After disabling posix-locks, the tar test runs properly. Zimbra backups still cause seqfaults. Here is a bt: inode_forget (inode=0x8059ac0, nlookup=0) at list.h:92 92 prev->next = next; (gdb) bt #0 inode_forget (inode=0x8059ac0, nlookup=0) at list.h:92 #1 0xb764ad0a in posix_forget () from /usr/lib/glusterfs/1.3.0-pre5/xlator/storage/posix.so #2 0xb763f676 in iot_forget_wrapper () from /usr/lib/glusterfs/1.3.0-pre5/xlator/performance/io-threads.so #3 0xb7fcef4a in call_resume_wind (stub=0x8059ac0) at call-stub.c:2027 #4 0xb7fcefd7 in call_resume (stub=0x8a7e630) at call-stub.c:2763 #5 0xb76437a5 in iot_worker () from /usr/lib/glusterfs/1.3.0-pre5/xlator/performance/io-threads.so #6 0xb7f9f3db in start_thread () from /lib/libpthread.so.0 #7 0xb7f2926e in clone () from /lib/libc.so.6 Harris ----- Original Message ----- From: "Harris Landgarten" <harrisl@xxxxxxxxxxxxx> To: "gluster-devel" <gluster-devel@xxxxxxxxxx> Sent: Friday, June 29, 2007 12:20:07 PM (GMT-0500) America/New_York Subject: io-threads seqfault on r/w I ran the following test: cd /mnt/glusterfs glusterfs # tar -cvjf ./test/test.tbz backups/ where backups is a large tree brick2, which exports /mnt/export and /mnt/namespace crashes almost immediately with the following bt: [Switching to Thread -1219015792 (LWP 1376)] 0xb7f1bb59 in truncate_stat_cbk () from /usr/lib/glusterfs/1.3.0-pre5/xlator/features/posix-locks.so (gdb) bt #0 0xb7f1bb59 in truncate_stat_cbk () from /usr/lib/glusterfs/1.3.0-pre5/xlator/features/posix-locks.so #1 0xb7f163ac in call_resume (stub=0x805ec70) at call-stub.c:2117 #2 0xb758a770 in iot_reply () from /usr/lib/glusterfs/1.3.0-pre5/xlator/performance/io-threads.so #3 0xb7ee63db in start_thread () from /lib/libpthread.so.0 #4 0xb7e7026e in clone () from /lib/libc.so.6 Harris _______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxx http://lists.nongnu.org/mailman/listinfo/gluster-devel
