[Stgt-devel] Open connections

Tomasz Chmielewski mangoo
Fri Apr 11 10:03:05 CEST 2008


FUJITA Tomonori schrieb:
> On Thu, 10 Apr 2008 16:46:57 +0200
> Tomasz Chmielewski <mangoo at wpkg.org> wrote:
> 
>> FUJITA Tomonori schrieb:
>>> On Thu, 10 Apr 2008 16:16:45 +0200
>>> Tomasz Chmielewski <mangoo at wpkg.org> wrote:
>>>
>>>> Niels de Carpentier schrieb:
>>>>
>>>> (...)
>>>>
>>>>>> Other problem I had with today's git is that tgtd "hangs", i.e., the
>>>>>> process still exists, but:
>>>>> There should be 2 processes, probably 1 died? I've had this happen during
>>>>> a initiator reconnect while the target was being configured. Putting the
>>>>> target offline during configuration has solved all these issues for me.
>>>> It happened again - only one initiator was connected and operational, 
>>>> after ~2 days of iscsi connection. I didn't restart anything, filter 
>>>> packets, unplug cables, reconfigure or anything like that.
>>> What kind of workload did you use?
>>>
>>> I need to reproduce the bug. If you can, please run tgtd via gdb.
>> It's about ~10 MB/s read or write; sometimes it goes up to ~20-25 MB/s.
>> Mostly random reads or writes.
> 
> What iSCSI parameters do you use? Can you show me the output of
> open-iscsi command to show the iSCSI parameters of a session?

# cat /etc/iscsi/nodes/iqn.2006*/*/default
node.name = iqn.2006-08.net.syneticon:superthecus.backup
node.tpgt = 1
node.startup = manual
iface.hwaddress = default
iface.iscsi_ifacename = default
iface.net_ifacename = default
iface.transport_name = tcp
node.discovery_address = 192.168.111.177
node.discovery_port = 3260
node.discovery_type = send_targets
node.session.initial_cmdsn = 0
node.session.initial_login_retry_max = 4
node.session.cmds_max = 128
node.session.queue_depth = 32
node.session.auth.authmethod = None
node.session.auth.username = user
node.session.auth.password = verysecretpass
node.session.timeo.replacement_timeout = 1000000
node.session.err_timeo.abort_timeout = 10
node.session.err_timeo.reset_timeout = 30
node.session.iscsi.InitialR2T = No
node.session.iscsi.ImmediateData = Yes
node.session.iscsi.FirstBurstLength = 262144
node.session.iscsi.MaxBurstLength = 16776192
node.session.iscsi.DefaultTime2Retain = 0
node.session.iscsi.DefaultTime2Wait = 0
node.session.iscsi.MaxConnections = 1
node.session.iscsi.MaxOutstandingR2T = 1
node.session.iscsi.ERL = 0
node.conn[0].address = 192.168.111.177
node.conn[0].port = 3260
node.conn[0].startup = manual
node.conn[0].tcp.window_size = 524288
node.conn[0].tcp.type_of_service = 0
node.conn[0].timeo.logout_timeout = 15
node.conn[0].timeo.login_timeout = 15
node.conn[0].timeo.auth_timeout = 45
node.conn[0].timeo.active_timeout = 5
node.conn[0].timeo.idle_timeout = 60
node.conn[0].timeo.ping_timeout = 5
node.conn[0].timeo.noop_out_interval = 10
node.conn[0].timeo.noop_out_timeout = 15
node.conn[0].iscsi.MaxRecvDataSegmentLength = 65536
node.conn[0].iscsi.HeaderDigest = None,CRC32C
node.conn[0].iscsi.DataDigest = None
node.conn[0].iscsi.IFMarker = No
node.conn[0].iscsi.OFMarker = No



>> With that amount of data exchanged I don't think I have enough storage 
>> for any kind of trace (it also doesn't happen very often - 2 days after 
>> installation of the newest git snapshot).
> 
> OK, I'll run some tests during this weekend with your iSCSI
> parameters.

It just happened again, after ~8-10 hours tgtd was started.

I'm reverting back to tgt-20071227...


-- 
Tomasz Chmielewski
http://wpkg.org




More information about the stgt mailing list