I'll take a look at the core. This is most odd. As part of my check
in tests I run a stress test which has multiple threads accessing the
device with various block sizes. During the I/O operations the tests
does 'ifconfig down ; sleep <some amount> ; ifconfig up'. This causes
timeout operations, which force retries and connection closures while
the system is doing I/O.
Another question about your target setup. What are you using as the
backing store for the target. I've had problems in the past with UFS
and large logical units; greater than 2GB. The indirect block lookup
for UFS is really slow and this can lead to time outs for the initiator.
On Dec 5, 2006, at 6:02 AM, Rutger Bevaart wrote:
Hello list,
I'm having stability problems with the iSCSI target software on
OpenSolaris (snv_49). The system I'm testing on is a SUN Fire
X2100M2 using a default installation (full etc.).
On the system I have configured one network interface specifically
for iSCSI interconnects using a private address. This all works.
Using the small TCP fix posted here in another post I'm able to
connect to iSCSI targets (static-config).
When I increase the load on the iSCSI target, create large targets,
etc. the network-iscsitgt service transtions to 'maintenance'
state. The following happened when deleting an iSCSI target that
was stuck in 'offline' state.
From /var/svc/log/system-iscsitgt\:default.log:
[ Dec 5 13:52:04 Stopping because process dumped core. ]
[ Dec 5 13:52:16 Executing stop method ("/lib/svc/method/svc-
iscsitgt stop 75")
]
[ Dec 5 13:52:20 Method "stop" exited with status 0 ]
[ Dec 5 13:52:23 Executing start method ("/lib/svc/method/svc-
iscsitgt start")
]
Entity: line 1: parser error : Extra content at the end of the
document
Somehow the configuration becomes corrupted?
/var/adm/messages:
Dec 5 13:50:54 vps01.intern scsi: [ID 243001 kern.warning]
WARNING: /scsi_vhci
(scsi_vhci0):
Dec 5 13:50:54 vps01.intern /scsi_vhci/
[EMAIL PROTECTED]
0 (sd1): Command Timeout on path /iscsi (iscsi0)
Dec 5 13:50:54 vps01.intern scsi: [ID 107833 kern.warning]
WARNING: /scsi_vhci/
[EMAIL PROTECTED] (sd1):
Dec 5 13:50:54 vps01.intern SCSI transport failed: reason
'timeout': retryin
g command
Dec 5 13:52:03 vps01.intern scsi_vhci: [ID 734749 kern.warning]
WARNING: vhci_s
csi_reset 0x1
Dec 5 13:52:03 vps01.intern scsi: [ID 107833 kern.warning]
WARNING: /scsi_vhci/
[EMAIL PROTECTED] (sd1):
Dec 5 13:52:03 vps01.intern SCSI transport failed: reason
'tran_err': retryi
ng command
So before the iscsitgt deamon crashes some commands are timing out.
Anybody have any clues? Access to the server possible... core file
can be downloaded at : http://www.illian.net/core-iscsitgt.gz (3.3MB).
Rgds,
Rutger
This message posted from opensolaris.org
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss
----
Rick McNeal
"If ignorance is bliss, this lesson would appear to be a deliberate
attempt on your part to deprive me of happiness, the pursuit of which
is my unalienable right according to the Declaration of
Independence. I therefore assert my patriotic prerogative not to
know this material. I'll be out on the playground." -- Calvin
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss
