[Bug 211990] iscsi fails to reconnect and does not release devices

bugzilla-noreply at freebsd.org bugzilla-noreply at freebsd.org
Thu Nov 10 23:47:16 UTC 2016


https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211990

--- Comment #31 from Julien Cigar <julien at perdition.city> ---
Y(In reply to Steven Hartland from comment #29)

Yes it seems so:

root at filer1:/var/log # vmstat -z | head -n 1; vmstat -z | grep mbuf
ITEM                   SIZE  LIMIT     USED     FREE      REQ FAIL SLEEP
mbuf_packet:            256, 3200070,    1024,    3431,94881716,   0,   0
mbuf:                   256, 3200070,     257,    3103,566359041,   0,   0
mbuf_cluster:          2048, 500010,    4455,     133,15367798,   0,   0
mbuf_jumbo_page:       4096, 250005,       0,     161,102798544,   0,   0
mbuf_jumbo_9k:         9216,  74075,     256,     120,52036055, 304,   0
mbuf_jumbo_16k:       16384,  41667,       0,       0,       0,   0,   0
mbuf_ext_refcnt:          4,      0,       0,       0,       0,   0,   0

> You'd have to catch it when the issue is occurring to confirm

not so easy :) but every time I have a WARNING: 10.20.30.32
(iqn.2016-08.lan.prod:target0): no ping reply (NOP-In) after 30 seconds;
reconnecting

I have a small increase in the "requests for 9k jumbo clusters denied" counter

> If you have plenty of ram you could just trying doubling kern.ipc.nmbjumbo9 to see if the problem goes away?

The machine has only 8GB of RAM (and ZFS), I think I'll first test with a 1500
MTU and -tso

-- 
You are receiving this mail because:
You are on the CC list for the bug.


More information about the freebsd-scsi mailing list