Hello Tomas,
Friday, January 5, 2007, 4:00:53 AM, you wrote:
TÖ> On 04 January, 2007 - Tomas Ögren sent me these 1,0K bytes:
>> On 03 January, 2007 - [EMAIL PROTECTED] sent me these 0,5K bytes:
>>
>> >
>> > >Hmmm, so there is lots of evictable cache here (mostly in the MFU
>> > >part of the cache)... could you make your core file available?
>> > >I would like to take a look at it.
>> >
>> > Isn't this just like:
>> > 6493923 nfsfind on ZFS filesystem quickly depletes memory in a 1GB system
>> >
>> > Which was introduced in b51(or 52) and fixed in snv_54.
>>
>> I've upgraded to snv54 now and even forced 500k dnlc entries in memory
>> (ncsize=500000,arc_reduce_dnlc_percent=0).. Let's see how it copes with
>> that ;)
TÖ> Not too well.. Just went into swap hell.. and why is anon/exec&lib only
TÖ> using 5+2 MB when there's about 50MB marked as Free?
>> arc_reduce_dnlc_percent/D
TÖ> arc_reduce_dnlc_percent:
TÖ> arc_reduce_dnlc_percent: 0
>> dnlc_nentries/D
TÖ> dnlc_nentries:
TÖ> dnlc_nentries: 499126
>> freemem/J
TÖ> freemem:
TÖ> freemem: 9ea
>> ::memstat
TÖ> Page Summary Pages MB %Tot
TÖ> ------------ ---------------- ---------------- ----
TÖ> Kernel 248982 1945 97%
TÖ> Anon 730 5 0%
TÖ> Exec and libs 271 2 0%
TÖ> Page cache 38 0 0%
TÖ> Free (cachelist) 738 5 0%
TÖ> Free (freelist) 6108 47 2%
TÖ> Total 256867 2006
TÖ> Physical 253028 1976
>> arc_reduce_dnlc_percent/W 0t3
TÖ> arc_reduce_dnlc_percent: 0 = 0x3
>> ::memstat
TÖ> Page Summary Pages MB %Tot
TÖ> ------------ ---------------- ---------------- ----
TÖ> Kernel 217511 1699 85%
TÖ> Anon 1502 11 1%
TÖ> Exec and libs 468 3 0%
TÖ> Page cache 48 0 0%
TÖ> Free (cachelist) 333 2 0%
TÖ> Free (freelist) 37005 289 14%
TÖ> Total 256867 2006
TÖ> Physical 253028 1976
>> dnlc_nentries/D
TÖ> dnlc_nentries:
TÖ> dnlc_nentries: 15183
>> freemem/J
TÖ> freemem:
TÖ> freemem: 9351
>> arc::print
TÖ> {
TÖ> size = 0xc34aa00
TÖ> p = 0x4000000
TÖ> c = 0x4000000
TÖ> c_min = 0x4000000
TÖ> c_max = 0x5e114800
>> ARC_mru::print
TÖ> {
TÖ> lsize = 0x4039400
TÖ> size = 0x8bf9e00
>> ARC_mru_ghost::print
TÖ> {
TÖ> lsize = 0x5a6c00
TÖ> size = 0x5a6c00
>> ARC_mfu::print
TÖ> {
TÖ> lsize = 0x2308800
TÖ> size = 0x2fd4800
>> ARC_mfu_ghost::print
TÖ> {
TÖ> lsize = 0xd8000
TÖ> size = 0xd8000
TÖ> Anything useful there? I did a savecore -L after this info was gathered
TÖ> (it was too slow to be useful before that.. had to wait a long time for
TÖ> mdb to show up), so a dump is available..
TÖ> I also have a logfile of ~every minute with ::kmastat, ::memstat and
TÖ> dnlc_nentries from boot up until it started having problems, if someone
TÖ> knows what to look for..
TÖ> Seems like Kernel used up around 84% steadily, then all of a sudden it
TÖ> started growing and within minutes Kernel was doing 97% and the machine
TÖ> more or less stopped moving..
I saw the same behavior here when ncsize was increased from default.
Try with default and lets see what will happen - if it works then it's
better than hung every an hour or so.
--
Best regards,
Robert mailto:[EMAIL PROTECTED]
http://milek.blogspot.com
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
