Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 16 Oct 2017 10:45:45 +0000
From:      bugzilla-noreply@freebsd.org
To:        freebsd-scsi@FreeBSD.org
Subject:   [Bug 222898] [iscsi]: panic: page fault - system crashes while working as iSCSI target
Message-ID:  <bug-222898-5312-ZCWkzv642Y@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-222898-5312@https.bugs.freebsd.org/bugzilla/>
References:  <bug-222898-5312@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D222898

--- Comment #8 from emz@norma.perm.ru ---
Well, I took a risk and installed a stock r315520. Unfortunately, the behav=
ior
is identical: crash within dozens of seconds after starting ctld, same panic
message on a debug kernel:

=3D=3D=3DCut=3D=3D=3D
WARNING: 213.152.134.114 (iqn.1991-05.com.microsoft:worker85): no ping reply
(NOP-Out) after 5 seconds; dropping connect
ion
WARNING: 213.152.138.158 (iqn.1991-05.com.microsoft:worker265): no ping rep=
ly
(NOP-Out) after 5 seconds; dropping connec
tion
WARNING: 213.152.134.62 (iqn.1991-05.com.microsoft:worker33): no ping reply
(NOP-Out) after 5 seconds; dropping connecti
on
WARNING: 213.152.134.114 (iqn.1991-05.com.microsoft:worker85): no ping reply
(NOP-Out) after 5 seconds; dropping connect
ion
panic: destroying session with 4 outstanding PDUs
cpuid =3D 24
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0xfffffe105789a=
8f0
vpanic() at vpanic+0x186/frame 0xfffffe105789a970
kassert_panic() at kassert_panic+0x126/frame 0xfffffe105789a9e0
icl_soft_conn_close() at icl_soft_conn_close+0x20a/frame 0xfffffe105789aa10
cfiscsi_maintenance_thread() at cfiscsi_maintenance_thread+0x100/frame
0xfffffe105789aa70
fork_exit() at fork_exit+0x84/frame 0xfffffe105789aab0
fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe105789aab0
--- trap 0, rip =3D 0, rsp =3D 0, rbp =3D 0 ---
Uptime: 1m55s
=3D=3D=3DCut=3D=3D=3D

I am running a non-debug version of r315520 right now (so there's no either
explicit regression or improvement over the r310734), it's still tending to
crash in random moments of time when loaded by 299 initiators.

As a workaround I will try to diminish the load by transferring some initia=
tors
to a spare SAN.

Crashdumps, binaries and infos:

http://files2.enaza.ru/ctld-stuck/r315520/

In the same time I'm still willing to help in any way in resolving this,
including the access on a production system, over ssh/IPMI/whatever.

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-222898-5312-ZCWkzv642Y>