On 14 May 2009 at 9:48, Mike Christie wrote:
>
> Ulrich Windl wrote:
> > Hi,
> >
> > investigating what happened during a power failure, I have a small problem
> > relating the connection numbers found in syslog to the output produced by
> > iscaiadm: Does the session number (in square brackets) correspond to the
> > connection number?
>
> It is the session number. The connection values for connection22:0 is
> the %SESSION_NUMBER:CONNECTION_NUMBER.
>
> iSCSI has this feature where you can have multiple connections per
> session. We used to support it which is why some of the iscsid.conf
> params have that [0] index. We do not support it anymore because
> upstream would not allow, so right now the connection number is always 0.
>
> >
> > i.e. Does "tcp: [22] 172.20.77.2:3260,0 iqn.1986-
> > 03.com.hp:fcgw.mpx100:rkdvmis2.1.50001fe1500c1f60.50001fe1500c1f6d"
> > correspond to
> > "iscsid: connection22:0 is operational after recovery (3 attempts)", and
> > does that
> > correspond to "kernel: sd 22:0:0:10: SCSI error: return code = 0x00010000"?
> >
> > AFAIK a session can have multiple connections to a target, and a target can
> > have
>
> You can also have multiple sessions to a target. We normally do a
> session to each portal on the target. With that iface stuff you can
> create more.
>
> > multiple LUNs. How can I find out what kernel objects depend on a specific
> > iSCSI
> > connection?
> >
>
> In newer tools we have
> iscsiadm -m session -P 3
Well, I knew, but...where is the connection number there (SLES10 SP2)?:
Target:
iqn.1986-03.com.hp:fcgw.mpx100:rkdvmis1.0.50001fe1500c1f60.50001fe1500c1f6c
Current Portal: 172.20.76.1:3260,0
Persistent Portal: 172.20.76.1:3260,0
**********
Interface:
**********
Iface Name: default
Iface Transport: tcp
Iface Initiatorname:
iqn.1992-04.de.uni-regensburg.klinik:testhost
Iface IPaddress: 172.20.76.13
Iface HWaddress: default
Iface Netdev: default
SID: 32
iSCSI Connection State: LOGGED IN
iSCSI Session State: Unknown
Internal iscsid Session State: NO CHANGE
************************
Negotiated iSCSI params:
************************
HeaderDigest: None
DataDigest: None
MaxRecvDataSegmentLength: 65536
MaxXmitDataSegmentLength: 65536
FirstBurstLength: 65536
MaxBurstLength: 262144
ImmediateData: Yes
InitialR2T: Yes
MaxOutstandingR2T: 1
************************
Attached SCSI devices:
************************
Host Number: 34 State: running
scsi34 Channel 00 Id 0 Lun: 0
>
> In older tools there was a debug option
> iscsiadm -m session -i
On a multiple-CPU machine with multiple iSCSI portals an storage systems
behind,
event ordering can be quite confusing, so that's why I'm asking. Again, when
seeing:
kernel: ping timeout of 15 secs expired, last rx 4748275159, last ping
4748277659,
now 4748281409
kernel: connection12:0: iscsi: detected conn error (1011)
How can I find out which connection (i.e. which SCSI target) is actually
affected?
Regards,
Ulrich
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"open-iscsi" group.
To post to this group, send email to open-iscsi@googlegroups.com
To unsubscribe from this group, send email to
open-iscsi+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~----------~----~----~----~------~----~------~--~---
