On 10/23/23 13:59, Bart Van Assche wrote:
On 10/23/23 13:41, Bob Pearson wrote:
On 10/23/23 15:21, Bart Van Assche wrote:
On 10/23/23 11:29, Bob Pearson wrote:
Lots of work. Thanks!! Did you mean 6.6 instead of 6.5? I assume
you are running an Ubuntu 23.10 VM and not bare metal while I am
not.
Yes, I meant 6.6 instead of 6.5. I'm indeed running Ubuntu 23.10
inside a VM. BTW, the test is still running and has just reached 231
iterations
...
Very good news! Maybe we can let rxe out of the doghouse.
Is there perhaps a misunderstanding? I ran my tests with the soft-iWARP
driver. Did you run your tests with the RXE driver?
(replying to my own email)
With the RXE driver I see a hang after 143 iterations in the same VM
(Ubuntu 23.10 VM, v6.6-rc7 kernel with kernel debugging disabled). No
SRP paths are available so the hang is not caused by a multipathd bug:
# lsscsi
[0:0:0:0] cd/dvd QEMU QEMU DVD-ROM 2.5+ /dev/sr0
[6:0:0:0] disk Linux scsi_debug 0191 /dev/sda
I couldn't obtain more information because kernel debugging is disabled.
But it is suspicious that this hang happens with the RXE driver and not
with the soft-iWARP driver.
Bart.
