius

From: William Strucke [mailto:strucke1@xxxxxxxxxxxxxxxxxxx] 
Sent: 2009 m. liepos 17 d. 15:34
To: Vilius Šumskas
Cc: xsan-users@xxxxxxxxxxxxxxx
Subject: Re: fsm crash again

We are seeing the same behavior -- this has been going on for months and I
can not find the culprit, only it's just one of our three volumes.  It's
constantly crashing and failing over, just going back and forth between the
two metadata controllers until eventually one of them fails to bring it back
up.  I've run cvfsck -C / cvfsck -vw twenty times and it hasn't fixed it. 
I've replaced both of the hard drives in my metadata array since that was
the only set that was reporting any type of error but it's still happening.

Does anyone have any ideas?
Could not find File System Manager for "netshare" on 172.16.4.3.
<snip>
Jul 17 01:40:47 xsan-2 ReportCrash[756]: Formulating crash report for
process fsm[527]
Jul 17 01:40:48 xsan-2 ReportCrash[756]: Saved crashreport to
/Library/Logs/CrashReporter/fsm_2009-07-17-014045_xsan-2.crash using uid: 0
gid: 0, euid: 0 egid: 0
Jul 17 01:40:48 xsan-2 fsmpm[89]: PortMapper: FSS 'netshare' disconnected.
Jul 17 01:40:48 xsan-2 fsmpm[89]: PortMapper: kicking diskscan_thread
-264712192.
Jul 17 01:40:48 xsan-2 fsmpm[89]: Portmapper: FSS 'netshare' (pid 527)
exited on signal 6
-- 
William Strucke
Systems Manager
Division of Arts and Humanities
College of Arts and Sciences at The Ohio State University
614/688.4447 or strucke.1@xxxxxxx


Vilius Šumskas wrote: 
Hi list,

It seems that our Xsan stories will never end. We have fsm crashes again on
our volumes. This time it's different error that that because of ACL
corruption:

Mar 10 13:37:57 xserve3 com.apple.xsan[91]: fsm(34918,0xf4f37000) malloc:
*** error for object 0x67f0800: incorrect checksum for freed object - object
was probably modified after being freed.
Mar 10 13:37:57 xserve3 fsm[34918]: fsm(34918,0xf4f37000) malloc: *** error
for object 0x67f0800: incorrect checksum for freed object - object was
probably modified after being freed.\n*** set a breakpoint in
malloc_error_break to debug
Mar 10 13:37:57 xserve3 com.apple.xsan[91]: *** set a breakpoint in
malloc_error_break to debug
Mar 10 13:37:57 xserve3 fsm[34918]: Xsan FSS 'EditSAN[0]': PANIC:
/Library/Filesystems/Xsan/bin/fsm "Segmentation Fault Signal." file fsm.c,
line 554
Mar 10 13:37:57 xserve3 KernelEventAgent[101]: tid 00000000 received
VQ_NOTRESP event (1)
Mar 10 13:37:57 xserve3 KernelEventAgent[101]: tid 00000000 type 'acfs',
mounted on '/Volumes/EditSAN', from '/dev/disk12', not responding
Mar 10 13:37:57 xserve3 fsm[34918]: PANIC: /Library/Filesystems/Xsan/bin/fsm
"Segmentation Fault Signal." file fsm.c, line 554

And

Mar 10 16:55:53 xserve3 com.apple.xsan[91]: fsm(44892,0xf4d2f000) malloc:
*** error for object 0x75a769e5: Non-aligned pointer being freed
Mar 10 16:55:53 xserve3 fsm[44892]: fsm(44892,0xf4d2f000) malloc: *** error
for object 0x75a769e5: Non-aligned pointer being freed\n*** set a breakpoint
in malloc_error_break to debug
Mar 10 16:55:53 xserve3 com.apple.xsan[91]: *** set a breakpoint in
malloc_error_break to debug

Any ideas?

  

 _______________________________________________
Do not post admin requests to the list. They will be ignored.
Xsan-Users mailing list      (Xsan-Users@xxxxxxxxxxxxxxx)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/xsan-users/maillists%40codeha.us

This email sent to maillists@xxxxxxxxx