ia64 "hello world" success; crfsd receives SEGV under heavier load

[Alex Chiang]

Hi Zach,

So the good news is, I've been able to successfully follow the
cookbook and get crfs working. My setup:

hp rx2660 (server)             hp rx6600 (client)
crfsd               <------->  2.6.25-rc7, crfs

[btw, both machines are ia64, the rx2660 has a debian 4.0
userspace, the rx6600 has a fedora 9 userspace running my
upstream git kernel]

Doing a 'hello world' test of:

	- mount on rx6600
	- echo "hello" > hello.txt
	- cat hello.txt
	- umount on rx6600
	- mount on rx6600
	- cat hello.txt

Was successful.

So the next obvious step was to copy a kernel tree over crfs. ;)

crfsd on my server died at some point during the transfer.

The logfile said:

Apr 01,2008|19:54:31|11462|sigaction_exit: caught SIGCHLD for our child pid 11463 (status 6), shutting down
Apr 01,2008|19:54:31|11461|sigaction_exit: caught SIGCHLD for our child pid 11462 (status 1), shutting down
Apr 01,2008|19:54:31|11460|sigaction_exit: caught SIGCHLD for our child pid 11461 (status 1), shutting down

Not too informative. Unfortunately, I didn't have the verbose
debug log turned on. Next trial...

Here's a backtrace. I'll keep poking at this later, but figured
you might want to take a look too.

Thanks.

/ac

achiang at blender:~/crfs$ gdb ./crfsd/crfsd /tmp/crfsd-logs/core 
GNU gdb 6.4.90-debian
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "ia64-linux-gnu"...Using host libthread_db library "/
lib/tls/libthread_db.so.1".


warning: Can't read pathname for load map: Input/output error.
Reading symbols from /lib/tls/libc.so.6.1...done.
Loaded symbols for /lib/tls/libc.so.6.1
Reading symbols from /lib/ld-linux-ia64.so.2...done.
Loaded symbols for /lib/ld-linux-ia64.so.2
Core was generated by `./crfsd/crfsd -c /tmp/crfsd.conf'.
Program terminated with signal 6, Aborted.
#0  0xa000000000010620 in __kernel_syscall_via_break ()
(gdb) bt
#0  0xa000000000010620 in __kernel_syscall_via_break ()
#1  0x2000000000093bb0 in raise () from /lib/tls/libc.so.6.1
#2  0x2000000000096df0 in abort () from /lib/tls/libc.so.6.1
#3  0x2000000000084180 in __assert_fail () from /lib/tls/libc.so.6.1
#4  0x4000000000035a50 in btr_dirty_to_leaf (upd=0x60000000087f0570, 
    root_block=0x60000000087f0588, dkey=0x6000000009dec75b, 
    parent_ret=0x60000fffffcc3618, leaf_ret=0x60000fffffcc3620, inserting=1)
    at crfsd/update.c:868
#5  0x40000000000396e0 in add_item (upd=0x60000000087f0570, 
    root_block=0x60000000087f0588, dkey=0x6000000009dec75b, 
    data=0x6000000009def133, len=4) at crfsd/update.c:1141
#6  0x40000000000399c0 in btr_add_item (upd=0x60000000087f0570, 
    dkey=0x6000000009dec75b, data=0x6000000009def133, len=4)
    at crfsd/update.c:1169
#7  0x40000000000259a0 in meta_range_write (btr=0x6000000000004fa0, 
    upd=0x60000000087f0570, payload=0x6000000009dec015)
    at crfsd/metawrite.c:208
#8  0x400000000002b6a0 in trans_commit (btr=0x6000000000004fa0, 
    vec=0x6000000008058b30, nr_frags=65) at crfsd/trans.c:133
#9  0x400000000003a960 in handle_incoming_frag (fd=6, btr=0x6000000000004fa0, 
    frags=0x60000fffffcc3838, hdr=0x60000fffffcc3820, buf=0x0)
    at crfsd/worker.c:100
#10 0x400000000003b910 in worker_main (client_fd=6) at crfsd/worker.c:200
#11 0x4000000000026bd0 in msger_main (client_fd=6, sin=0x60000fffffcc3964)
    at crfsd/msger.c:190
#12 0x4000000000003200 in accept_main () at crfsd/accept.c:168
#13 0x4000000000014e00 in main (argc=3, argv=0x60000fffffcc3d18)
    at crfsd/crfsd.c:212