Lustre devel - Apr 2007 - [Bug 12228] New: LBUG after : ptlrpc_check_set() bad phase ebc0de00

Please don''t reply to lustre-devel. Instead, comment in Bugzilla by
using the following link:
https://bugzilla.lustre.org/show_bug.cgi?id=12228



One of our customer with 1.4.6.1 + Bull patches for Lustre + Quadrics meet the
following problem : one of its Lustre Client met a LBUG after a
"ptlrpc_check_set()) bad phase ebc0de00". 

Here is a targeted extract of the Dump report :

00000100:00020000:5:1175080659.947287:864:19441:0:(client.c:577:ptlrpc_check_status())
@@@ type == PTL_RPC_MSG_ERR, err == -122 req@e0000023fcb2e380 x33516079/t0
o4->ost_galibier19.da0.19_UUID@galibier19_UUID:28 lens 328/288 ref 2 fl
Rpc:R/0/0 rc 0/-122
00000100:00020000:1:1175236082.477747:864:19441:0:(client.c:577:ptlrpc_check_status())
@@@ type == PTL_RPC_MSG_ERR, err == -122 req@e0000063ff642d80 x38661496/t0
o4->ost_galibier19.da0.27_UUID@galibier19_UUID:28 lens 328/288 ref 2 fl
Rpc:R/0/0 rc 0/-122
00000100:00020000:1:1175507735.909770:656:19441:0:(client.c:751:ptlrpc_check_set())
@@@ bad phase ebc0de00 req@e0000063ff64cf80 x44935727/t0
o4->ost_galibier19.da0.27_UUID@galibier19_UUID:28 lens 328/288 ref 3 fl
New:I/0/0 rc 0/0
00000100:00040000:1:1175507735.928262:656:19441:0:(client.c:752:ptlrpc_check_set())
LBUG
[...]
<1>LustreError: dumping log to /tmp/lustre-log.1175507735.19441
[...]
==========================STACK TRACE OF FAILING TASK
=========================================================================================STACK
TRACE FOR TASK: 0xe00000037c710000 (ptlrpcd)

 0 __dump_save_context+0x16c [0xa0000001004352cc]
 1 dump_lcrash_save_context+0x2c [0xa00000010043124c]
 2 dump_lcrash_configure_header+0xaec [0xa0000001004311ac]
 3 dump_generic_execute+0x43c [0xa000000100434f3c]
 4 dump_execute+0x12c [0xa00000010042db6c]
 5 panic_event+0x76c [0xa00000010042fe8c]
 6 notifier_call_chain+0x6c [0xa0000001000a302c]
 7 panic+0x16c [0xa00000010007c4ec]
 8 ptlrpc_check_set+0x4f6c [0xa000000209bd180c]
 9 ptlrpcd_check+0x68c [0xa000000209c78f8c]
10 ptlrpcd+0x6ac [0xa000000209c7988c]
11 kernel_thread_helper+0x2c [0xa0000001000127ec]
12 start_kernel_thread+0x1c [0xa000000100008d1c]

This problem is similar to Bugzilla #11083. However, there were 2 LBUG in #11083
: it is not our case.

It seems that before this LBUG, RPCs can not be sent because of "quota
exceeded"
: is it a Quota problem that could be at the origin of this LBUG ?