Bug 1359131 - Glusterd crashes upon receiving SIGUSR1
Summary: Glusterd crashes upon receiving SIGUSR1
Keywords:
Status: CLOSED DUPLICATE of bug 1364329
Alias: None
Product: GlusterFS
Classification: Community
Component: glusterd
Version: 3.8.1
Hardware: x86_64
OS: Linux
high
high
Target Milestone: ---
Assignee: Atin Mukherjee
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks: 1359190 1364329 1364331
TreeView+ depends on / blocked
 
Reported: 2016-07-22 11:36 UTC by Pavel Malyshev
Modified: 2016-08-05 05:55 UTC (History)
4 users (show)

Fixed In Version:
Clone Of:
: 1359190 (view as bug list)
Environment:
Last Closed: 2016-08-05 05:53:59 UTC
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Embargoed:


Attachments (Terms of Use)
coredump (387.52 KB, application/x-bzip)
2016-07-22 13:56 UTC, Pavel Malyshev
no flags Details

Description Pavel Malyshev 2016-07-22 11:36:15 UTC
Description of problem:
I try to generate glusterdump file with SIGUSR1 to find out maximum op-version for the glusterd on this host.
Glusterd crashes upon receiving SIGUSR1


Version-Release number of selected component (if applicable):
# cat /etc/lsb-release 
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=14.04
DISTRIB_CODENAME=trusty
DISTRIB_DESCRIPTION="Ubuntu 14.04.4 LTS"

# dpkg -l | grep gluster
ii  glusterfs-client                      3.8.1-ubuntu1~trusty1               amd64        clustered file-system (client package)
ii  glusterfs-common                      3.8.1-ubuntu1~trusty1               amd64        GlusterFS common libraries and translator modules
ii  glusterfs-server                      3.8.1-ubuntu1~trusty1               amd64        clustered file-system (server package)


How reproducible:
Glusterd crashes each time I send SIGUSR1


Steps to Reproduce:
1. rm -f /var/run/gluster/glusterdump.*
2. service glusterfs-server start
3. kill -SIGUSR1 $(pidof glusterd)
4. ps -ef | grep glusterd ; ls /var/run/gluster | grep glusterdump

Actual results:
No glusterd in the ps output.
No glusterdump file created.

glusterd crashes with the following lines in the log:
[2016-07-22 11:22:50.923915] I [MSGID: 106493] [glusterd-rpc-ops.c:691:__glusterd_friend_update_cbk] 0-management: Received ACC from uuid: c6654b8f-3ca8-424f-8d0f-7c0b7d8108cc
pending frames:
patchset: git://git.gluster.com/glusterfs.git
signal received: 11
time of crash: 
2016-07-22 11:23:09
configuration details:
argp 1
backtrace 1
dlfcn 1
libpthread 1
llistxattr 1
setfsid 1
spinlock 1
epoll.h 1
xattr.h 1
st_atim.tv_nsec 1
package-string: glusterfs 3.8.1
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(_gf_msg_backtrace_nomem+0x92)[0x7f21f4b87b02]
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(gf_print_trace+0x31d)[0x7f21f4b9104d]
/lib/x86_64-linux-gnu/libc.so.6(+0x36cb0)[0x7f21f3f81cb0]
/lib/x86_64-linux-gnu/libc.so.6(_IO_vfprintf+0x1d03)[0x7f21f3f96943]
/lib/x86_64-linux-gnu/libc.so.6(__vsnprintf_chk+0x95)[0x7f21f4054165]
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(gf_proc_dump_add_section_fd+0x69)[0x7f21f4bb18a9]
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(gf_proc_dump_add_section+0xa5)[0x7f21f4bb1ab5]
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(+0x4ce3e)[0x7f21f4bb1e3e]
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(gf_proc_dump_xlator_info+0x58)[0x7f21f4bb2658]
/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(gf_proc_dump_info+0x500)[0x7f21f4bb2d00]
/usr/sbin/glusterd(glusterfs_sigwaiter+0xce)[0x55ffe99d2bae]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x8184)[0x7f21f4318184]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7f21f404537d]
---------



Expected results:
Glusterd creates glusterdump file in /var/run/gluster


Additional info:

Comment 1 Atin Mukherjee 2016-07-22 12:47:08 UTC
Does this happen everytime? I've seen it earlier as well but couldn't reproduce.

