FUJITA Tomonori schrieb:
No more I/O errors from open-iscsi side as well.
I'm running a bit different "test" now and I'm getting host resets on
open-iscsi side every few minutes:
connection3:0: iscsi: detected conn error (1011)
iscsi: host reset succeeded
The "test" is using DRBD over a slow link.
In such case (background synchronization running over a slow link), for
tgtd, DRBD device for look like one that returns reads fast (say, 50
MB/s), but any write, even small, can potentially wait several seconds
to complete (with average speed of about 200 kB/s and long delays).
How can I check what causes these resets?
Have you already checked the tcpdump log?
Not yet - I was transferring several gigabytes, I wouldn't like to have
tcpdump logs of this size...
Other way to reproduce it is establishing a link with a slow speed
between the target and the initiator (i.e., "--shaper 50000" option to
openvpn), and reading/writing data.
You will see device resets, and eventually, I/O errors on the initiator;
with no errors on the target.
Anyway, it is something with open-iscsi, as it's reproducible with IET
as well, so I'll ask at open-iscsi list.
--
Tomasz Chmielewski
http://wpkg.org
--
To unsubscribe from this list: send the line "unsubscribe stgt" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
