Date: Wed, 30 Dec 2020 21:46:47 +0000 (UTC) From: doug <doug@fledge.watson.org> To: "Steve O'Hara-Smith" <steve@sohara.org> Subject: Re: Observations on virtual memory operations Message-ID: <e3505776-cd-4373-1456-1f5234e2da98@fledge.watson.org> Resent-Message-ID: <1a1be3d6-15fc-d581-292b-da8b47c2d6@fledge.watson.org> In-Reply-To: <20201230060647.38938a75f69e6c045802f655@sohara.org> References: <167603f-a82a-7031-6850-2d08f17a36@fledge.watson.org> <8f3a278a-56cd-c732-68a0-cf6fa5d50a3f@nomadlogic.org> <CADqw_gKWdMaMpCq=%2By=_pvd6wp2bJvDsWkv5GYTwrznhCSpfeQ@mail.gmail.com> <af8ceb5-8ac-57f5-24f8-72b114fa8a@safeport.com> <03553c65-c1c2-db2c-6ab9-a0f4d09c3e2d@nomadlogic.org> <c5cbc8d0-dccd-cb66-a35c-93bcb6cd676@safeport.com> <20201230060647.38938a75f69e6c045802f655@sohara.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 30 Dec 2020, Steve O'Hara-Smith wrote:
> On Wed, 30 Dec 2020 01:03:05 +0000 (UTC)
> doug@safeport.com wrote:
>
>> I do not seem to be able to say clearly enough, there is no memory
>> problem, the is a problem with the paging subsystem filling up swapspace.
>
> Are you using tmpfs or swap backed md ?
>
Neither:
swapctl -l
Device: 1024-blocks Used:
/dev/aacd0p3 4194304 1857780
from vmstat:
procs memory page faults cpu
r b w avm fre flt re pi po fr sr in sy cs us sy id
2 0 16 13G 1.7G 2557 0 0 0 3044 435 509 5523 875 7 1 93
0 0 16 14G 1.5G 8582 0 0 0 7366 428 491 6410 731 6 1 92
0 0 16 14G 1.5G 2945 0 0 0 2934 480 511 4313 780 6 1 93
0 0 16 14G 1.6G 3795 0 0 0 3950 457 434 4910 716 5 1 94
1 0 16 14G 1.6G 589 0 0 0 1128 431 777 2128 1481 2 1 98
2 0 16 14G 1.6G 3273 0 1 0 3329 446 715 5871 1266 5 1 94
And the perl pgm that develops counts from vm.stats.vm
SYSTEM MEMORY INFORMATION:
mem_wire: 1520701440 ( 1450MB) [ 18%] Wired: disabled for paging out
mem_active: + 957276160 ( 912MB) [ 11%] Active: recently referenced
mem_inactive:+ 3454586880 ( 3294MB) [ 41%] Inactive: recently not referenced
mem_cache: + 0 ( 0MB) [ 0%] Cached: almost avail. for allocation
mem_free: + 1823854592 ( 1739MB) [ 21%] Free: fully available for allocation
mem_gap_vm: + 549380096 ( 523MB) [ 6%] Memory gap: UNKNOWN
-------------- ------------ ----------- ------
mem_all: = 8305799168 ( 7921MB) [100%] Total real memory managed
mem_gap_sys: + 240508928 ( 229MB) Memory gap: Kernel?!
-------------- ------------ -----------
mem_phys: = 8546308096 ( 8150MB) Total real memory available
mem_gap_hw: + 43626496 ( 41MB) Memory gap: Segment
Mappings?!
-------------- ------------ -----------
mem_hw: = 8589934592 ( 8192MB) Total real memory installed
SYSTEM MEMORY SUMMARY:
mem_used: 3311493120 ( 3158MB) [ 38%] Logically used memory
mem_avail: + 5278441472 ( 5033MB) [ 61%] Logically available memory
-------------- ------------ ----------- ------
mem_total: = 8589934592 ( 8192MB) [100%] Logically total memory
The system I have the save a command on is older so an apache/php memory
leak is not out of the question. The dipicted here is 11.3, apache24 and
php7. I keep this one up by restarting one or more jails with swap space
usage >50%. When it so it is almost instantaneous in both cases.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?e3505776-cd-4373-1456-1f5234e2da98>