repeatable statd segfault

DISCUSSION michael 1 participant 2 messages Sep 5, 2003

Subsystems (2)

NFS client NFS server

lore

Share & tools

lore

michael

author

2003-09-05 03:01 UTC #0 lore raw

(gdb) run -Fd

Starting program: /usr/local/src/nfs-utils-1.0.5/utils/statd/statd -Fd
09/04/2003 19:44:57 statd[12873]: Version 1.0.5 Starting
09/04/2003 19:44:57 statd[12873]: Flags: No-Daemon Log-STDERR
09/04/2003 19:44:57 statd[12873]: statd running as root. chown /var/lib/nfs/sm to choose different user

09/04/2003 19:44:57 statd[12873]: New state: 15467 09/04/2003 19:44:57 statd[12873]: Waiting for client connections.

--> at this point i run ./testlk /home/.raby/nfs_lock_test/blah --> the file is mounted on a netapp 760 and the following happens:

--- Begin notify list dump ---
Entry 1: 10.3.19.5
--- End notify list dump ---
09/04/2003 19:45:12 statd[12873]: MONITORING 10.18.34.19 for gir

Program received signal SIGSEGV, Segmentation fault. 0x40119e3a in svc_getreqset () from /lib/libc.so.6 (gdb)

--> 10.18.34.19 is the netapp.

one question i have is, does /var/lib/nfs/sm/10.18.34.19 get unlinked after being created? it appears on occaision when i create a working lock, but then dissapears from the directory before the lock is released.

thanks!
michael

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs

michael

author

2003-09-05 03:24 UTC #1 lore raw

Here is a slight variation of the segfault. in this case locking worked for about 5-10 minutes, then upon a test suddenly... (note: this is a slightly older version of statd.)

(gdb) run -Fd
The program being debugged has been started already.
Start it from the beginning? (y or n) y
Starting program: /sbin/rpc.statd -Fd
(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)... 09/04/2003 20:03:58 rpc.statd[18964]: Version 1.0 Starting
09/04/2003 20:03:58 rpc.statd[18964]: Flags: No-Daemon Log-STDERR
09/04/2003 20:03:58 rpc.statd[18964]: New state: 15471
09/04/2003 20:03:58 rpc.statd[18964]: Waiting for reply... (timeo 5)
09/04/2003 20:03:58 rpc.statd[18964]: Waiting for reply... (timeo 5)
09/04/2003 20:03:58 rpc.statd[18964]: Notification of 10.7.10.8 succeeded.
09/04/2003 20:03:58 rpc.statd[18964]: Unlinked /var/lib/nfs/sm.bak/10.7.10.8
09/04/2003 20:03:58 rpc.statd[18964]: Waiting for client connections.
09/04/2003 20:18:11 rpc.statd[18964]: Received SM_UNMON request from gir for 10.7.10.8 while not monitoring any hosts. (no debugging symbols found)...(no debugging symbols found)...
Program received signal SIGSEGV, Segmentation fault.
0x40122e3a in svc_getreqset () from /lib/libc.so.6
(gdb)

Quoted 14 lines · michael

(gdb) run -Fd
Starting program: /usr/local/src/nfs-utils-1.0.5/utils/statd/statd -Fd
09/04/2003 19:44:57 statd[12873]: Version 1.0.5 Starting
09/04/2003 19:44:57 statd[12873]: Flags: No-Daemon Log-STDERR
09/04/2003 19:44:57 statd[12873]: statd running as root. chown /var/lib/nfs/sm to choose different user
09/04/2003 19:44:57 statd[12873]: New state: 15467 09/04/2003 19:44:57 statd[12873]: Waiting for client connections.
--> at this point i run ./testlk /home/.raby/nfs_lock_test/blah --> the file is mounted on a netapp 760 and the following happens:

Quoted 8 lines

--- Begin notify list dump ---
Entry 1: 10.18.34.19
--- End notify list dump ---
09/04/2003 19:45:12 statd[12873]: MONITORING 10.18.34.19 for gir
Program received signal SIGSEGV, Segmentation fault. 0x40119e3a in svc_getreqset () from /lib/libc.so.6 (gdb)