RPI3 swap experiments ["was killed: out of swap space" with: "v_free_count: 5439, v_inactive_count: 1"]
Mark Millard
marklmi at yahoo.com
Thu Aug 2 04:27:40 UTC 2018
[I have a top-posted introduction here in reply
to a message listed at the bottom.]
Bob P. meet Mark J. Mark J. meet Bob P. I'm
hopinh you can help Bob P. use a patch that
you once published on the lists. This was from:
https://lists.freebsd.org/pipermail/freebsd-current/2018-June/069835.html
Bob P. has been having problems with an rpi3
based buildworld ending up with "was killed:
out of swap space" but when the swap partitions
do not seem to be heavily used (seen via swapinfo
or watching top).
[I will avoid the long, complicated history of
investigations here and also any past hypothesis
about contributing causes.]
Bob P. recently introduced Mark J.'s patch to report
the likes of: v_free_count: 5439, v_inactive_count: 1 .
The issue happens during time periods were simple
means of observation suggest that their is lots of
swap space available. Sorting by time (a looping
script was running, logging output, including
swapinfo output):
Wed Aug 1 18:08:10 PDT 2018
Device 1K-blocks Used Avail Capacity
/dev/da0b 1048576 28248 1020328 3%
/dev/mmcsd0s3b 1048576 28256 1020320 3%
Total 2097152 56504 2040648 3%
Aug 1 18:08:13 www kernel: v_free_count: 5439, v_inactive_count: 1
Wed Aug 1 18:08:21 PDT 2018
Device 1K-blocks Used Avail Capacity
/dev/da0b 1048576 31768 1016808 3%
/dev/mmcsd0s3b 1048576 31640 1016936 3%
Total 2097152 63408 2033744 3%
Aug 1 18:08:25 www kernel: pid 93301 (c++), uid 0, was killed: out of swap space
Wed Aug 1 18:08:35 PDT 2018
Device 1K-blocks Used Avail Capacity
/dev/da0b 1048576 24840 1023736 2%
/dev/mmcsd0s3b 1048576 25404 1023172 2%
Total 2097152 50244 2046908 2%
The above is a clean up of the output which had more
and had repeated information from the tail of a log
until it gets new messages. The original script was
something like:
#!/bin/sh
while true
do gstat -abd -I 10s ; date ; swapinfo ; tail -n 2 /var/log/messages
done
I ran out of ability indicate what more to investigate. For
example I'm unsure of when to do the "sysctl vm" that you
[Mark J.] have suggested back in June. It is unlikely that
Bob P. will happen to be there when buildworld has the
kill(s) occur. When it happens in the build sequence is not
stable from one try to the next.
This introduction is a reply to the following.
On 2018-Aug-1, at 6:51 PM, bob prohaska <fbsd at www.zefox.net> wrote:
> The patch to report OOMA information did its job, very tersely. The console reported
> v_free_count: 5439, v_inactive_count: 1
> Aug 1 18:08:25 www kernel: pid 93301 (c++), uid 0, was killed: out of swap space
>
> The entire buildworld.log and gstat output are at
> http://www.zefox.net/~fbsd/rpi3/swaptests/r336877M/
>
> It appears that at 18:08:21 a write to the USB swap device took 530.5 ms,
> next top was killed and ten seconds later c++ was killed, _after_ da0b
> was no longer busy.
>
> This buildworld stopped a quite a bit earlier than usual; most of the time
> the buildworld.log file is close to 20 MB at the time OOMA acts. In this case
> it was around 13 MB. Not clear if that's of significance.
>
> If somebody would indicate whether this result is informative, and any possible
> improvements to the test, I'd be most grateful.
===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)
More information about the freebsd-arm
mailing list