Bug 762454 (GLUSTER-722)

Summary: Segfault in glusterfs 3.0.3 replicate/afr_readdirp_cbk
Product: [Community] GlusterFS Reporter: Jan Boysen <jan>
Component: replicateAssignee: Vikas Gorur <vikas>
Status: CLOSED CURRENTRELEASE QA Contact:
Severity: medium Docs Contact:
Priority: low    
Version: 3.0.2CC: amarts, gluster-bugs, vijay
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
the clients glusterfs.vol
none
the servers glusterfsd.vol none

Description Vijay Bellur 2010-03-15 03:06:22 UTC
Can you please provide the volume files that were in use?

Comment 1 Jan Boysen 2010-03-15 05:08:02 UTC
We have just installed the new released glusterfs 3.0.3 and are facing segfault with the following backtrace:

pending frames:
frame : type(1) op(READDIRP)
frame : type(1) op(READDIRP)
frame : type(1) op(LOOKUP)
frame : type(1) op(LOOKUP)
frame : type(1) op(LOOKUP)
frame : type(1) op(TRUNCATE)
frame : type(1) op(TRUNCATE)
frame : type(1) op(TRUNCATE)
frame : type(1) op(TRUNCATE)
frame : type(1) op(TRUNCATE)
frame : type(1) op(TRUNCATE)

patchset: v3.0.2-41-g029062c
signal received: 11
time of crash: 2010-03-15 04:57:14
configuration details:
argp 1
backtrace 1
dlfcn 1
fdatasync 1
libpthread 1
llistxattr 1
setfsid 1
spinlock 1
epoll.h 1
xattr.h 1
st_atim.tv_nsec 1
package-string: glusterfs 3.0.3
/lib/libc.so.6[0x7f5aad076f40]
/usr/lib/glusterfs/3.0.3/xlator/cluster/replicate.so(afr_readdirp_cbk+0x4d)[0x7f5aabdd0e7d]
/usr/lib/libglusterfs.so.0(default_readdirp_cbk+0x79)[0x7f5aad7d1fc9]
/usr/lib/libglusterfs.so.0(default_readdirp_cbk+0x79)[0x7f5aad7d1fc9]
/usr/lib/libglusterfs.so.0(default_readdirp_cbk+0x79)[0x7f5aad7d1fc9]
/usr/lib/glusterfs/3.0.3/xlator/protocol/client.so(client_readdirp+0x1a3)[0x7f5aac635b33]
/usr/lib/libglusterfs.so.0(default_readdirp+0xe2)[0x7f5aad7d1df2]
/usr/lib/libglusterfs.so.0(default_readdirp+0xe2)[0x7f5aad7d1df2]
/usr/lib/libglusterfs.so.0(default_readdirp+0xe2)[0x7f5aad7d1df2]
/usr/lib/glusterfs/3.0.3/xlator/cluster/replicate.so(afr_do_readdir+0x334)[0x7f5aabdd2614]
/usr/lib/glusterfs/3.0.3/xlator/cluster/replicate.so(afr_readdirp+0xf)[0x7f5aabdd295f]
/usr/lib/glusterfs/3.0.3/xlator/cluster/distribute.so(dht_do_readdir+0x171)[0x7f5aabbb77b1]
/usr/lib/glusterfs/3.0.3/xlator/cluster/distribute.so(dht_readdirp+0xf)[0x7f5aabbb7a9f]
/usr/lib/glusterfs/3.0.3/xlator/mount/fuse.so[0x7f5aab98f176]
/usr/lib/glusterfs/3.0.3/xlator/mount/fuse.so[0x7f5aab98da11]
/lib/libpthread.so.0[0x7f5aad39ffc7]
/lib/libc.so.6(clone+0x6d)[0x7f5aad11483d]
---------

Comment 2 Jan Boysen 2010-03-15 05:42:30 UTC
Created attachment 155 [details]
This is the program that the customer is running accross nfs to exhibit the problem.

attached the clients glusterfs.vol

Comment 3 Jan Boysen 2010-03-15 06:42:47 UTC
Created attachment 156 [details]
printcap

All the servers are identical configured....

Comment 4 Vijay Bellur 2010-03-15 09:54:52 UTC
(In reply to comment #2)
> Created an attachment (id=155) [details]
> the clients glusterfs.vol
> 
> attached the clients glusterfs.vol

Can you please try glusterfs-volgen to generate configuration for you? Loading performance translators between replicate and client-protocol is not recommended by us.

Comment 5 Anand Avati 2010-03-17 07:19:46 UTC
PATCH: http://patches.gluster.com/patch/2936 in release-3.0 (cluster/afr: Fix possible NULL dereference in readdirp_cbk)

Comment 6 Anand Avati 2010-04-20 05:51:00 UTC
PATCH: http://patches.gluster.com/patch/3139 in master (cluster/afr: Fix possible NULL dereference in readdirp_cbk)