Comment 2 Atin Mukherjee 2016-07-22 12:48:06 UTC
coredump file?

Comment 3 Atin Mukherjee 2016-07-22 13:17:57 UTC
A fix http://review.gluster.org/14987 is posted in master branch.

Comment 4 Pavel Malyshev 2016-07-22 13:56:13 UTC
Created attachment 1182856 [details]
coredump

coredump for /usr/sbin/glusterd from glusterfs-server deb package, version 3.8.1-ubuntu1~trusty1, ppa: "deb http://ppa.launchpad.net/gluster/glusterfs-3.8/ubuntu trusty main"

Comment 5 Pavel Malyshev 2016-07-22 13:58:36 UTC
Atin,

Thank you for the quick reply!
This is a release build without debug symbols.
I've attached coredump.
The backtrace for the crashed thread is:
(gdb) bt
#0  0x00007f591bfde943 in _IO_vfprintf_internal (s=s@entry=0x7f59196d01d0, format=<optimized out>, format@entry=0x7f591cc6aa10 "%s.%s - usage-type %s memusage", 
    ap=ap@entry=0x7f59196d1398) at vfprintf.c:1661
#1  0x00007f591c09c165 in ___vsnprintf_chk (s=0x7f59196d0362 "mgmt/glusterd.management - usage-type ", maxlen=<optimized out>, flags=1, slen=<optimized out>, 
    format=0x7f591cc6aa10 "%s.%s - usage-type %s memusage", args=0x7f59196d1398) at vsnprintf_chk.c:63
#2  0x00007f591cbf98a9 in gf_proc_dump_add_section_fd () from /usr/lib/x86_64-linux-gnu/libglusterfs.so.0
#3  0x00007f591cbf9ab5 in gf_proc_dump_add_section () from /usr/lib/x86_64-linux-gnu/libglusterfs.so.0
#4  0x00007f591cbf9e3e in ?? () from /usr/lib/x86_64-linux-gnu/libglusterfs.so.0
#5  0x00007f591cbfa658 in gf_proc_dump_xlator_info () from /usr/lib/x86_64-linux-gnu/libglusterfs.so.0
#6  0x00007f591cbfad00 in gf_proc_dump_info () from /usr/lib/x86_64-linux-gnu/libglusterfs.so.0
#7  0x0000560afcb5dbae in glusterfs_sigwaiter ()
#8  0x00007f591c360184 in start_thread (arg=0x7f59196d7700) at pthread_create.c:312
#9  0x00007f591c08d37d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

List of threads:
(gdb) info threads
  Id   Target Id         Frame 
  7    Thread 0x7f59152df700 (LWP 12079) pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
  6    Thread 0x7f5919ed8700 (LWP 11999) 0x00007f591c367b9d in nanosleep () at ../sysdeps/unix/syscall-template.S:81
  5    Thread 0x7f5914ade700 (LWP 12080) 0x00007f591c08da13 in epoll_wait () at ../sysdeps/unix/syscall-template.S:81
  4    Thread 0x7f59186d5700 (LWP 12002) pthread_cond_timedwait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_timedwait.S:238
  3    Thread 0x7f591d0a4780 (LWP 11998) 0x00007f591c36165b in pthread_join (threadid=140020575758080, thread_return=0x0) at pthread_join.c:92
  2    Thread 0x7f5918ed6700 (LWP 12001) pthread_cond_timedwait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_timedwait.S:238
* 1    Thread 0x7f59196d7700 (LWP 12000) 0x00007f591bfde943 in _IO_vfprintf_internal (s=s@entry=0x7f59196d01d0, format=<optimized out>, 
    format@entry=0x7f591cc6aa10 "%s.%s - usage-type %s memusage", ap=ap@entry=0x7f59196d1398) at vfprintf.c:1661

Comment 6 Pavel Malyshev 2016-07-22 13:59:52 UTC
And yes, it happens every time on this specific node.
Other nodes (Ubuntu 14.04.4/i686, CentOS 7/x86_64) process SIGUSR1 correctly.

Comment 7 Atin Mukherjee 2016-08-05 05:53:59 UTC

*** This bug has been marked as a duplicate of bug 1364329 ***


Note You need to log in before you can comment on or make changes to this bug.